diff --git a/.gitignore b/.gitignore index a2ecafe..d60e911 100644 --- a/.gitignore +++ b/.gitignore @@ -1,3 +1,11 @@ *.pyc /lib/* -*.env \ No newline at end of file +*.env +*.csv +*.js +*.map +*.json +*.css +*.jsx +*.ts +founder/node_modules/* \ No newline at end of file diff --git a/1sim_vc_responses.py b/1sim_vc_responses.py index 6694b7e..8a07cd4 100644 --- a/1sim_vc_responses.py +++ b/1sim_vc_responses.py @@ -1,3 +1,11 @@ +#!/usr/bin/env python3 +""" +1sim_vc_responses.py +- Generates startup descriptions using prompts from 'VC Design Grid.csv' via OpenAI API. +- Computes text embeddings and cosine similarities for the generated startups. +- Simulates VC responses based on loaded VC contexts and saves results to CSV files. +""" + import os import time import pandas as pd @@ -80,7 +88,7 @@ def get_vc_responses(self, df: pd.DataFrame, vc_context: str, vc_name: str) -> p for company in range(1, 11): desc = df.loc[row, f'Response_{company}'] response = self.client.chat.completions.create( - model="gpt-4", + model="gpt-4o-mini", messages=[ {"role": "system", "content": f"You are {vc_name}. Use this context for decisions:\n{vc_context}"}, {"role": "user", "content": f"Evaluate this startup:\n{desc}"} @@ -108,10 +116,10 @@ def main(): # Get VC responses - marc_context = simulator.load_vc_context('decode(venturing)/Pmarca Blog Archives.pdf') + marc_context = simulator.load_vc_context('Pmarca Blog Archives.pdf') marc_responses = simulator.get_vc_responses(startups_df, marc_context, "Marc Andreessen") - bill_context = simulator.load_vc_context('decode(venturing)/bill gurley.md') + bill_context = simulator.load_vc_context('bill gurley.md') bill_responses = simulator.get_vc_responses(startups_df, bill_context, "Bill Gurley") # Save outputs diff --git a/Attractive_Market_pairplot.png b/Attractive_Market_pairplot.png new file mode 100644 index 0000000..35e6b94 Binary files /dev/null and b/Attractive_Market_pairplot.png differ diff --git a/Attractive_Market_posteriors.png b/Attractive_Market_posteriors.png new file mode 100644 index 0000000..33a40fa Binary files /dev/null and b/Attractive_Market_posteriors.png differ diff --git a/Attractive_Market_predictions.png b/Attractive_Market_predictions.png new file mode 100644 index 0000000..da82bab Binary files /dev/null and b/Attractive_Market_predictions.png differ diff --git a/Attractive_Market_results.png b/Attractive_Market_results.png new file mode 100644 index 0000000..d872c7c Binary files /dev/null and b/Attractive_Market_results.png differ diff --git a/README.md b/README.md index 39a3a6e..d6224a4 100644 --- a/README.md +++ b/README.md @@ -1,29 +1,210 @@ # decode-venturing -## 1sim_vc_responses.py - -| Input/Output | File Name | Path | Type | Description | -|---------------|------------|------|------|-------------| -| **Input**: Scenario templates
**Output**: Generates 10 detailed startup descriptions per template | **VC Design Grid.csv** | `decode-venturing/VC Design Grid.csv` | Input (CSV) | Template file with startup scenario parameters | -| **Input**: Training data for Marc's investment style | **Pmarca Blog Archives.pdf** | `decode-venturing/Pmarca Blog Archives.pdf` | Input (PDF) | Marc Andreessen's investment philosophy | -| **Input**: Training data for Bill's investment style | **bill gurley.md** | `decode-venturing/bill gurley.md` | Input (MD) | Bill Gurley's investment approach | -| **Contains**: Numerical embeddings for similarity analysis | **embeddings.csv** | `embeddings.csv` | Output (CSV) | Vector representations of startup descriptions | -| **Contains**: Pairwise similarity scores between all startups | **cosine_similarity.csv** | `cosine_similarity.csv` | Output (CSV) | Startup similarity matrix | -| **Contains**: Execution score (1-10), Market score (1-10), Investment decision (Yes/No), Reasoning | **vc_investor_responses.csv** | `decode-venturing/vc_investor_responses.csv` | Output (CSV) | Investment decisions | - -## 2sim_based_2step_calib.py - -| Input/Output | File Name | Path | Type | Description | -|---------------|------------|------|------|-------------| -| **Input**: Execution/Market scores, Investment decisions | **vc_investor_responses.csv** | `decoding_decision/vc_investor_responses.csv` | Input (CSV) | Investment decisions data | -| **Input**: Stan model code for two-step calibration | **two_step_model.stan** | `decoding_decision/stan/two_step_model.stan` | Input (Stan) | Statistical model specification | -| **Contains**: Parameter posterior distributions | **two_step_stan_output** | `decoding_decision/two_step_stan_output` | Output (CSV) | Model fitting results | - -## 3plot_diagnostics.py - -| Input/Output | File Name | Path | Type | Description | -|---------------|------------|------|------|-------------| -| **Input**: Raw investment decisions | **vc_investor_responses.csv** | `decoding_decision/vc_investor_responses.csv` | Input (CSV) | Investment decisions data | -| **Input**: Parameter posterior distributions | **two_step_stan_output** | `decoding_decision/two_step_stan_output` | Input (CSV) | Model results | -| **Shows**: Predicted vs actual investment probabilities | **calibration_plot.png** | Generated during runtime | Output (PNG) | Diagnostic visualization | -| **Shows**: Investment probability by execution/idea scores | **heatmap_plot.png** | Generated during runtime | Output (PNG) | Diagnostic visualization | \ No newline at end of file +**decode-venturing** is a powerful toolkit designed to help entrepreneurs refine their startup strategies by simulating venture capitalist (VC) responses, calibrating predictive models, and visualizing key diagnostics. This project leverages AI and Bayesian statistics to align startup ideas with investor preferences, enhancing the likelihood of securing funding. + +--- + +## 📚 Table of Contents + +- [Overview](#overview) +- [Python Files](#python-files) + - [`1sim_vc_responses.py`](#1sim_vc_responsespy) + - [`2sim_based_2step_calib.py`](#2sim_based_2step_calibpy) + - [`3plot_diagnostics.py`](#3plot_diagnosticspy) + - [`path_finding_algorithm.py`](#path_finding_algorithmpy) + - [`state_elicitation_algorithm.py`](#state_elicitation_algorithmpy) +- [Workflow Diagram](#workflow-diagram) +- [Getting Started](#getting-started) +- [Entrepreneurial Implications](#entrepreneurial-implications) +- [Contributing](#contributing) +- [License](#license) +- [Contact](#contact) + +--- + +## 🚀 Overview + +**decode-venturing** streamlines the process of aligning startup ideas with VC investment criteria through: + +1. **Startup Generation:** Creating detailed startup descriptions. +2. **VC Simulation:** Mimicking VC evaluations based on their investment philosophies. +3. **Model Calibration:** Understanding the factors influencing investment decisions. +4. **Diagnostics Visualization:** Assessing model performance. +5. **Strategic Optimization:** Refining strategies to increase investment success. + + +--- + +## 🛠️ Python Files + +### `1sim_vc_responses.py` + +```python +""" +1sim_vc_responses.py +- Generates startup descriptions from 'VC Design Grid.csv' using OpenAI API. +- Computes embeddings and cosine similarities for analysis. +- Simulates VC responses and saves results to CSV files. +""" +``` + +| **Input/Output** | **File Name** | **Path** | **Type** | **Description** | +|------------------|-----------------------------------|--------------------------------------------|----------|---------------------------------------------------------------| +| **Input** | `VC Design Grid.csv` | `decode-venturing/VC Design Grid.csv` | Input (CSV) | Templates with startup scenario parameters | +| **Input** | `Pmarca Blog Archives.pdf` | `decode-venturing/Pmarca Blog Archives.pdf`| Input (PDF) | Marc Andreessen's investment philosophy | +| **Input** | `bill gurley.md` | `decode-venturing/bill gurley.md` | Input (MD) | Bill Gurley's investment approach | +| **Output** | `embeddings.csv` | `decode-venturing/embeddings.csv` | Output (CSV) | Vector representations of startup descriptions | +| **Output** | `cosine_similarity.csv` | `decode-venturing/cosine_similarity.csv` | Output (CSV) | Similarity matrix between startups | +| **Output** | `vc_investor_responses.csv` | `decode-venturing/vc_investor_responses.csv`| Output (CSV) | VC investment decisions and reasoning | + +--- + +### `2sim_based_2step_calib.py` + +```python +""" +2sim_based_2step_calib.py +- Reads VC responses from 'vc_investor_responses.csv'. +- Prepares data for a two-step Stan model. +- Fits the Stan model and saves posterior distributions. +""" +``` + +| **Input/Output** | **File Name** | **Path** | **Type** | **Description** | +|------------------|-------------------------------|--------------------------------------------|----------|---------------------------------------------------------------| +| **Input** | `vc_investor_responses.csv` | `decode-venturing/vc_investor_responses.csv`| Input (CSV) | Data on VC investment decisions | +| **Input** | `two_step_model.stan` | `decode-venturing/stan/two_step_model.stan`| Input (Stan)| Bayesian model specification | +| **Output** | `two_step_stan_output.csv` | `decode-venturing/two_step_stan_output.csv`| Output (CSV)| Posterior distributions from Stan calibration | + +--- + +### `3plot_diagnostics.py` + +```python +""" +3plot_diagnostics.py +- Loads investment data and Stan calibration results. +- Generates calibration scatter plots and heatmaps. +""" +``` + +| **Input/Output** | **File Name** | **Path** | **Type** | **Description** | +|------------------|-----------------------------|--------------------------------------------|----------|---------------------------------------------------------------| +| **Input** | `vc_investor_responses.csv` | `decode-venturing/vc_investor_responses.csv`| Input (CSV) | VC investment decisions data | +| **Input** | `two_step_stan_output.csv` | `decode-venturing/two_step_stan_output.csv`| Input (CSV)| Calibration results from Stan model | +| **Output** | `calibration_plot.png` | Generated during runtime | Output (PNG)| Scatter plot of predicted vs actual investment probabilities | +| **Output** | `heatmap_plot.png` | Generated during runtime | Output (PNG)| Heatmap of investment probabilities by scores | + +--- + +### `alg1_find_path.py` + +```python +""" +path_finding_algorithm.py +- Implements the Path-Finding Algorithm (Local Search) to optimize startup strategies. +- Iteratively generates startups, simulates VC responses, calibrates models, plots diagnostics, and adjusts strategies based on insights. +""" +``` + +| **Input/Output** | **File Name** | **Path** | **Type** | **Description** | +|------------------|---------------------------------|--------------------------------------------|----------|---------------------------------------------------------------| +| **Input** | `1sim_vc_responses.py` | `decode-venturing/1sim_vc_responses.py` | Input (Python) | Data generation and simulation script | +| **Input** | `2sim_based_2step_calib.py` | `decode-venturing/2sim_based_2step_calib.py`| Input (Python) | Model calibration script | +| **Input** | `3plot_diagnostics.py` | `decode-venturing/3plot_diagnostics.py` | Input (Python) | Diagnostic plotting script | +| **Output** | Various CSV and PNG files | `decode-venturing/` | Output (CSV/PNG)| Generated data and diagnostic plots | + +**Functionality:** + +- **Iterative Optimization:** Continuously refine startup strategies based on simulated VC feedback and calibration insights. +- **Strategy Adjustment:** Modify startup parameters to enhance alignment with VC preferences. + +--- + +### `alg2_elicit_state.py` + +```python +""" +state_elicitation_algorithm.py +- Implements the State Elicitation Algorithm (Pattern Recognition) to uncover latent VC preferences. +- Generates startups, simulates VC responses, calibrates models to identify latent states, analyzes patterns, updates policies, and visualizes diagnostics. +""" +``` + +| **Input/Output** | **File Name** | **Path** | **Type** | **Description** | +|------------------|------------------------------------|--------------------------------------------|----------|---------------------------------------------------------------| +| **Input** | `1sim_vc_responses.py` | `decode-venturing/1sim_vc_responses.py` | Input (Python) | Data generation and simulation script | +| **Input** | `2sim_based_2step_calib.py` | `decode-venturing/2sim_based_2step_calib.py`| Input (Python) | Model calibration script | +| **Input** | `3plot_diagnostics.py` | `decode-venturing/3plot_diagnostics.py` | Input (Python) | Diagnostic plotting script | +| **Output** | Various CSV and PNG files | `decode-venturing/` | Output (CSV/PNG)| Generated data and diagnostic plots | + +**Functionality:** + +- **Latent State Identification:** Discover hidden patterns in VC responses to understand underlying investment factors. +- **Policy Updating:** Adjust startup strategies based on identified latent states to better align with VC preferences. + +--- + +## 📈 Workflow Diagram + +```mermaid +graph TD + A[1sim_vc_responses.py] -->|Generates Data| B[2sim_based_2step_calib.py] + B -->|Calibrates Model| C[3plot_diagnostics.py] + A -->|Provides Data| G[Path-Finding Algorithm] + B --> G + C --> G + A -->|Provides Data| H[State Elicitation Algorithm] + B --> H + C --> H + +``` + +--- + +## 🏁 Getting Started + +1. **Clone the Repository:** + ```bash + git clone https://github.com/yourusername/decode-venturing.git + cd decode-venturing + ``` + +2. **Install Dependencies:** + ```bash + pip install -r requirements.txt + ``` + +3. **Set Up Environment Variables:** + - Create a `.env` file in the root directory. + - Add your OpenAI API key: + ``` + OPENAI_API_KEY=your_openai_api_key_here + ``` + +4. **Prepare Input Files:** + - Place `VC Design Grid.csv`, `Pmarca Blog Archives.pdf`, and `bill gurley.md` in the project root. + +5. **Run Algorithms:** + - **Path-Finding Algorithm:** + ```bash + python3 path_finding_algorithm.py + ``` + - **State Elicitation Algorithm:** + ```bash + python3 state_elicitation_algorithm.py + ``` + +6. **View Outputs:** + - Check generated CSV files and diagnostic plots in the `decode-venturing/` directory. + +--- + +## 💡 Entrepreneurial Implications + +- **Refine Strategies:** Iteratively adjust startup ideas based on simulated VC feedback to better align with investor preferences. +- **Understand VC Preferences:** Uncover hidden factors that drive investment decisions, enabling more targeted pitches. +- **Data-Driven Decisions:** Utilize robust statistical models and visual diagnostics to inform strategic choices and enhance investment prospects. +- **Optimize Pathways:** Continuously improve the pathway to securing investments through informed strategy adjustments. + +--- diff --git a/Team_Execution_pairplot.png b/Team_Execution_pairplot.png new file mode 100644 index 0000000..3381051 Binary files /dev/null and b/Team_Execution_pairplot.png differ diff --git a/Team_Execution_posteriors.png b/Team_Execution_posteriors.png new file mode 100644 index 0000000..23fb88a Binary files /dev/null and b/Team_Execution_posteriors.png differ diff --git a/Team_Execution_predictions.png b/Team_Execution_predictions.png new file mode 100644 index 0000000..a5b2114 Binary files /dev/null and b/Team_Execution_predictions.png differ diff --git a/Team_Execution_results.png b/Team_Execution_results.png new file mode 100644 index 0000000..a9083e8 Binary files /dev/null and b/Team_Execution_results.png differ diff --git a/Would_Invest_pairplot.png b/Would_Invest_pairplot.png new file mode 100644 index 0000000..e14bbc9 Binary files /dev/null and b/Would_Invest_pairplot.png differ diff --git a/Would_Invest_posteriors.png b/Would_Invest_posteriors.png new file mode 100644 index 0000000..d591779 Binary files /dev/null and b/Would_Invest_posteriors.png differ diff --git a/Would_Invest_predictions.png b/Would_Invest_predictions.png new file mode 100644 index 0000000..ab04c4d Binary files /dev/null and b/Would_Invest_predictions.png differ diff --git a/Would_Invest_results.png b/Would_Invest_results.png new file mode 100644 index 0000000..8928692 Binary files /dev/null and b/Would_Invest_results.png differ diff --git a/alg1_find_path.py b/alg1_find_path.py new file mode 100644 index 0000000..8b85383 --- /dev/null +++ b/alg1_find_path.py @@ -0,0 +1,106 @@ +import numpy as np +import pandas as pd +from typing import Dict, List +from dataclasses import dataclass + +# using [Investor Classification Framework for Pathfinding](https://claude.ai/chat/37933b85-ebce-4099-9886-196ad36ec653) +@dataclass +class Investor: + id: str + capacity: float + idea_weight: float + execution_weight: float + +@dataclass +class Founder: + id: str + capital_needed: float + idea_signals: Dict[str, float] + execution_signals: Dict[str, float] + +class PathFindingAlgorithm: + def __init__(self, investors: List[Investor], founders: List[Founder]): + self.investors = investors + self.founders = founders + self.state_space = ['idea', 'execution'] + self.matches = pd.DataFrame() + + def compute_benefit(self, investor: Investor, founder: Founder) -> float: + idea_score = sum(founder.idea_signals.values()) * investor.idea_weight + execution_score = sum(founder.execution_signals.values()) * investor.execution_weight + return idea_score + execution_score + + def find_paths(self, max_iterations: int = 10, epsilon: float = 0.01): + # Initialize matching distribution + n_investors = len(self.investors) + n_founders = len(self.founders) + pi = np.ones((n_investors, n_founders)) / (n_investors * n_founders) + + for t in range(max_iterations): + # Step 2: Primal (Path-Finding) + benefits = np.zeros((n_investors, n_founders)) + for i, investor in enumerate(self.investors): + for f, founder in enumerate(self.founders): + benefits[i,f] = self.compute_benefit(investor, founder) + + # Optimize matching (simplified version) + pi_new = self._optimize_matching(pi, benefits) + + # Step 3: State Elicitation/Certification + certified_paths = self._certify_paths(pi_new) + + # Step 4: Augment state space if needed + if not all(certified_paths): + self._augment_state_space() + continue + + if np.all(np.abs(pi_new - pi) < epsilon): + break + + pi = pi_new + + self.matches = self._convert_to_dataframe(pi) + return self.matches + + def _optimize_matching(self, pi, benefits): + # Simplified matching optimization + # In practice, use linear programming solver + return pi * benefits / np.sum(pi * benefits) + + def _certify_paths(self, pi): + # Check if matches satisfy investor constraints + return [True] * pi.shape[0] # Simplified + + def _augment_state_space(self): + # Add more granular signal interpretation + self.state_space.append('market_validation') + + def _convert_to_dataframe(self, pi): + matches = [] + for i, investor in enumerate(self.investors): + for f, founder in enumerate(self.founders): + if pi[i,f] > 0.01: # Threshold + matches.append({ + 'investor_id': investor.id, + 'founder_id': founder.id, + 'match_probability': pi[i,f], + 'idea_score': sum(founder.idea_signals.values()), + 'execution_score': sum(founder.execution_signals.values()) + }) + return pd.DataFrame(matches) + + def elicit_states(self, match_data): + """Analyze patterns in successful/failed matches""" + pattern_data = [] + for _, match in match_data.iterrows(): + successful = match['successful'] + # Extract patterns in investor preferences + pattern = { + 'investor_type': match['investor_type'], + 'sector': match['sector'], + 'stage': match['stage'], + 'success': successful + } + pattern_data.append(pattern) + + return pd.DataFrame(pattern_data) \ No newline at end of file diff --git a/alg2_calibrate_inv.py b/alg2_calibrate_inv.py new file mode 100644 index 0000000..a3e7408 --- /dev/null +++ b/alg2_calibrate_inv.py @@ -0,0 +1,100 @@ +""" +calibration algorithm (from investor's perspective). +the need of this procedure is explained as follows: +Imagine you're playing a video game where you need to match heroes with the right teammates. +The game shows you a percentage score for each potential team-up, like "90% good match!" But if that number isn't calibrated properly, +you might keep picking teams that look great on paper but lose in actual battles. In the startup world, this matters most when founders +are trying to match with investors based on their team's background - like having PhDs or previous startup experience. +A well-calibrated system means if it tells you "you have an 80% chance of matching with science-focused investors because of your PhD," that prediction should be trustworthy +because looking at past similar cases, about 80% of PhD teams actually did get investment from those investors. This helps founders make better decisions about which +investors to approach instead of wasting time on matches that look good but rarely work out in reality." + +""" +import numpy as np +import pandas as pd +from sklearn.isotonic import IsotonicRegression + +class InvestorCalibration: + def __init__(self, n_bins=10, epsilon=0.05): + self.n_bins = n_bins + self.epsilon = epsilon + self.bin_edges = np.linspace(0, 1, n_bins + 1) + self.calibration_map = None + + def fit(self, predicted_probs: np.array, actual_outcomes: np.array): + """Fit calibration model using historical data""" + + # Step 1: Bin probabilities + bin_indices = np.digitize(predicted_probs, self.bin_edges) - 1 + + # Step 2: Compute empirical frequencies + self.bin_frequencies = [] + for bin_idx in range(self.n_bins): + mask = bin_indices == bin_idx + if np.any(mask): + freq = np.mean(actual_outcomes[mask]) + self.bin_frequencies.append(freq) + else: + self.bin_frequencies.append(np.nan) + + # Step 3: Measure calibration error + bin_midpoints = (self.bin_edges[:-1] + self.bin_edges[1:]) / 2 + calibration_errors = np.abs( + np.array(self.bin_frequencies) - bin_midpoints + ) + + # Step 4: Fit isotonic regression for calibration + ir = IsotonicRegression(out_of_bounds='clip') + valid_mask = ~np.isnan(self.bin_frequencies) + if np.any(valid_mask): + self.calibration_map = ir.fit( + bin_midpoints[valid_mask], + np.array(self.bin_frequencies)[valid_mask] + ) + + def calibrate(self, predicted_probs: np.array) -> np.array: + """Apply calibration to new predictions""" + if self.calibration_map is None: + return predicted_probs + return self.calibration_map.predict(predicted_probs) + + def plot_reliability(self, predicted_probs, actual_outcomes): + """Generate reliability diagram data""" + bin_indices = np.digitize(predicted_probs, self.bin_edges) - 1 + plot_data = [] + + for bin_idx in range(self.n_bins): + mask = bin_indices == bin_idx + if np.any(mask): + mean_predicted = np.mean(predicted_probs[mask]) + mean_actual = np.mean(actual_outcomes[mask]) + n_samples = np.sum(mask) + + plot_data.append({ + 'bin_start': self.bin_edges[bin_idx], + 'bin_end': self.bin_edges[bin_idx + 1], + 'predicted_prob': mean_predicted, + 'actual_freq': mean_actual, + 'n_samples': n_samples + }) + + return pd.DataFrame(plot_data) + + def get_calibration_stats(self): + """Return calibration metrics""" + valid_freqs = [f for f in self.bin_frequencies if not np.isnan(f)] + if not valid_freqs: + return {} + + bin_midpoints = (self.bin_edges[:-1] + self.bin_edges[1:]) / 2 + valid_bins = ~np.isnan(self.bin_frequencies) + + return { + 'max_calibration_error': np.max(np.abs( + np.array(valid_freqs) - bin_midpoints[valid_bins] + )), + 'mean_calibration_error': np.mean(np.abs( + np.array(valid_freqs) - bin_midpoints[valid_bins] + )), + 'n_reliable_bins': len(valid_freqs) + } \ No newline at end of file diff --git a/categorical_effects.png b/categorical_effects.png new file mode 100644 index 0000000..9ce460e Binary files /dev/null and b/categorical_effects.png differ diff --git a/front.js b/front.js new file mode 100644 index 0000000..34ac74f --- /dev/null +++ b/front.js @@ -0,0 +1,209 @@ +import React, { useState } from 'react'; +import { Card, CardContent } from "@/components/ui/card"; +import { Tabs, TabsContent, TabsList, TabsTrigger } from "@/components/ui/tabs"; +import { LineChart, Line, XAxis, YAxis, CartesianGrid, Tooltip, ResponsiveContainer } from 'recharts'; + +const InvestorDashboard = () => { + const [matches, setMatches] = useState([ + { + id: 1, + type: "Technical Patents Group", + investors: ["Marc Andreessen", "Sam Altman", "Jeff Bezos"], + rawScore: 0.8, + calibratedScore: 0.65, + signals: { + idea: { + "Technical IP": 0.9, + "Novel Approach": 0.85 + }, + execution: { + "Team Background": 0.7, + "Market Validation": 0.4 + } + }, + history: { + similarCases: 120, + actualSuccess: 78 + } + }, + { + id: 2, + type: "Monopoly Patents Group", + investors: ["Peter Thiel", "Josh Wolfe", "Mike Maples Jr"], + rawScore: 0.75, + calibratedScore: 0.60, + signals: { + idea: { + "Monopoly Potential": 0.95, + "Market Defensibility": 0.80 + }, + execution: { + "Scientific Depth": 0.85, + "IP Strategy": 0.75 + } + }, + history: { + similarCases: 90, + actualSuccess: 54 + } + } + ]); + + const InvestorMatrix = () => ( +
+ +

Patents as Execution

+
+

Andreessen, Altman, Bezos

+
    +
  • Technical depth focus
    • +
  • Product shipping history
    • +
  • Team execution ability
    • +
+
+ + + +

Patents as Moat

+
+

Thiel, Wolfe, Maples

+
    +
  • Monopoly potential
    • +
  • Market defensibility
    • +
  • Scientific breakthrough
    • +
+
+ + + +

Team Over Patents

+
+

Graham, Ravikant, Dixon

+
    +
  • Rapid execution
    • +
  • Market understanding
    • +
  • Customer traction
    • +
+
+ + + +

Innovation Patents

+
+

Srinivasan, Boyle

+
    +
  • Technical novelty
    • +
  • Regulatory fit
    • +
  • Market timing
    • +
+
+ +
+ ); + + const CalibrationPlot = ({ match }) => { + const calibrationData = [ + { bin: "0-20%", predicted: 0.1, actual: 0.08 }, + { bin: "20-40%", predicted: 0.3, actual: 0.25 }, + { bin: "40-60%", predicted: 0.5, actual: 0.45 }, + { bin: "60-80%", predicted: 0.7, actual: 0.62 }, + { bin: "80-100%", predicted: 0.9, actual: 0.71 } + ]; + + return ( +
+ + + + + + + + + + +
+ ); + }; + + const MatchCard = ({ match }) => ( + + +
+
+

{match.type}

+
{match.investors.join(", ")}
+
+
+
Raw Score: {(match.rawScore * 100).toFixed()}%
+
Calibrated: {(match.calibratedScore * 100).toFixed()}%
+
+
+ +
+
+

Idea Signals

+ {Object.entries(match.signals.idea).map(([key, value]) => ( +
+ {key} + {(value * 100).toFixed()}% +
+ ))} +
+
+

Execution Signals

+ {Object.entries(match.signals.execution).map(([key, value]) => ( +
+ {key} + {(value * 100).toFixed()}% +
+ ))} +
+
+ +
+

Historical Calibration

+
+ Based on {match.history.similarCases} similar cases, + {match.history.actualSuccess} succeeded + ({((match.history.actualSuccess / match.history.similarCases) * 100).toFixed()}% actual success rate) +
+ +
+ + + ); + + return ( +
+ + + Investor Matrix + Match Analysis + Calibration Details + + + + + + + + {matches.map(match => ( + + ))} + + + + + +

Overall Calibration Analysis

+ + + + + +
+ ); +}; + +export default InvestorDashboard; \ No newline at end of file diff --git a/investment_decision_heatmap.png b/investment_decision_heatmap.png new file mode 100644 index 0000000..123be8a Binary files /dev/null and b/investment_decision_heatmap.png differ diff --git a/investment_probability_by_factor.png b/investment_probability_by_factor.png new file mode 100644 index 0000000..e8fc33e Binary files /dev/null and b/investment_probability_by_factor.png differ diff --git "a/investor/\360\237\227\204\357\270\217\360\237\227\243\357\270\217belief.md" "b/investor/\360\237\227\204\357\270\217\360\237\227\243\357\270\217belief.md" new file mode 100644 index 0000000..4d18dbe --- /dev/null +++ "b/investor/\360\237\227\204\357\270\217\360\237\227\243\357\270\217belief.md" @@ -0,0 +1,15 @@ + +| **Investor Name** | **Title** | **Gist of Content** | **Examples** | **URL** | +| ------------------- | ------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------- | +| **Marc Andreessen** | **Making the Future** | Emphasizes the importance of actively creating and shaping the future rather than passively waiting for it to happen. Advocates for innovation, long-term thinking, and perseverance to drive societal progress. | Marc discusses the rise of software-driven businesses like Airbnb, which transformed the hospitality industry, and Tesla, which disrupted the auto industry by focusing on electric vehicles and software integration. | [Link](https://a16z.com/2011/08/20/why-software-is-eating-the-world/) | +| **Marc Andreessen** | **Big Breakthrough Ideas and Courageous Entrepreneurs** | Champions bold, transformative ideas that challenge the status quo. Highlights the importance of entrepreneurs who solve hard problems, often facing resistance or skepticism, but ultimately delivering outsized impact. | Cites the example of early internet pioneers, such as Amazon (redefining retail) and Facebook (redefining communication), who faced skepticism but proved their ideas could scale. Also praises biotech startups using AI to discover new drugs faster and solve unmet medical needs. | [Link](https://a16z.com/2011/08/20/why-software-is-eating-the-world/) | +| **Marc Andreessen** | **It's Time to Build** | Urges society to prioritize building infrastructure, technology, and systems to tackle global challenges like healthcare, housing, and education. Criticizes societal inertia and the lack of investment in transformative projects. | References the U.S.’s lagging infrastructure compared to other countries and the need to rethink housing policies to reduce shortages. Points to the success of SpaceX in reviving space exploration as an example of what bold investment and private entrepreneurship can achieve when government lags behind. | [Link](https://a16z.com/2020/04/18/its-time-to-build/) | +| **Marc Andreessen** | **Why You Should Be Optimistic About the Future** | Highlights the rapid pace of technological progress and the proliferation of startups addressing critical problems in society. Encourages belief in innovation and entrepreneurship as vehicles for abundance and prosperity. | Discusses how renewable energy costs are dropping rapidly due to solar and battery breakthroughs, and how advances in AI are revolutionizing industries from healthcare to logistics. Praises startups tackling environmental challenges, such as reducing carbon footprints or advancing clean energy adoption. | [Link](https://a16z.com/2011/08/20/why-software-is-eating-the-world/) | +| **Peter Thiel** | **Competition is for Losers** | Advocates for monopolistic businesses that dominate markets by creating unique value, arguing that competition erodes profits and innovation. | Uses Google as an example of a company that created a near-monopoly in search by focusing on proprietary technology (PageRank) and leveraging network effects. Also contrasts this with the restaurant industry, where fierce competition leaves businesses unprofitable. | [Link](https://www.youtube.com/watch?v=yODORwGmHqo) | +| **Peter Thiel** | **Successful Businesses are Based on Secrets** | Emphasizes that great businesses are built on discovering "secrets" — unique insights that others have overlooked. Secrets are the foundation for innovation and creating breakthrough companies. | Highlights Facebook's success, which was based on the secret that "real identity" social networking was more compelling than anonymous or pseudonymous platforms like MySpace. Discusses PayPal's early success by focusing on eBay power sellers, a niche market others ignored. | [Link](https://www.youtube.com/watch?v=yODORwGmHqo) | +| **Peter Thiel** | **You Are Not a Lottery Ticket** | Rejects the notion that success is purely based on luck. Argues that entrepreneurs and innovators succeed through vision, planning, and execution rather than randomness or iteration alone. | Uses Elon Musk as an example, citing SpaceX and Tesla as ventures built on deliberate, high-risk, but well-planned strategies. Contrasts this with startups that rely on endless A/B testing, arguing this leads to marginal improvements rather than transformative change. | [Link](https://www.youtube.com/watch?v=iZM_JmZdqCw) | +| **Peter Thiel** | **Definite Optimism** | Advocates for a mindset of "definite optimism," where entrepreneurs envision a better future and work with conviction toward achieving it, rather than relying on luck or vague hope. | Highlights the construction of the transcontinental railroad and SpaceX as examples of definite optimism, where bold visions were paired with concrete execution. Contrasts this with "indefinite optimism," where societies or companies rely on vague hope without substantive action. | [Link](https://www.youtube.com/watch?v=iZM_JmZdqCw) | +| **Bill Gurley** | **Venture Capital Red Flag Checklist** | Outlines key governance and operational warning signs in startups, emphasizing the importance of proper governance, financial discipline, and transparency to avoid fraud or failure. | Highlights FTX as a case study, where the absence of a legitimate board, aversion to audits, and opaque financial practices led to its collapse. Advises watching for governance lapses like dual-class stock structures and excessive founder liquidity. | [Link](https://abovethecrowd.com/2022/11/28/venture-capital-red-flag-checklist/) | +| **Bill Gurley** | **On Bubbles** | Warns against unchecked speculation and risk-taking during frothy markets, emphasizing the cyclical nature of markets and the importance of assessing risk accurately. | Mentions the dot-com bubble, during which irrational exuberance led to overfunding and unsustainable valuations. Highlights employees and investors working for profitless companies as a sign of systemic risk and warns against dismissing the importance of profitability. | [Link](https://abovethecrowd.com/2014/01/24/on-bubbles/) | +| **Bill Gurley** | **Late-stage Private Rounds vs. IPOs** | Critiques late-stage financing trends, arguing that private rounds lack the rigor of IPOs, leading to improper valuation, governance issues, and unproven business models being overfunded. | Cites Fab.com as an example of a company that burned through $330M without proving its business model. Warns against late-stage investors accepting inflated valuations without scrutiny, which would have been corrected in a rigorous IPO process. | [Link](https://abovethecrowd.com/2015/07/20/todays-100m-late-stage-private-rounds-are-very-different-from-an-ipo/) | +| **Bill Gurley** | **It's Time to Invest in Tough Markets** | Advocates for investing in tough economic environments, as they weed out opportunists and create better opportunities for long-term-focused entrepreneurs. | Highlights 2008 as a period when strong startups like Uber emerged. Notes that tough markets offer benefits like lower costs for hiring and rent, reduced competition, and a clearer focus on building sustainable businesses. | [Link](https://abovethecrowd.com/2008/12/01/open-for-business/) | diff --git "a/investor/\360\237\227\204\357\270\217\360\237\244\234act.md" "b/investor/\360\237\227\204\357\270\217\360\237\244\234act.md" new file mode 100644 index 0000000..e69de29 diff --git a/bill gurley.md "b/investor/\360\237\227\243\357\270\217belief/BG.md" similarity index 100% rename from bill gurley.md rename to "investor/\360\237\227\243\357\270\217belief/BG.md" diff --git "a/investor/\360\237\227\243\357\270\217belief/MA.md" "b/investor/\360\237\227\243\357\270\217belief/MA.md" new file mode 100644 index 0000000..1daa77e --- /dev/null +++ "b/investor/\360\237\227\243\357\270\217belief/MA.md" @@ -0,0 +1,856 @@ +## Making the Future + +### Introduction + +Patrick + +Our guest today is Silicon Valley icon, Marc Andreessen. Before co-founding the venture capital firm Andreessen Horowitz, Marc was an early pioneer of the internet. At age 22, he built Mosaic, the first widely adopted web browser and the technology that underpinned Netscape communications. Marc was an early proponent of cloud computing, social networks, and the software business model. In each case, Marc seemed to be well ahead of the crowd. During our conversation, we explore how software is making the world better, how slow sectors like education, healthcare and housing are eating the economy, and Marc's vision for the future of a16z. Please enjoy my conversation with Marc Andreessen. Before we transition to the episode, I also wanted to highlight our newest series Business Breakdowns. Each week, we do a deep dive into an individual business to understand what makes it great. Find more information on joincolossus.com or search for and sign up to the Business Breakdowns feed on your preferred podcast player. + +### Software Eating The World + +Patrick + +Marc, we're going to cover a ton of territory today. I'm going to start with some of your iconic ideas, and we'll start with maybe your most famous, which is this concept of software eating the world. I want to focus on the middle word, eating. I think everyone understands software. People understand the world. Could you describe exactly what you mean by eating, like what is literally happening here? + +Marc + +Current events, just trying to make it colorful, to make it interesting. But the thing that's happening is, quite literally, it's actually a concept from Buckminster Fuller, believe it or not, a hundred years ago. And he called it... I think his term for it was ephemeralization. The reason I'm getting confused is because economists have a separate term they call dematerialization. Think of those two terms. And so Buckminster Fuller had this thing that he saw. It's amazing that he saw this. He saw this in, I forget what it was, like 1910s of 1920s or something, way before the computer was invented. But he looked at the progress of industrial development at a time where people generally had a different point of view. He said basically with science and technology, we're getting better and better at doing more and more with less and less, until ultimately one day we'll be able to do everything with nothing. + +And the literal meaning of what he meant, and this is why economists now have this idea of dematerialization, the literal idea is like in the beginning of industrial development you're using raw materials in this grossly ineffective way. You're just mining for huge amounts of natural resources. You've got just this incredibly inefficient process for making steel, so you're just using huge amounts of ore, you're using incredibly engines. You're using huge amounts of oil to generate very small amounts of power. And then what basically happens is you get more and more advanced and you get much more efficient. You use less natural resources, you use less energy, and you're getting better and better results. + +It's funny that he observed that was actually happening, it's just simply the art of science and technology even before computers. And then software's that alchemy that finishes that, and ultimately I think in the far future delivers ultimately his vision of doing everything + +Invest Like The Best - Making the Future Page of 28. + +1 with nothing, which is quite literally taking activities, productive processes, and entire categories of products and services that have a real-world footprint, a footprint made of atoms today and ultimately turned them into bits. We've all been through this in the last 18 months, the degree to which travel, driving to the office and the actual physical office building quite literally got dematerialized into video conferencing. It's sort of one of these classic, incredible examples of that phenomenon. That's the long-term part. That's ultimately where it's headed. + +Patrick + +I've seen you describe it elsewhere as just the ultimate lever on the physical world, and I think of Vaclav Smil's work on energy where you've got sources of energy and you've got the efficiency. How much of the energy can you get out of the fossil fuel or the nuclear rod or whatever? It seems like software is the best possible way, not just to make atoms more efficient, but also remove them from the equation. I think that's what you're saying. Do I have that roughly right? + +Marc + +The way I'd put it is basically it's like, "Look, a physical product can be replaced by software." The example I always use for people who are trying to think about this from scratch is to get up in the morning, most of us used to have an alarm clock, a little plastic box that sat on our bedside table. It plugged into the wall. It would yell at us to get up in the morning. There was this entire company, Radio Shack, that was in the business of selling us all these little plastic boxes that did all these things. Now that little plastic box, that alarm clock, it's an app for your phone. And so is the answering machine, cd player, and so is the DVD players, and so is literally dozens or hundreds of little things. + +And so for anything that can be quite literally transformed or replaced and transitioned from atoms to bits like that, happens. That caused kind of a form of reaction, that original essay and that argument, which basically is okay smart guy, that makes sense for alarm clocks or whatever and electric shavers, but what about cars? What about houses? You can't drive bits. If you do need to go from one place to another you at some point need that metal, glass and rubber in the form of a car. There are certainly things that will always have some real-world footprint, at least for a very long time. But to your use of this term lever, basically the way to think about it then is software is a lever to be able to have those physical systems and products become much more efficient and much more powerful way in advance of them ever actually being eliminated as physical products. + +And that's happening with the car in real time, which is first of all, the car is being transformed into basically a rolling supercomputer, but it's still bent metal and glass and rubber. But more and more the value of what the car is, and you have quite literally functional value, quite literally the economic value of the car is being basically turned into software. These things are supercomputers on wheels. They're now starting to drive themselves. The entire basis for competition in the auto industry is to transition to software. The entire basis of how people are going to evaluate one car versus another is going to be on the basis of its software, which is very different than how people used to think about these things. + +The other way to think about it is, how is the activity of all those cars coordinated? This is where you get this kind of magical phenomenon in the form of these systems like Lyft and Uber where, just think conceptually for a second, how did Lyft and Uber come into being? The thing that literally happened was a programmer sat down, typed in a bunch of computer code on a computer, and pressed enter, and the code started to run. And then the next day, cars and their drivers started to drive to different places and pick up passengers who they + +normally never would have even known about, and those passengers all of a sudden had a completely new way to get a ride. + +You configured an entirely new revolutionary real-world transportation system. Not by building a new kind of car, not even by buying a lot of cars, not by building train tracks or a monorail or sub-orbital rockets or any of these things, but just having software recoordinate existing physical resources into a far more efficient form. Airbnb is the same thing for housing, and so forth and so on. And so software is the lever of the world. Somebody types in code, that code runs the real-world changes. And the real-world changes not just arbitrarily but in a way that's more efficient, that's more organized, that's more productive, that provides more opportunity to people. + +There's just this enormous future runway as software is becoming pervasive in our world, just like every physical item that's getting a chip, as every chip is getting connected to the network, as this technology kind of saturates every kind of physical product around us, which is what's happening. The software is increasingly the coordination layer. You have this long arc of improvement in the physical world that actually doesn't require a complete replacement of the bits, but just lets you use them much, much better. + +Patrick + +Do you think there are credible downsides or potential downsides to this trend? You might call this like the Luddite question or South Park, they took our jobs, like the automation problems that people talk about. What, if any, do you think are credible concerns about this inexorable trend? + +Marc + +It is economic change. It is change. Of course, this is a process that's been running for a very long time. Once upon a time, all of mankind was doing subsistence farming. The good news is subsistence farming jobs went away. Those were not very good jobs. All the jobs that everybody has, who's listening to this podcast, these are all jobs that really created this consequence of technology at some point in the last 300 years. It is a process of change. It is a process that results in creating more and better jobs. We're all relieved that we don't live in a world of pre-automation, for that reason, because we really seriously equate quite dramatically without the jobs that preceded technology. We are all the beneficiaries of this trend. + +But it is change kind of in the moment. There are entire fields and professions that do get eliminated through this process. A classic historical example is there was once a thriving professional world of blacksmiths. The job of the blacksmith was to put shoes on horses. Horses were the dominant transportation technology of the time, and when the car came out the blacksmiths were not happy. And there were blacksmiths that went out of work, and there were blacksmiths that had to retool into a different career kind of late in their lives, in a way they never would have wanted to. And there were blacksmiths that never had the same standard of living on the other side of that. + +It is change, and it is a disruption of the status quo. There are a few things that you just have to kind of put balance against that. I think one is the point I already made, which is that the result of this process is more and better jobs than you had before. That is good for many people in their own lifetimes that get to have a better life as a consequence. And by the way, it's doubly true of their kids, and I think this is the big difference between people who think about this process abstractly and people who think about it in the real world, which is if you ask any responsible parent, ask any responsible parent would you like your kid to have + +the same job you do, or would you like them to have a better and different job, a different experience? + +Ask a parent who works at a blue collar profession. Ask a parent who works on an assembly line or works in the frontline of manual labor, the harder work. Do you want your kid to be doing that, or would you rather your kid be a software developer or an artist or a job in which you're in a very comfortable physical environment. You're not running the risk of workplace accidents and so forth, higher paid, and able to provide better for that kid's family and for ultimately your grandchildren. And virtually all parents will say that they are in favor of that. This is progress. This is how it happens. So I think that's important. + +The other really important thing I think people just miss, and this is sort of this massive disconnect between the narrative and then the facts, or the statistics, is basically like there's this perception and narrative that we live in this unprecedented technological disruption. You have these both very triumphless kind of analysis of software is eating the world, or you've got these very negative blasts. The New Yorker did that big piece awhile ago about how awful disruption is and how technology is destroying everything and making everything awful. There's this narrative that this is this time of unprecedented time of disruption, technology-driven disruption. + +Then you actually look at the numbers. The number that you look at to try to get a handle on what's actually happening is productivity growth. Productivity growth is the economic statistic that captures... It captures a bunch of things, but primarily it captures the impact of technological change on the economy. You can quite literally just get a gauge and be like, "How much is technology transforming the economy in any given year by what the productivity growth number is for that year?" And productivity growth... You'll know this, productivity growth was much higher from, call it, the 1920s to the 1970s than it was from the 1970s to today. Generally speaking, the rate of technological change in economy has been decelerating for the last 40 years, not accelerating. + +And everybody thinks it's been accelerating. It's actually been decelerating. By the way, productivity growth has been decelerating. Several other important related metrics have been decelerating. The rate of new business formation has been decelerating. And again, this is one of these contra narrative things where you think entrepreneurialism is everywhere. It's like, "No, no actually across the whole economy entrepreneurialism is in decline." Company turnover is in decline. Job turnover is in decline. There's this narrative that your kids are going to have a lot more jobs in their lifetime than you do. That was true at one point. It's not true any more. + +And then there are these two other really interesting metrics I always looked at when you get to the jobs questions, which is if you look under the net employment numbers, which are the headline numbers, you'll read the headline numbers every month or whatever of X hundred thousand jobs were created this quarter. That's a net number, but that's net of job growth creation and net job growth destruction. It's literally creation minus destruction. And it turns out the creation and destruction numbers are much higher than the net numbers, but they're also falling. And so the rate of new job creation in the economy has been falling for 40 years. The rate of new job destruction has been falling in the last 40 years. + +By the way, another interesting thing, the aggregate level... I think I have this right from memory, the aggregate job creation and destruction rates in the US are something like 4X per capita the respective rates in Europe. Europe has a far more rigorous draconian regulatory system aimed at basically stalling economic change. They have succeeded in that mission, and the way that we know that is they've succeeded by basically dramatically + +reducing both the rates. They've done what they wanted to do. They've reduced the rate of job destruction. There's a lot less job turnover. They've also correspondingly really sharply reduced by an equivalent amount the rate of new job creation. And of course the result of that is unemployment rates in Europe are higher, wage growth is slower, new job creation rate is lower. + +Generally speaking, this has been, I would argue, a bad trade for Europeans in general. You can make an argument either way, but it's like you also have the counterexample of society that's decided that they want even less change, and I don't know that it works out that well. You want to parametrize that, which is like we're not actually in this era of just enormous technological change the way people think, and we can discuss why that is. It's a very interesting question. To me, just to the conclusion that criticisms are actually 100% up based. They're completely wrong. If you want more job creation, if you want higher wages, and if you want more of a sense of opportunity and potential and the ability to be fulfilled and the ability for people to provide for their family and for their kids to have great jobs in the future, you actually want more technology in the economy. You want more productivity growth, you want more economic change, which again is very contrary to current narrative. + +### Major Impediments Over the Coming Decades + +Patrick + +I think this productivity growth thing is the most fascinating thing in the world, because how odd it is to put the numbers against what feels, especially in 2021 and maybe some things changing or has changed post-COVID, but what feels just like explosive experimentation in lots of sectors beyond software, that we're sort of stagnated for a while. It begs the question, you kind of alluded to it, if we agree that productivity growth is good and want it to go a lot faster, what are the major category impediments to that potentially happening over the next 10 to 20 years? + +Marc + +To you point, let's talk about this in two phases. Let's talk about it up until 2019, and then let's talk about what might happen from here, because I think the world maybe did just change in this respect in a pretty interesting way. Basically from, call it, 1971... If you look at the charts, from 1971 to basically 2019 is this era of... Stagnation's the wrong word, because there was a lot of change, but this sort of decelerating rate of change. My analysis is basically the thing that's confusing on that is economy's not homogenous. All sectors of the economy aren't the same, and what's happened over that 40 year period is this really sharp divergence between different parts of the economy. + +There's a chart, and we don't have it in front of us right now, but there's a chart that maybe you could link to. You've probably seen it before, which is this chart, the economist Mark Perry posts updates on a regular basis. He calls it the chart of the century. And it's the chart of price curves by sector in the economy. It's average prices across these huge economic sectors. And basically what you see is, it's a Tale of Two Cities. Basically, there are a whole bunch of sectors, let's call them the super fast change sectors, if you will, or kind of super fast growth sectors, or super fast sectors getting disrupted on a regular basis. These are sectors like computers and media and retail, and by the way, cars, and clothes and food, most of the stuff you buy. + +Those sectors basically over a 20-, 30-, 40-year period experience basically it's this really rapid of price declines. The price declines and then also and/or dramatic improvements to + +the product. The common example might be just the television set. If you spent $1000.00 on a television set 40 years ago, you got a 13-inch or whatever... A 21-inch CRT thing. If you spend $1000.00 on a television today, you get a 100-inch super skinny wall-mounted, this most amazing in-home cinematic experience you can possibly imagine. And in fact, they're having trouble in the TV industry even holding the prices at $1000.00. A lot of the top-of-the-line TVs now are $300.00, $400.00. They're just spectacular. They're just amazing. That's representative... That's a great example, that representative of a sector that has just had a ferocious rate of technological change over the last 40 years. + +So, very rapid price declines. Very rapid product quality improvements. By the way, very rapid volume growth. Way more people have TVs today than had 40 years ago, and have way better TVs. I mean, 40 years ago people didn't even necessarily have color TVs, and now they've got these cinematic things. That's representative of, maybe call these the fast economic sectors. And by the way, in a lot of those sectors, there's been an enormous amount of job growth, so the amount of job growth that's taken place in making entertainment for those TVs or video industry] . Huge job growth in those sectors. + +Then you take the other sectors. We might call those the slow sectors. There are a bunch of those, but three really big ones. Housing, healthcare and education. And those sectors exhibit the opposite behavior, the opposite behavior of your TV set. The 100-inch TV set that covers your wall is going to cost you 100 bucks. The four-year college degree is going to cost $1 million. And it's like the four-year college degree is basically... Just think about technologically, it's an unchanged experience from 100 years ago. Getting a degree from a college university today, it's the same set of activities, same format, it's teachers in the classroom, it's written exams. There's been no technological change whatsoever, and prices have exploded. + +It's a great example of you've got a sector where you would think the Luddites would be happy, because it's like technological change is not happening but then you see these explosive prices. And then you kind of look at the nature of these categories. Let's say housing, education, healthcare, it's like, "Well those sound familiar. Why do those sound so familiar?"It's like, "Oh yeah, that's what we call the American dream. That's what we call being middle class." What's the definition of the American dream of sort of being middle class, upper middle class? I've got a nice house in a nice neighborhood. I send my kids to really good schools, and I've got really good healthcare. In those three sectors, prices are exploding through the roof. You have either very little technological change happening in those sectors, its actually possible you're actually going backwards. It's possible you're actually losing productivity in those sectors as time passes, because the administrative bloat is driving this huge increase in cost. Now, here's the thing. There's a bunch of things that are also sort of interesting about those sectors. One is very little technological innovation happening in those sectors. So one is very little technological innovation happening in those sectors. Houses don't really get built differently, hospitals don't really get run differently, and schools are exactly the same. Two is, the sectors are huge. These are giant slices of GDP. These are far bigger industries than like televisions or books, food. These are huge. Healthcare is a sixth of the American economy and on its way to being half. And then of course housing and education is also super expensive. This is the answer to the question you asked, which is the reason why they came to me as a whole is not going where you would think it's going is because these sectors are most of the economy. And then think mathematically what's happening, mechanically what's happening, which is that the prices of the fast sectors are falling and the prices of the slow sectors are rising, and the slow sectors are eating the economy. + +Mechanically they're becoming a bigger and bigger percentage of the economy. The non- + +productivity growth, non-price declining, non-technologically innovative sectors are becoming the economy. So this is the ephemeralization point. All this stuff that has a lot of technological change is becoming ephemeral and cheap, leaving everything else to become where all the money goes. And then you get to this kicker on it, which is housing, education, healthcare, the other thing they have in common is this very specific form of government intervention in those markets. And that government intervention comes in two parts. One is restricted supply. In all three of those sectors, there are very sharp government-driven restrictions on the provision of the supply to the economy and we can talk about that. But of course the problem with restricting supply is you drive up prices. Because you drive up prices, voters get very unhappy. And so then the government subsidizes demand. + +This is the phenomenon, you just take colleges as an example, you have a literal college cartel in the US. You have an accreditation agency that accredits colleges, universities, run by the incumbent colleges and universities. The accreditation is necessary to get access to federal student loans. Without federal student loans, you can't make these business models work in the current structure, but the consequence of that bottleneck of supply, which is supported by the government, which is enabled by the government, because otherwise they would just break it up under antitrust grounds. They let the federal student loan program be kind of leashed by it. Because of the restricted supply, prices rise. Because prices rise, voters get upset. Because of that, politicians put more money into federal student loans. But if you put more demand, if you put more money in to an economic sector with restricted supply, the result is prices go up even more. + +And so then prices go up even more and then voters get even more upset and then politicians subsidized even more and then prices go up even more. And so in all three of those sectors, they're in this doom loop that's going in the opposite direction of what everybody thinks they're worried about. They're not in a doom loop of technological innovation and automation. They're in a doom loop of lack of technical innovation, steadily escalating prices, steadily escalating government subsidies, right, with this incredibly restricted supply. + +And in that formulation, there's no limit to how much these things can cost. And there is no limit to how much a house in San Francisco can cost. There is no limit to how much heart surgery in a hospital can cost. There's no limit to what a four year college degree can cost. It can potentially cost infinite because the government will just keep subsidizing it. I look at that and I'm like, well, it's obvious what we need to do. We need to technologically disrupt those sectors. We need to do to the slow sectors what we've done to the fast sectors. We need more technological innovation. If we don't do that, this story evolves in to the exact opposite direction that everybody's worried about, which is what's actually happening right now. + +### Real Change vs False Positives + +Patrick + +That brings us through to 2021. You mentioned that maybe something has happened that might cause a shakeup in this inertia. I almost want to go example by example here because they're so big, they're so important, and it could give us a couple of different lenses to view what the future might look like. So I guess the first as an umbrella concept, what has happened as a result of COVID or in 2021 that you think is different? And then maybe we'll start with education to dig in a little bit on how we might attack this problem. Because to me, it seems like the most obviously ridiculous one. There's a network, there's accreditation, there's actually the stuff you + +learn, and we can attack those things with technology. But before we get to the example of education, I'd love just to hear your thoughts on what really has changed versus what's a change mirage as a result of COVID. + +Marc + +I think the big thing is just system shock. Start by saying, look, we're going to talk about some positive things that are going to come out of COVID, so important to kind of note up front, COVID in general was very bad. It was very bad for a lot of people. It was very bad for a lot of families, had severe health consequences for a lot of people, a lot of people died. It was also very economically disruptive. And a lot of people have got very badly damaged by it. A lot of small businesses have gotten destroyed. So overall it's a big negative, and we wish that we hadn't gone through it. That said, it is going to have consequences, and I think some of those consequences are actually potentially quite positive. So the big one is just, it's a system shock. + +You know how life works is most days, weeks, and months are a lot like the ones that came before, but it just kind of rolls along and every once in a while you kind of tweak things a little bit, but generally tomorrow is actually a lot like today. And then you may, either as a worker or as a manager or a CEO or as investor or whatever, you're an entrepreneur, you may have some idea on how to do something radically different. Maybe I want to go live somewhere different from where I work and telecommute. That's okay, that's kind of weird and different. Or maybe I want to start a new kind of online retail business that doesn't even have a physical storefront. Or maybe I'm a CEO and I realize I actually have 10,000 too many people. Maybe I have 10,000 people answering the phones and I really am ready to replace that all with software but I'm worried about the effect of my company if I cut 10,000 people out of a customer support function. + +Or maybe I'm a restaurant chain, maybe I'm the CEO of McDonald's or whatever, a restaurant chain, and maybe I think it's time to actually build a big delivery component to my business, but I think my shareholders will get mad at me because it's a large amount of cost to get a delivery business up and running. So people have had all these ideas, I think for a long time, of things that they might like to do differently, they just haven't had catalysts. There's been no kind of environmental setting for which I think a lot of people could pursue the more radical ideas that they had. And so I think COVID as a system shock, as an opportunity all of a sudden for a lot of those ideas, first for a lot of experiments to get run during COVID right along those lines, which have been really interesting for a lot of people, both personally and businesses. + +But the other is, now it's like everybody has a cover story. If I ever wanted to reorganize my company, now is the time. If I ever wanted to fundamentally eliminate job departments over here and create new lines of business over there in a way that would've freaked people out, now, basically it's just like, oh, because of COVID, we're going to do this. If I always wondered, why as the CEO do I have all these offices and everybody commutes to the office and they kind of dress up in these kind of funny clothes and they sit there at their desks and they just stare at their screen and do email all day. Maybe they could have done that at home. Now I have an excuse to say, hey, we're not going to have an office anymore. And all of a sudden, instead of being a weird thing that freaks everybody out, it's like, oh, of course, in COVID we ran the experiment, and it turns out maybe we don't need these offices. + +It's this really unusual opportunity to take stock from first principles of how we think things should work and how we want to live and possibly drive very different conclusions. And so that's the kind of macro backdrop in that things get really interesting. So there's + +implications kind of all throughout the economy, but the really big one is just for the last, whatever, 3000 years, if you were an ambitious, capable young person and you grew up in a village or some remote location or whatever, and even a mid-sized city, if you wanted access to top-flight economic opportunity, you moved to the biggest city you could find. And this has been a long-running, this way predates technology, that's true of London and Venice and many other big cities for a long time, for thousands of years of core technology. Economic opportunity has always been in the city. + +And then of course, that also means there's always been this lifestyle trade-off, which is like, okay, I can make more money in the city. I'm going to live worse for the same amount of money. And then of course that problem has gotten particularly acute in the last 20 years because of all these crazy housing policies where the cities where everybody wants to live don't want to build more housing. And then of course, now you have just this general collapse of civil order and this incredible spike in crime. Decided to turn our superstar cities in to Gotham for some reason. Except apparently with no Batman. It's like a running joke out here. I have all these friends who like live in San Francisco, and I've been teasing them for years. It's just like, okay. I've been accusing my friends who live in San Francisco, specifically, the specific city of San Francisco. I've been telling my friends for years, you only live in San Francisco for the stories. You want the crazy stories. + +I came home last night and there were homeless people having sex in my front yard. Usually get people to kind of admit that they kind of like that up to a point. But then it's like, I took my three-year-old out yesterday, went down to the park and we got somebody, a junkie tried to stab my three-year-old with a used syringe. The jokes start to get less funny over time. You just have this generalized collapse of these environments. That was a problem that was already developing and of course that's really been exacerbated over the last 18 months. So it's like, okay, what's that trade off? What is it worth, a hellscape, in order to have this spectacular job. If you have to make that trade off, it's a difficult choice. + +But all of a sudden, lo and behold, you don't have to make that trade off. And maybe it actually turns out that remote work, and this is what every CEO found during the pandemic, is actually remote work... Remote work is not perfect. There's going to be a lot of changes that companies are going to make because of remote work. And there's some experiments and we could talk a lot about that. But generally speaking, it worked really well and it worked much, much better than anybody thought it would work. There's no company of any size that I'm aware of where any of the knowledge work, there's a production plant that might've gotten shut down for a while or something, but every company I know of that has knowledge work and knowledge workers, they all had good experiences with remote work across the board. + +If you think about just economy, no bank went down, no stock exchange went down, no internet company went down, nothing, no insurance company. This might fundamentally change what we thought we knew from 3000 years of economic history, which is this idea that where you work and where you live to be connected. And maybe fundamentally they really don't. And then at that point, now you're in a whole new world. Housing, as an example, right, it completely changes this housing question because it's like, okay, now maybe people are going to maybe separately optimize their work lives and their home lives. Maybe they're going to live in a very different kind of built environment. My favorite example is, if I were in the real estate business, what I'd be doing right now, which I'm not, but if I was, I'd be trying to figure out, find really nice midsize cities or larger towns that are really nice physical environments, really safe, good schools, just generally really nice places to live, and then build basically new kinds of multi-generational housing. + +Basically, compounds, neighborhoods where you could have 12 or 15 or 20 homes interlinked where you could have basically all the members of extended family. Every member of the family could have 3000, 4000, 5,000 square foot house and you could all be clustered together, and collectively you would be paying less than you'd be paying if you were all living in one bedroom apartments in San Francisco. Kids growing up in that environment might have a fundamentally kind of more pleasant upbringing, especially surrounded by extended family, but then they might still have access to all the top economic opportunities in the years ahead because they can just video in and they can get the internship at Google or whatever, and have the same career that they would have if they were in San Francisco. And that's the kind of thing that I think becomes possible now that might be a really fundamental break from how things have worked until now. + +### Future of Education + +Patrick + +You mentioned this idea of knowledge work and how swimmingly any company based on knowledge work went through this period, shockingly so. It brings me to education, which of the three categories, to me is by far the most confusing, because on the one hand you can learn anything you want online in much more depth and speed and fun than you ever could in history. Yet the cost is going up. How should we think about this? Should we think about it as that bundle of accreditation network and actual education? Should that be unbundled? How could we attack this such that, because on the other side you have those stats around if you go to college, everything's better. And I don't know which way the causality runs. Your incomes are way higher and your health is better and all these other things. So how should we square this circle? It just seems so confounding to me that we could have the internet and yet this cost keeps going up. + +Marc + +Yeah, that's right. So a couple of specific things happened over the last 18 months. So one is, my wife teaches at Stanford Business School, and I saw her actually go through this. In one week Stanford was like, everybody's going home. Professors, they basically said, which professors are willing to teach online? And it was like, whatever, half the professors were like, yeah, no way, I'm not doing that. And then the other half of the professors were like, okay, I'll give it a shot. For the professors that were willing to teach online, they literally had to construct their own methods and system for doing it. All of a sudden on the fly create a new way to teach courses online. And then their courses actually got a lot more popular because a lot of the other courses got canceled. They all got just thrown in the deep end of the pool for online learning. + +And again, it's one of those things, like it went better than anybody had a right to expect, but it was by no means perfect, because having a week to retool a class from in person to Zoom is not exactly optimal pedagogical methodology. You might want to have a chance to think more systematically than that. But again, it worked fairly well. It worked a lot better than people thought. The education system kind of ran this experiment. A bunch of things happened. I think one is, I think a lot of parents of K through 12 students, I think it's the first time parents saw what their kids are getting in the classroom at the K through 12 level, in many, many years. Most parents, if you're in your thirties or forties and your kids are in sixth grade or eighth grade, you were taught in the classroom 30 years ago, it turns out some things have changed. So the current curricula is quite a bit different at a lot of schools. + +I know a lot of parents were just shocked, absolutely shocked at the stuff that was coming + +across. There were these great viral videos that have gone, basically like teachers getting caught on video basically saying, yeah, this is a big problem. The parents are seeing what we're teaching. We have to figure out some way to get back to the classroom so they can't see it anymore because you don't want parents interfering with whatever's happening. From the extreme end you've got parents who are just like, oh my god, this is so much different than I thought it was. And by the way maybe you like that, maybe you don't. It certainly feels like we're on the front end of a pretty dramatic homeschooling boom the way some set of parents are like, I'm not sending my kids back to that. That's one set of things. At the college level, it's actually funny, the phenomenon you described, it really stresses this question of what is college you just mentioned. + +A sort of advanced view on college is, college is not one thing. It's a bundle of things as quote unquote education, but it's also a daycare for young adults. And then it's also like a social dating environment. And then it's also a testing gateway kind of thing on the way in. And then there's a certification, branding thing on the way out. So it's a bundle of these things. And so I think that the streaming experience, there was this meme that went out that was like, popular internet streaming services, Hulu, $60 a year, Netflix, $100 a year, Harvard, $54,000 a year. I think it really kind of stressed to people, it's like just sitting at home and getting a bunch of video lectures and doing a bunch of exams is nice, and I'm sure that's some of those classes are useful, but that obviously is not most of the value, to your point. + +Because otherwise you would just be watching courses on YouTube. You'd be getting basically the same thing at this point. So it's basically not that. So what do we know about what college and universities actually are? What we know, and Bryan Caplan has probably written the best on this, what we know is it's mostly the stamp of approval. And we actually know it's mostly the stamp of approval because of this thing that Bryan Caplan talks about called the sheepskin effect, which is basically this interesting question, which is like, suppose you go to Harvard for seven semesters out of the eight, but you drop out right before that last semesters. You have seven-eighths of the learning. Do you then go out in the workforce and make seven-eighths of the salary? And the answer is no, you make half the salary of somebody who completed that last semester. And so there's two possibilities from the sheepskin effect. One is all the actual skills are taught in the last six months, which is probably not the case, and at least if your college experience was like mine, that's not the semester where you're studying the hardest. + +So that's not the case. And so what's the case? Which is, it's the stamp of approval. It's the brand that you get when you successfully complete the degree. And then you kind of dig underneath that. It's like, okay, what exactly does that brand mean? And it actually turns out it means a couple of things very specifically, in my analysis, at least. It means that you passed an IQ test on the way in, which is basically, and it's quite literally, it's SAT, ACT, admissions process. You made it through the front door. And basically what happened was employers used to do IQ testing directly 40 years ago, and then it became very politically unacceptable to do that. So employers for the last 40 years have been laundering the IQ testing through the universities. They've been letting the universities do that on the front end, the university admission process, using scores like the SAT and ACT. + +But that signal persists, which is if I'm hiring somebody who's 22 who got a degree at a top school, I know they made it through that admissions process. That means they're smart because they had SAT or ACT scores high enough to clear that. So it's literally a laundered IQ test. And that's significant because the universities are currently in the process of dumping the tests, and they're dumping the SAT and ACT. And this is amazing to me. The universities are voluntarily dumping half the value of universities , right? + +Patrick + +Right. + +Marc + +For their own political reasons. So that's going to have consequences. We can talk about that. And then the other thing the sheepskin effect, the reason why the sheepskin effect is real, the deeper reason that that happens, the psychologists think, and people that study this stuff, is it's basically, it's a test for the personality trait of conscientiousness. Quite literally the personality trait of I finish what I start. I'm a responsible adult and I'm capable of finishing complex projects. And this is why graduates out of top schools are so attractive to employers is because it's somebody who's very smart, who has a demonstrated track record of applied conscientiousness. Those are the ideal employees. That's what you want for the highest end jobs. Those are the two qualities that you want. And so basically, I think there's an argument that a huge amount of the sort of activity at the university level is a kabuki dance around fundamentally these two tests, the test for IQ and the test for conscientiousness. + +And then I think if you put your employer hat on, you're like, okay, that works pretty well up to a point, but maybe that's not working as well anymore because first of all, these universities are dumping the tests, the IQ tests. And so I'm not going to get that signal anymore. And then the other is, is a four-year college degree the only way to prove conscientiousness? And maybe there are other ways to do that. And by the way, maybe finishing online course work is actually the same signal. And it's not the same signal today because people aren't used to it yet, and people are still getting used to this new world. But sitting here 5 years or 10 years or 20 years, do we really believe that the only way to kind of check that box that says I'm a conscientious adult is that I went and sat in a classroom for four years? It is hard to believe that that persists, especially with all of the other consequences of the educational system that are kind of getting so pathological. + +And that's the long run opportunity. And we're excited about that as VCs. We're excited about that. We're trying to back online education companies that are kind of behind that trend. And again, it's like, COVID doesn't change all of that overnight, but COVID's enough of a system shock where I think it's going to cause more people to be asking these questions. + +Patrick + +The job being done here in large part is reduced search costs for employers. And it's a kind of a workaround, like you said, a laundered testing. Why not start universities? Like why doesn't Google have a captive university and just own the stack, make it more efficient, make it more tailored for the actual jobs they need to fill? Do you think that might start happening, that big companies will actually create replacements for this system? + +Marc + +So it's starting to happen. And there are companies that have done this for a long time, GE for many years, I think they still do, they have actually their own management school, literally its own offsite facility. McDonald's actually has had this for a long time. They have this thing called Hamburger University. In theory, you can enter McDonald's as a line worker and that at some point you go to Hamburger University and you get literally the training and at some point you become a franchisee or a manager or an executive at the company. GM had, what did they call it? Kettering University in Michigan, in Detroit? It had a different name for a long time. And it was for GM, basically trained up their own people. There is a + +history of this. Arguably this is part of this general transition towards, I don't know, complacency or something over the last 40 or 50 years where I think a lot of big employers just decided that they could outsource all this. + +Here would be one interpretation, which is basically the US government decided postWorld War II to just flood the university system with money. It's like, okay, if that's going to happen, then we might as well just piggy back on top of that. That did work for a long time, but maybe it's just working less well now. So yeah, no, look, I think a lot of companies are going to do that. Google, I forget the latest, but Google has a program. They've at least announced, I think they have... I believe I recall correctly, I think Google has now formally removed college degrees from their recruiting process. I mean, it's actually funny, which is they've actually done that from the left. They've actually done that, it's part of their "equity" thing, which is basically it's "unfair" to discriminate against people who didn't have the opportunity to go to college. + +But it results in the same thing that we're talking about, which is, I think they have a way of testing and evaluation now for their incoming funnel that doesn't require you to have a college degree and you would get evaluated, at least they claim you get evaluated fairly, as compared to people who do have college degrees. They seem like they're pushing on that pretty hard. Look, I think the dumping of the SAT ACT, I think that alone, I think any employer with their head screwed on straight is going to be like, wait a minute. Now what is this signal? It's literally half the signal is being nuked. So I guess the good news is, how is your general conscientiousness test but the IQ test is being tossed out. That stresses, this is an incredibly politically charged topic, and so there aren't a lot of employers who can stand up in public and say, we need to start IQ testing again, or take the SAT as part of your application process to work with company. + +But they're going to need some signal. At the end of the day, they do need performance. And to get performance, at some point they need intelligence. They're going to look for some other proxy. We've got various companies that are operating in this space. One of them is our own company, Udacity has been doing this for quite a while now with this concept of nanodegrees. So here's another thing that you see. So here's another thing that happened where I think the education system went pathological. And this is always this thing, whenever people who run universities kind of talk in public, it's this interesting thing where the university presidents always seem to take great pains to point out that the purpose of the college degree is not to get a good job. The purpose is not to get a good job. The purpose is not filthy money. The purpose is not this and that. They sing these odes to living a good life and achieving social justice and all this stuff, and people grew up in different settings. I grew up in the rural Midwest, I always knew what the entire purpose of the college degree was to get a good job. I was never confused by that because I always knew what the alternatives were. I knew what the default was. It was working at a gas station which, God bless people who work at gas stations, that was not for me. + +But what some of these new private sector solutions have is they get away from maybe the flight of fancy of the whole thing and they get much more down to brass tacks, which is, okay, skills for a job for an employer. And then, by the way, it turns out employers are pretty interested in that, and so there's a lot of partnerships, and Udacity is an example where they have explicit partnerships with employers and they work with the employers upfront on crafting the curriculum. + +And it's like, okay, if somebody goes through this program and if they get these skills, they know how to do, let's say, mobile app development, and machine learning, and data analysis, then there are these jobs available and we'll hire these people for these jobs. It's a much + +more fundamental... I don't know. It's like the knowledge work at going to trade school or something like that. If you can get the good jobs coming out the other side of that, for a lot of people, that's like, wow, that's fantastic. And maybe I don't have whatever, and Ivy League school on my resume, but I've got a great job, and a great life, and a great ability to provide for my family and all the same skills, and it's great. + +And by the way, if I like philosophy, I can spend my spare time reading about philosophy, I guess, put it this way. The more abstract, and weird, and detached the legacy system gets, the more practical and pragmatic I think the private sector will be in filling the gap.And at the end of the day, employers need people who are smart and who can do the work. There is an opportunity here for a new kind of system that just works better. + +Patrick + +From a selfish investing point of view, do you think that focus of entrepreneurs on these big areas that have not seen the cost deflation, the slow sectors, can or will result in higher returns for technology like investments because there's so much white space to make improvements? + +Marc + +Good news, bad news. We talk about this all the time, so this is very much a good news, bad news story. The good news with these slow sectors is the markets are gigantic. The good news is that competitors are not good. They're generally not only not good, they generally are heading the wrong direction. They've got the Thelma and Louise strategy, it's driving themselves off the cliff for reasons I don't fully understand, but they're going for it. + +I don't understand how the universities are going to continue their claim on this amount of federal student loan funding with all their other behavior. I don't understand how that continues, but they seem on board with it. So the incumbents are headed in the wrong direction. And then, there's very little technological change happening, and so you would think that if you could inject technology and have a better way to do things, you could potentially build something very large. I think that's all true, by the way. + +The bad news is these sectors are just harder to operate for all the reasons we've discussed. They're very entrenched. There's all these systematic issues you have to work through and then, again, you go back to the government, the government is very involved. + +Take education as an example. If you're going to start a new education company and try to compete with colleges, universities, in a way you have this fundamental question which is, are you going to have access to the federal student loan money or not? If you do have access to it, you have to fit into the existing system in which case, can you really innovate? If you want to do something different, such that you're not going to have access to it, then you're at an immediate economic disadvantage to the incumbents because you don't have access to that money. + +So you've got this real, basically, economic disconnect for the business model that you'd have to figure out a way through. And so basically, what we find is these founders, the founders in the slow sectors are different than the founders in the fast sectors. + +The founders of the fast sectors are just like, "I'm going to build a better product. I'm going to take market. I'm going to focus on my customers. I may have some theory, but most of what I'm doing is just I'm going to hit the market as hard as I can with the best product I can, and I'm going to kick butt and it's going to be great." + +The slow sector founders have to be, I would say, much more sophisticated. More sophisticated because their business models are more complex and there are more issues. And so, they have to have a fundamentally a government relations kind of approach in the very beginning. + +They have to have some sort of more advanced theory on business model, different kind of fundraising plan. The company will stage capital differently, it might take longer to develop. There are more things that can go wrong. By the way, the incumbents are very powerful. The incumbents have achieved almost complete regulatory capture in these spaces. + +And so, new education startups should be ready to come under just withering assault from Washington or from Sacramento because all of the teacher unions, and all of the universities, and all of the people who are basically wired into those systems are going to just try to kill it. + +I don't know, it's hard mode. It's attractive to a certain kind of founder. Some of those founders are very naive and they just get crushed through the process. On the other hand, some of those founders are very sophisticated and very ambitious, and then there's this giant brass ring at the other end so they're going for it. + +### Hardware and its Role in Shaping Society + +Patrick + +Obviously, you've been extremely focused on funding, this software, eating the world trend. What about hardware? It seems that this doesn't fall in either of the categories we've talked about so far. Hardware as a technology has been a key part of the world story over the last 300 years. How do you think about who finances this, who builds it? What's important to build that we haven't? What are your thoughts on hardware tech, more generally? + +Marc + +Hardware is just more difficult. The easiest tech startups are always a hundred percent software. It gets more difficult from there. There are 10 things in hardware that can kill you as a company that just don't exist if you're just doing software. Well, the most notable one, especially playing out right now for a lot of companies, which is, because of COVID, there's this so-called supply chain interruption happening right now for chips, and so there are a lot of components. + +If you're building a hardware product, the product is usually made out of components from multiple places. There's a whole bunch of chips and other kinds of things you need to build these things. And then, the nature of hardware is if you're missing one of those components you can't build your thing. + +If your thing has 40 components and you have access to 39 of those, you're stuck, which is why Elon, for example, running Tesla, he's always put this big focus on, he's trying to verticalize their supply chain, he's been trying to bring everything in house as much as he possibly can specifically to get past this risk where any one vendor can hold them up either intentionally or unintentionally, because there's a bunch of hardware companies right now that can't build products because they can't get chips. + +By the way, there's another big issue, which is this is a startup challenge, which is when Apple goes out to buy chips in Asia, they open the negotiation with the check for $5 billion. They buy out all future manufacturing capacity of a certain kind of chip for the next three + +years, and then you show up as a startup and you want to buy 10,000 of those chips and you're opening gambit as a check for $2 million, you get laughed out of the room. It just puts you in hard mode. + +Many of the great entrepreneurial successes have been hardware startups because it is harder. Arguably, there's less competition, and then if you win, the prize is bigger. It's a harder road, but the best of those founders, they tend to be very good at fundraising. + +Let me put it this way. There's a reason why you end up with these multinational, basically industrial conglomerates. There's a reason you end up with these companies like, in the old days, GE, or in the new world, Apple, or whatever, that are operating in all these different sectors and do hardware at such massive scale, and it's because there is a big scale advantage in that industry so it's harder. + +And again, this is why software has been such a blessing in some sense which is, software as a technology has just really opened up entrepreneurship in a lot more sectors of the economy because it's really this , and so, I think generally that's been a net positive. I don't think we've lost anything because that's happened. + +I think hardware was always hard to do. It's just now we have something that's somewhat easier to do. A lot of hardware is commoditizing. There's a really interesting, actually, economic phenomenon that I've been trying to figure out recently, which is there's this concept of what's called the flattening of the supply chain. Another example of this ephermeralization concept, which basically is, well, just think of computers as an example. + +If you open up an old computer from 20 or 30 years ago, just the number of parts in the computer is just a lot more than the number of parts in a computer today. You pick up a computer today, and it's a huge concept now in the computers called system on chip. You literally have these single chips that have the equivalent of what used to be a dozen, or two dozen, or three dozen other chips, and they've just all been integrated onto a single chip. + +That idea, basically, that's happening broadly in the hardware world, which is basically the number of parts is shrinking, which is actually pretty exciting. It's basically like a lot of the constant stress and drama around global trade has to do with very complex supply chains with all of these different kinds of components that go into making a car or a phone, or something like that. + +And if you get these hardware products to the point where they just have a lot fewer pieces from a lot fewer places, then you can start to strip away a lot of that complexity. You can start to strip away more of this dependence on foreign partners that might not be reliable, and I also think there's this underlying thing that's happening, which is these supply chains are being flattened, and so maybe in the years ahead, it will be easier to build hardware just because you need fewer pieces. It'll still be harder than doing software, but maybe it gets easier. + +### Investing in Immensely Complicated, Large-Scale Projects + +Patrick + +Thinking back to this idea of factor productivity growth and the idea of just really complicated projects, your partner, Alex, had a great suggestion, which was to ask you how you think about financing things that are very complex. The specific from he recommended, which I love, is what would a series A round into the Manhattan Project have looked like? How do you think about + +this financing of very complex and therefore meaningful and worthwhile projects? + +Marc + +You'd do it by process of elimination, work backwards, which is why did you need Apple to do the iPhone? Why wasn't the iPhone a startup? There are a lot of big advances over the last 30 or 40 years. To your point, especially in these more complex fields and they come out of these big companies that entrepreneurs tend to be building the software. It goes back to this thing of, why do you have these large scale industrial conglomerates? + +Why can't you have more startups that are Tesla and SpaceX, is ultimately the question. Part of that is you read the story of Tesla and SpaceX and it's the story of just a couple of things jump out at you. One is this story of just unbelievable drama, and stress, and risk. Elon almost lost both companies at different points multiple times. Very tenuous whether those companies could raise money at this point in time. + +The classic venture backed financing model, you're raising your round of money every two or three years, you need to assume you're doing it with a new lead investor each time. If you need five, or six, or seven rounds of financing to get a complex product to market, but if you can't raise that third or fourth round of financing, it's game over, you lose the whole company. + +So I would say this is a criticism of Silicon valley that I give a little bit of credence to, which is these big, hairy, audacious projects that involve lots of moving parts, they are just hard to do with the classic venture financing model because of this need to raise multiple rounds of capital. + +What would be the alternative to that? What would be the counterfactual where you could have told Elon Musk up front, "We're just going to pre-commit $5 billion to build Tesla or SpaceX." By the way, oh, this is the other thing with Tesla and SpaceX, which is they're the only examples. Why are there a hundred more Teslas and SpaceXes? Why are there not companies building automated machine factories building thousands of houses for a fraction of the cost of today? Or why are there not these huge new universities or robotic automated hospitals? All these things that the world pretty clearly needs, where are the startups to do those things? + +Why do we only have Elon and why do we only have cars and rockets? And so it's like, okay, well, what if you had a system in which you could pre-commit $5 billion and you could precommit $5 billion to a project like Tesla, or SpaceX, or any of these other projects, and you were willing to do that because you knew that if the thing works, it would do what Tesla have SpaceX have done, which is you would generate equity value far in excess of that. + +So you put five billion into Tesla over the course of, whatever, its first 10 years, and then if it worked, it's worth 500 billion on the other side and you make a hundred X. That's a venture scale venture capital style proposition that actually looks quite attractive. + +So then it's like, well, what would it take to basically make an upfront commitment to somebody for $5 billion? Okay, that's interesting. So question number one is, could you raise the money to do that? And actually in the old days, the answer to that probably was no. In the current world, the answer that is probably yes. There are a set of high end VCs or private equity firms that could probably raise that money to have a program like that. There's this global savings glut, there's just trillions of dollars in accumulated capital out in the economy that just doesn't know where to do. + +There's 15 trillion of negative yield in government debt in Europe. There's these giant sovereign wealth funds that have no idea how to invest their money. And so, if they had a chance to invest in another dozen or a hundred attempts to have a new Tesla, they probably would go for it if you had a responsible plan. So, okay, could you raise some money to do it? And then, there's the what's the responsible way to pre-commit $5 billion, right? + +Patrick + +Right. How to you let the entrepreneur draw it or something? + +Marc + +Right, exactly. And there's an old idea that I think could be brought back and it's a term called project finance. And it's basically how you finance a dam, or a bridge, or a new highway, or something like that. It's tranche money. It's tranche money corresponding to milestones. You basically have a plan, the plan has very specific milestones along the way, the burn rate escalates over time. You've got these milestone check-ins over time, and then you pre-commit the money, but the money gets released in chunks so the money gets released conditional on progress. + +You predefine the plan in a much more rigorous way than the general venture-backed startup does, and then you unlock the money as you go, and then basically you would build a portfolio of these investments. So you would probably start, you'd raised some huge funds, you would nominally commit five billion each to a whole bunch of projects, and then you would basically take its course, and a bunch of the projects would not hit their milestones, hopefully relatively early in the process. You'd kill those projects. You would end up funneling most of the money into the set of projects that actually that were making the right kind of progress. + +And this is a doable thing, it's a doable methodology. It's actually like how, if you've ever read Augustine, what was his name? It was the former CEO of Lockheed wrote these great books about large-scale project management and it's basically, at least how they used to build the fighter planes, it was a very similar process. + +Patrick + +Not Skunkworks. Is it Skunkworks? + +Marc + +Skunkworks was a lightweight version of this. It was the venture capital labs version of this. I'm talking about the main line, how they did the main line projects, how they built the F-16. It's on a government contract, but it was on a government contract with milestones. So then, it's like, okay, what would it take? Because this is a different way of operating than the venture-backed startups. + +The way a venture-backed startups work is not like this. The way venture-backed startups work is they reformulate to plan on each race. At each step, they have a plan that's designed to appeal to the next investor, as compared to a plan that's designed to appeal to the investor who pre-committed upfront, and that's a very big difference. + +And then, you need a different kind of founder as a result of that, which is, you need a founder who's willing to basically preplan these things years in advance. You need somebody who's really systematic and really rigorous. You probably want to amp down a little bit on the creativity and wildness that characterizes high-tech and you want to probably amp up on the rigor and methodology side of things, and operating discipline side + +of things. And then, at least in theory, if you found a founder like that, you can line up against this. + +And then of course, that begs the question of our times, which is how many more Elons are there? How many more you Elons are there, and then, it's like, well, are Elons nature or nurture? + +Patrick + +Can we grow them? + +Marc + +Right, exactly. Where are the cloning tanks? How should they be trained? And then, how would you possibly find those people, evaluate those people? The counterargument to everything that I'm saying is basically the venture capital ecosystem, the way it exists today, did finance Tesla and SpaceX. So Elon didn't need this new method. He was able to make the old method work. + +Maybe we just need more people who are more ambitious, like Elon, and they just need to go make it work. And God knows there's no shortage of venture capital today so if you can't do it in today's venture capital environment, when could you ever do it? + +On the other hand, you go right back to the original question, which was why aren't there more Elons? There's something in there. There is some possibility here of, I would, say a major kind of conceptual breakthrough that it feels like maybe it's sitting right in front of us, but nobody's doing it yet. + +### A New Crypto Fund + +Patrick + +You are not in your firm a non-insignificant player in this question, and it's a good excuse to talk a little bit about what you're building with Andreessen Horowitz, the firm, I think a lot about velocity and vector for all kinds of companies, and it feels like most investment firms skate away from all the hard questions that we ask of the companies that we invest in. + +You're thinking about building the firm in what seems much more like an enduring, long-term way, and I'd love to hear you, I guess, first just frame the scope of the ambition for the audience in what you're building on the A16Z, and I'd love to talk a bit about how you came to these ideas that you're trying to deploy. + +The morning that we're talking, you announced both a very large new crypto fund, a new general partner, my good friend, David Haber, and it just seems like the velocity is incredibly high at Andreessen Horowitz, so I'd love to learn a bit more about the vector. What direction are you going? + +Marc + +There's the obvious one, which is just technology is neutral in some sense, technology is a tool in every case and it's up to people how people use it, and there are technologies used for very bad things, but on net, technology is good, on net technology leads to the world becoming better, as we've been discussing today. + +There are all these fields of human activity in which very clearly I think we need more + +technology and it's going to come from this kind of process, these kinds of entrepreneurs, these kinds of startups. So we're big believers that the world needs more of this and we're positioned to do more of this to help these companies get financed and built in the right way. + +So honestly, part of it is just conviction. We really believe in what we're doing and why it's good. There is a structural, I wouldn't say it's a secret plan, but I don't talk about it much just because it gets a little bit abstract, but it's a structural idea that we have underneath this. Your audience might actually be more interested in than a normal audience so I can go through that if you'd like. + +Patrick + +Yeah, bring it on. + +Marc + +This is going to sound really weird but we've thought about it a lot. I call it HP 2.0, Hewlett Packard 2.0. So if you look at what we started talking about earlier, if you look at basically the history of technological development, industrial production, basically up until the 1970s, the model was basically the industrial conglomerate. There were these companies, GE was kind of the classic for a long time, GM was an example. + +And then, even in the computer industry for the first 30 or 40 years of the computer industry, you had IBM, you had Digital Equipment, which was another one of these big ones at the time, and then you had HP. + +HP was the founding company of Silicon Valley, literally founded in the 1930s. It was put on ice for World War II. They resurrected it when they came back from the war and they built HP. And the point is between basically, call it, 1945 and 1970, I don't know, one, or two, or three, the way new technology products got invented and developed was primarily not by a startup. + +It was primarily by a new division of HP, or a new division of IBM. And so literally, Bill and Dave at HP, I'll just focus on them, they would pick a high-performing, up-and-comer engineer, or product manager, or general manager, and they would basically anoint that person and say, "You go forth and build the laser printer," and then that person would create a new division inside HP and they would go and do that, and it would either work or it wouldn't, and if it worked, they would feed it money and resources and talent, and if it didn't work, they would wind it down and transfer those people to other projects. + +And people would work for HP for their entire careers, but they would go from division to division. They would be in some divisions that were growing fast, they be in some divisions that were winding down, maybe starting divisions, so forth and so on. + +So, and then, from the perspective of HP itself, Bill and Dave, they basically ran, it's the classic organizational structure. It's the one that GM used to have that Alfred Sloan wrote his whole book about which basically is you have the central office and the central office basically has a set of functions. It's got this people, talent evaluation function, centralized HR. You have a common pool of talent. + +These companies were famous for moving their high potential people from division to division. I was an intern at IBM. If I had stayed at IBM, hopefully I would have been part of one of these programs, but I was an intern, I saw it happen, which is if you were an up-and- + +coming executive at IBM in the '40s, '50s, '60s, '70s, they were hopscotching you from business to business every two years for the first 20 years of your career, because they were trying to get you to be as well-rounded and generalist and skilled in as many areas of the business as they possibly could because you were a future potential CEO of the company. + +And so, they had this really serious effort devoted to leadership development talent. They knew the same thing for engineers, they knew who the really good engineers were, they would move them around. + +If you were in a business of HP that had a particular technical problem, they would transfer in an engineer from another division that knew how to solve that problem. So they knew how to allocate talent, and it was great. If you're going to work at one company for your whole life, which was the model then, you wanted your employer to know who you were and to care about you and to try to develop you like that. + +So that was great. They had capital allocation, they would decide where the money goes. They would have strategy office, so they would have some sense of the future market trends. They would have a sales overlay function. They would have large account management out of the central office. So they'd have relationships with other big companies. There would be somebody at the HP central office who owned the overall relationship with General Motors as a customer, or J.P. Morgan as a customer and had overall responsibility for that, so they knew what all the other big companies were thinking. + +They had an M&A office, mergers and acquisitions, regulatory affairs. They had branding, marketing. They'd have these big marketing campaigns about innovation and all the great stuff HP was doing, but then they'd have all these divisions. + +And then, in each division, the division would own the product. They would own the future roadmap for the product. They would often have their own sales force, their own marketing campaign, they'd have their own HR internal division. They'd have their own finance internal division. And they actually would give the divisions quite a bit of autonomy. + +And in fact, it's actually funny. They would actually put the divisions, for the most part, not in Silicon Valley. For the most part, the divisions were off in other cities. The same at IBM, same thing DEC, they'd have 20 or 30 other geographic locations where they actually had all these divisions running. + +And then, you'd have these general managers of the divisions who actually had a lot of control. They had a lot of control and autonomy over running the business. They had to report up to Bill and Dave, and they were expected to do a good job with the business and hit certain financial metrics and so forth. But they were like mini CEOs running their companies. + +And in fact, basically what happened was venture capital emerged in the 1970s. And basically what happened was venture capital strip-mined the talent out of these companies, especially the entrepreneurial talent. So basically, the founders and CEOs of a lot of the great tech companies in the '70s and '80s were either top general managers or top engineers out of HP, IBM, and DEC, venture capital basically was a siren song, basically saying, look, do what you're doing today, run a business, but instead of being a mini CEO, be a real CEO, and instead of working for the greater glory of Bill and Dave, work for the greater glory of yourself. Instead of having to constantly have your best people be stripped away from you to move to another division, how about you just keep all your best people + +forever? And why don't you build your own company? And by the way, if you're Mr. HP general manager, start your own company, we'll finance it, we'll venture finance it. You'll control it, you'll be the CEO, but we'll be a majority investor. And you know, you go do your thing. And by the way, if you grow up, ultimately maybe you could be the next HP, and then within your company, you can have multiple lines of business. + +And then the implicit underneath that was then at some point, other venture capitalists will come and they'll try to take all those people to start more new companies. And that was the bootstrapping process for Silicon Valley. That's why it exists. And that's why that conglomerate model kind of stopped working, which is basically the venture capital strip mines the talent and created this entrepreneurial ecosystem out of it. And I think the net result of that was good. Having 1000 or 10,000 companies in Silicon Valley is better than having three. Then we just have a much broader range of products and much larger number of jobs than would have if everything was still in these conglomerates. + +But there were advantages to that structure that just basically got lost. And so then you kind of say, okay, what if I'm just a startup today? So I'm just like a raw startup company, and I'm just starting, and it's just me and my friends are trying to build a product. What is it that we're lacking? And of course what we're lacking is all the stuff the HP central office used to do. We're lacking brand, we're lacking distribution, we're lacking money, we're lacking talent. We're lacking any stroke with the government whatsoever. We're lacking any ability to deal with press. We're lacking all these things. We're lacking somebody we can call when the shit hits the fan. We're just kind of on our own. We're on an isolated island by ourselves. + +Basically, you can kind of tell where this is going, is this idea of sort of HP 2.0, which is basically, can we construct in the context of a venture capital firm that is a minority investment and impact companies, can we construct the modern equivalent of what would have been that HP central office from like 1960, and provide our companies with that kind of overlay capability, with all of that reinforcement and brand and help and power that you would have had as a division of HP? But then of course it's a different deal. We're not in control. HP was in control of all his divisions. We're a minority investor, right? So we ride along with the founder. + +So it's a different kind of relationship. They don't work for us, but with the huge advantage that we have no need for strategic consistency. We're not an operating company. And so as a consequence, we are completely unbounded in terms of how many different sectors we can invest in. We're completely unbounded in terms of how many different companies we can invest in, assuming we can find quality companies. + +I'll just give you an example. We now have 250 companies active in the portfolio today where we own more than 10% of the company. HP or IBM at the peak of their power, they didn't have 250 divisions. They didn't have anywhere close to 250 divisions. There is a scalability in the venture model that in theory is much broader. We can provide a much bigger umbrella across all these companies, with the trade off that we don't have majority ownership. We have minority ownership, and we don't have control. + +And so yeah, it's sort of a back to the future thing, but completely evolved for the more appropriate model today, which is the incredibly aggressive world of power entrepreneur, who really wants to run their own thing. The last thing they want is a boss. We don't propose to be their boss, but we propose to do all the things to help them along the way that otherwise they'd be missing. + +### The Little Boy, Big Boy Problem + +Patrick + +I love the inspiration from HP. Do you take any inspiration from what I'll call traditional east coast financial firms, sort of traditional Wall Street banks or private equity firms, or it seems like a very, again, different set or way of doing things. Do you intend to borrow concepts from east coast and blend them into the west coast style of investing? + +Marc + +Yeah, so the east coast examples are, they're all from more than 100 years ago. So the answer is not from the last 100 years. Abstractly, what has been the role of the east coast in kind of American business culture or society for the last 100, 150 years? It's been basically consolidation, consolidation, professionalization, and scale. What has been the role of the west? The west is the frontier. And this is kind of how the country got built out. It's not an accident the coasts ended up in these roles. The west was always the frontier, up until ultimately the physical frontier was closed. Now we've created these new virtual frontiers. + +The west or the western mentality, and by the way, there are plenty of people on the east coast with the western mentality or with the frontier mentality, so there's no law that says people act a given way on either side of the Mississippi, but just generally speaking, you could say there's a west coast mentality that says let's invent something new. By the way, it's not just Silicon Valley. It's also, like LA is another great example. It's like LA is where all the movies are made. Make a new movie, that's an LA thing. Let's own the movie studio is a New York thing. + +It's the same thing in tech, right? Let's start a new tech company, that's a west coast thing. That's a frontier thing. Let's take that company public and then interrogate the shit out of it, quarterly earnings for the next 30 years, and demand, give them all the cash back and stop doing new things, that's an east coast thing, consolidation and control. They both have their roles. The American engine has been fantastic on both fronts for a very long time. We're best at it. + +It's actually fairly amazing, if you think about it, we're best at both. We're best at the new ventures and we're also best at the scaled multinationals. Our Fortune 500 multinationals are just outstandingly capable of doing what they do in terms of servicing customers all over the world. They're primarily run out of New York, and they for the most part have stopped doing new things, but they're great. They're at enormous scale and they've had a huge impact on the world and they generate huge amounts of cash. And then they get that cash back to shareholders who don't know what to do with it. + +On the west coast, we've been really good at the startup part. Of course, what the west coast has not been good at is the scaling, consolidation, control part, historically. My term for this, I call this the little boy, big boy problem, which is we're the little boy. We start these companies, we get them to a certain point, and then we hand them off. They get adopted, and they get adopted by new parents. Those parents live in fancy places in New York. And those parents have very different expectations, the parents have very different requirements. And maybe those parents are going to love our companies as much as we do, but probably not. Maybe they're going to want them to continue to do new things and invest in new R&D programs. Probably not. + +One of my ambitions for the Valley, west coast perspective, Silicon Valley kind of mentality, one of my ambitions is that we grow up over time and that we basically scale our ability to + +have our company scale. And this is one of the reasons we've gone so heavily into growth investing ourselves at the firm, is basically to be able to be with our companies for years later, beyond the classic venture capital stages. We don't need to do the handoff. There's an opportunity to kind of cross that, but we're going to give it a shot. + +Patrick + +Do you think that, obviously the traditional Wall Street firms, especially the younger ones, so maybe the west coast mentality ones, are more flexible, have gone from public investing down into private very aggressively and very quickly, and often they're more full stack style of investors, meaning they'll do debt, they'll do equity, they'll do all sorts of stuff in between. Do you think that the biggest Silicon Valley firms will go in the opposite direction? Meaning, you know, I'll take a radical example, like Andreessen has a portfolio that's all public equities or something like that. Do you think that you'll meet the east coast infringing on private investing with you investing in public companies? + +Marc + +Oh yeah, for sure. And by the way, the best of hedge funds that do private growth, the best of them are also working their way down to venture. They're coming earlier. We're working our way up. Right now, we're all private, but it wouldn't be crazy if at some point we decided to continue to expand up the capital stack. The big thing there, I guess what I would say, it's like we don't view it, we compete with some of these east coast firms from time to time on a deal. That's not most of how we experience them, though. Most of how we experience them is as partners on our companies. + +So generally speaking, they're investing in companies that either we've backed or that other firms like ours have backed. So we end up working with them a lot more than we end up getting with them. I would actually say there's actually been a lot more cooperative kind of synergy, as opposed to them being some huge threat. And I think it's worked out well for them. It's worked out well for us. The big thing that happened is I would say they are the aberrations of the east coast mentality. The big thing that happened was the east coast mentality kicked in really hard after Enron, particularly Enron and WorldCom. + +This is the thing that just amazes me about how this history played out, which is you have the.com crash of 2000. Yeah, there were startups that didn't work, but there was also this like assets, you know, most of the money was telecom. That's what it was called at the time, TMT, tech, media, telecom. Most of the money that was lost was telecom. Telecom was like another zero on top of tech. And then WorldCom was the big scandal. There were a bunch of others, but WorldCom was the big scandal. And then Enron was this kind of bolt from the blue, and they were in theory in the telecom business with this kind of bullshit bandwith trading thing, where it turns out the phones on the trading floor weren't actually hooked up. + +The Enron, WorldCom kind of became this giant scandal. And then there was this just like enormous freak out, by the way, justifiably so, by investors and then politicians and Wall Street and the regulators, and they were just like, holy shit, we can't let Enron, WorldCom happen again. And that led to these regulatory reforms like Sarbanes Oxley, and ultimately in the later years, financial services, Dodd-Frank, after the financial crisis. And then those regulatory reforms, it's like Sarbanes-Oxley was supposed to prevent another Enron or WorldCom. Dodd-Frank was supposed to stop banks from getting to be too big to fail. What ended up happening is those regulations ended up being massive subsidies to the big companies, and they ended up being massively punitive to startups. + +For AT&T to deal with Sarbanes-Oxley, it's easy because AT&T has 10,000 lawyers and + +accountants, 10,000 people in regulatory affairs. And they've got the whole system figured out already. And so it's just another process for them to work through. For a startup to deal with Sarbanes-Oxley is totally different things. It's just a much, much harder fundamental gate towards building a company. It's not the only one, because there are a bunch of other things like this that happen. + +The east coast mentality to prevent these big blow ups basically implemented a set of reforms that had the effect of punishing startups. The result of that was the startups basically said, okay, fine, we just want the public. We don't need to deal with SarbanesOxley and all this other stuff if we don't go public as early. And so we'll just stay private for longer because if we're private, we only have accredited investors and all these regulations don't apply and it's going to be all fine. And so that led to this phenomenon where all these private companies just started staying private a lot longer, but they needed the financing, obviously, that they would have gotten once they had gone public or done secondary and corporate market. + +And so then that led to the creation of this category of growth equity. And then of course the result of all this was an exacerbation of income inequality, because it was a removal of high growth stocks, young growth stocks, in the stock market, where anybody's retirement money can be invested in them. And it was removal of all of those new companies for public market and transfer only to people like us, who run private money. So the result of the reform was a withdrawal of economic opportunity from the population at large and the concentration of that economic opportunity in the hands of private equity managers. And so it's one of these like weird, the blow back- + +Patrick + +Reverse outcomes, right? Yeah. + +Marc + +It's a regulatory capture playing out in another model. Anyway, my point is the result of that was the east coast mentality won. They call it the New York, DC mentality, won in the sense of, they just basically put a choke hold on IPOs for 20 years. They got what they wanted, fundamentally. They got companies to stop going public. Companies kept developing, firms like ours scaled up to do more private capital to keep them going. And then to your point, some east coast investment firms, basically, I would say a small number of them have adapted, but it's like a dozen or so of those firms that adapted to come in to private tech investing. + +Most east coast firms, most money in the stock market, is retirement money. Most of that money is managed by public managers that don't do privates and can't do privates. And most of that money is in mutual funds or index funds. They can't do privates. They're not allowed. Or they're index funds, so they can't make any conscious choices whatsoever, as if that's a good idea. + +And so most of the money basically, it's really dramatic, most of the money that people need, whatever, seven, eight, nine, 10% gains over time on, to be able to finance their retirements, most of that money doesn't have access to the most dynamic parts of the economy. I don't understand how that makes sense. If somebody were looking at that on a sheet of paper today, I don't think you would want that outcome, but it is the outcome that it is. And the implication from that is that this private growth investing phenomenon probably has real legs. + +### Time to Build + +Patrick + +Two closing questions for you. We started with the idea, the observation that you had famously that software's eating the world. Last year, you issued one that maybe seemed less an observation, more of an imperative, which is it's time to build. I'd love you to just sum up that message. I think it's a great place to end, thought it was a great article that came at a very interesting time. What is this imperative? Why is it time to build? + +Marc + +Yeah, there was like the micro thing at the time, which was, I wrote it the week that New York City put out the call for people, New York residents, to take their rain ponchos to the local hospitals because they were running out of surgical gowns. This is on top of the ventilator crisis. And then this was on top of like, oh my God, are we going to be able to make drugs? And so it was just like, how has the most advanced industrial civilization in history worked herself into the position where we can't make surgical gowns? I kind of couldn't take it that week. + +The big, broad thing is it's a civilizational question, and it's actually the same. It's related to what we just talked about with the frontier. We had a culture in the United States from conception through to like 1970. The culture was basically production. The culture was production. It was achieving things. It was building things. It was construction, development of nature for the purpose of basically making the life of man better. It was unapologetic. And it was just like, we're going to do these things. We're going to build houses, and we're going to have those houses. They're going to have heat and then they're going to have this new thing called air conditioning, and it's going to be great because it's going to make everybody's lives better. + +And then something basically changed in the sixties and seventies where we culturally decided to really strongly tilt in a different direction. And I guess I would say that I think there were people with very good intent, strong arguments based on their own kind of moral views, that it was time to make that change. And that change is represented by, environmentalist was kind of the big one, but then there's other theories of basically just everything from animal rights to urban sprawl, all these arguments basically against building that emerged. + +Maybe it should be somewhere in the middle. Maybe there should be, I don't know, like a smart growth agenda or something like that. I just think we seem to have ended up too much in the ditch on this stuff. We seem to have ended too much in a state where basically we have a system, some people would call it a democracy , in which to get basically approval to build anything of significance, you have to get so many people to say yes that practically speaking, it's becoming impossible. Right back to housing, it's like, what would it take to build, the most obvious thing in the world is build another, whatever, million houses within the short commute distance to San Francisco, but it's just impossible. Inconceivable, with the way the political system is set up. It's just like, there's just no way. I mean, much less a new freeway, much less a new bridge, much less anything of any size and significance. + +I was involved in building a new hospital at Stanford, so I was involved in the original project planning. I was on the board at the time, the early 2000s. And then we helped finance it and the hospital was finally open a couple years ago. It was like an 18 year project. The process that you have to go through and the amount of money involved is just mind boggling. It's so + +crazily extreme and tilted in the direction of not doing these things. Maybe I'm sounding like a Kafka novel. It's just absurd, what you have to go through. And the number of authorities that have veto rights and the number of negotiations you have to have, and the things that get extracted that have nothing to do with healthcare, it's just crazy. + +The point of the essay was, it's time to rebalance this. It's time to get back to the idea of what counts is not the process. What counts is not the debate, the discussion, the this, the that, all the concerns. At the end of the day, what counts is did something actually happen? Did something get built? Did things get better as a result of that? I went on to argue, basically, it's like you could have a left wing view on this and a right wing view on this. The left wing view on this is to the extent the left wing argument generally is government should have more power, then the government should be scored more on its ability to build things. + +The right wing view on this is that the market should build things. But in that case, then if you're on the right, you really need to live up to being pro-capitalism, anti-corporatism, anticorruption, anti-regulatory capture, anti-monopoly, anti-oligopoly type crony capitalism. The institutional right has become, let's say impure, on those topics over the last 50 years, right? It's too entangled with big business. So I think both sides of the political spectrum could actually adapt to this. Interestingly, I got very positive feedback on this message, both from AOC's chief of staff, on the left, who thought it was great, and Kevin McCarthy, Republican Speaker of the House, who thought it was great. I succeeded in crossing the trade and appealing to both parties. + +And then actually, Boris Johnson, actually, Dominic Cummings actually picked it up there for about a year. The big messaging in the UK was Boris Johnson's podium was build, build, build. So we'll see where it goes. But at least I was trying to tilt the conversation at least a little bit in that direction. + +Patrick + +Well, I think it's a wonderful closing imperative. The last question I ask everybody is what is the kindest thing that anyone's ever done for you? + +Marc + +It's hard to say any answer for me other than Jim Clark originally picking me up off the ground in Silicon Valley and starting a company with me. Let me generalize it out, because I get super nervous and rattled when I have to talk about emotional things, as an engineer. The most amazing thing about the Silicon Valley ethos that people in the Valley all get and understand and people outside the Valley really don't get, and then I hear it all the time from people who come to the Valley, from people who come from LA or New York or DC, and they come to the Valley and they join one of these companies, they become part of this ecosystem, you always hear the same thing three years later, which is, I never realized what it would be like to be in an environment in which people genuinely want to help each other. + +They genuinely want to help each other advance in their careers in a way where it's not out of, because they want something from you in the moment. It's not because they have some hidden agenda. But every entrepreneur has experienced this. You know, I'm starting a company, it's like people come out of the woodwork and say, "Hey, can I help? Can I send you my friends? Can I do this? Can I introduce you to this investor? You're having a security issue. Can I come in this weekend and help you figure it out?" There's this incredible pay it forward ethic to the whole thing. + +My Hollywood friends, always, they know about this. They're always laughing because it's like Hollywood is famously the exact opposite. If there are two people in Hollywood and they're friends and each of them is trying to make a movie, the famous thing is one of them will happily knife the other. They'd rather lose the friendship than see their friend make the movie. + +Anyway, so I've been a beneficiary of this for a long time. I think part of it is just it's in the water. It's the pay it forward thing, which is other people have been nice and supportive to you and so you feel an obligation to contribute. But also a big part of it, and this goes back to this frontier concept, a big part is the sense of unbounded future possibility. So like in Hollywood, the reason why they're so mad at each other all the time is because there's only so much financing for so many movies. There's a very large number of people who want to make movies. There's only a small number of movies that are going to get made. And so basically it's like a zero sum kind of mentality. + +In the Valley, there's no practical limit on the number of new ideas for tech and there's no practical limit on the number of new startups. And by the way, there's no practical limit on the number of people who can participate. There's no practical limit on the number of people who can learn to write code, learn to do all the other things involved in building these companies. And so there's this sense and a track record that validates this, there's this sense of there's no limit. If I have a friend, and his startup succeeds, mine can also succeed. And in fact, we might succeed together. In fact, we might really help each other along the way. It's a positive sum view of the world. + +And then you just look at the history of the industry. That's how it's worked. For a place that's so kind of hyper capitalist and aggressive and sharp edged, it's just an amazingly helpful, supportive, happy, generous environment. I'd just say, yeah, I've been a big beneficiary of that for a very long time. And I try to, as best I can, to do that. + +## Big Breakthrough Ideas and Courageous Entrepreneurs +https://www.youtube.com/watch?v=JYYsXzt1VDc + +Transcript: +(00:00) [SOUND] Thank you very much for taking the time to come in and speak to us. Many of us, are aspiring entrepreneurs, so we'd really quite like to be like you. And, many others, would also like to pitch to you. >> [LAUGH] >> Actually sitting here makes me, gives me a sense of how intimidating that must be, so, I won't, I won't wish it for much longer. +(00:25) >> [LAUGH] >> And perhaps, perhaps we could, just start by outlining the, the three main topics I'd love to cover today. The first is your views on tech and venture capital trends. The second is, how you assess entrepreneurial DNA. And the third is your views on leadership and your leadership experiences, that, that you've had throughout your, your esteemed career. +(00:50) And so, if we could perhaps start with the, that first, tech trends, and go with something topical. You mentioned last month at the Goldman-Sachs conference, that tech was not in a bubble. Rather, it was in a mature deployment phase. And then the WhatsApp deal happened. And Mark is on the board of Facebook. +(01:08) So I just wanted to ask you, what do you think about that deal and how are you thinking about evaluations? >> So I, unfortunately, I can't, ten years, ten years from now I can come back and tell you all about the WhatsApp deal, but right now I'm on the, I'm on the, I'm on the Facebook board and I know that you all would not come visit me in jail. +(01:22) So I will, I will keep that one to myself. so, there's a couple of big things. So, just in terms of thinking about what we've been through in the last 20 years in Silicon Valley, some people in the room are old enough, you may remember there was a bubble. and, it was a fairly big deal, in sort of 1998 to 2000, and there was a very profound crash, which was deeply traumatizing, for those of us who went through it. +(01:48) And then we went through this extremely long period of, basically, you know, years of pain followed by then, sort of, what I think of as, as very slow recovery. I think it's actually been an object lesson in the psychology of markets and bubbles. I think that, people are much more highly sensitized to bubbles after a bubble. +(02:05) If you could be sensitized to them before a bubble, you could make a lot more money. But people get highly sensitized and so there's this phenomena of, of trying to close the, the barn door after the horses have escaped. And that, that is a lot of what all the bubble talk in the last, ten years has been about. +(02:20) And so we, we could talk at length about kinda why I think, in fact, tech is not now in a bubble and has not been in a bubble since 2000. the, the deeper thing, the more interesting is this follows a historical pattern, which is what I talked about at the Goldman Conference, which is based on the, the best thinker on this topic is an, is an economist named Carlotta Perez, who wrote a book called Technological Revolutions. +(02:38) It's probably the single best book. Like, that book and The Innovator's Dilemma are probably the two key books that are really critical to understanding how this industry works. And so she describes in her book, she describes a general model for the deployment of new technologies. And then how technologies intersect with financial markets. +(02:54) And so she's got this whole thing, and it's basically this multi-generational process. And there's what, it's basically these two big, sort of phases of it. There's what's called the installation phase and there's what's called the deployment phase. And it turns out in every single case and this includes railroads and, like, lots, electricity and steam engines and lots of prior new fundamental technologies, there's always this just gigantic bubble and then crash kind of halfway through. +(03:18) And historically that marks the transition from the installation phase to the deployment state. The deployment stage, you could argue, is where the actual interesting thing, things happen. It's where all the tech-, all the new technologies actually start to work. They actually make it into everybody's hands. +(03:30) They actually become cost effective and we actually find out how to actually use all these things. And so that's the phase I think we're in, in now. You know, without talking about the Whatsapp deal in particular, it is interesting to note that the companies that people think are overvalued today, generally either have billions of dollars of revenue, which was not the case, in, in, in, in the 90s. +(03:50) For example, Facebook, people argue Facebook as an example. Facebook went from $0 to $10 billion of revenue in less than ten years. And so that is definitely not what happened in the 90s. The other thing is the companies that people debate today, for the most part, have extraordinarily high customer, count. +(04:06) user, user count. Market sizes have expanded gigantically and so you've got these things now that people are arguing about that have, in some cases, a half billion users, on their way to a billion users. And if people want to take a position that you can have a large scale internet service that's worth a billion users that's not gonna be worth anything, you, you could take that position, I'm not sure you would recommend it. +(04:26) >> Yeah, no, that makes sense. When you, as you say, when you look at the, the cost per user, it's actually only $36, which is much, much less than in many others for the What'sApp deal. But another thing you, you previously mentioned was that, MBAs flocking into the tech sector is a sign of the bubble. +(04:42) So to play devil's advocate. >> Yeah. >> Many of the people here are flocking to the tech sector. >> Yeah. >> So, could that, perhaps, be a sign of a bubble? >> So things are heating up. And so, [LAUGH] Historically, there's actually been, and I suspect everybody in the room knows this, there has been a direct correlation between, PE multiples and, MBAs. +(04:59) tilting, tilting, tilting into the, tilting into the, the tech industry, for sure. So I think something different is actually happening. I think something different is happening with how companies are getting built. And maybe I can do the long version, kinda the, the slightly long version of this, which I, I think there's actually a whole new, a whole new way companies are being built in the last ten years and, and I think that business people and MBAs turn out to be very central to it in a way that's different than the past. +(05:21) So I kinda divide the story of how technology, the great technology companies got built kind of in the three phases and I think we're in the third phase now. The first phase was in the 40s, 50s, 60s, 70s. And it was so crazily hard. If you talk to people who were in business then or you read the stories, it was so hard to build a new tech company. +(05:36) It was such an unbelievably, sort of exceptional thing to do that you, you, you only really have these really extreme characters who, who would do it. And there were a pretty small number of them. And they were extreme, extreme characters, like they were, they make all the current, like, high octane entrepreneurs look like wusses. +(05:51) And the ones I'm thinking of, Thomas Watson Senior. If you want to read, like, what it's like to work for somebody who's harsh, read the book on Thomas Watson Senior. You know, he makes, he makes all of today's entrepreneurs look like cream puffs. >> [LAUGH] >> He would just literally sit in his staff meetings for like five hours and just scream at his, scream at his guys, there's just this, then he built this astonishing company, IBM, off the other side of that. +(06:12) David Packard. David Packard, actually, was quite a character. He, David Packard, people now remember for the HP way and for kind of that whole warm and fuzzy, you know, kind of approach to running companies. When, when David Packard was actually running HP, he had two nicknames. One was Pappy, which is kind of what people remember in a kind of paternal instinct, type. +(06:30) His other nickname was the Mean One. And he similarly would just, you know, tear people apart. And then Ross Perot is my favorite example. Ross Perot built the first great outsourcing company, one of the big tech successes in the 60s. And of course, you know, he was fantastic as a business builder when he came into contact with the American public, people went, what? and, you know, again this sort of extreme personality. +(06:50) So you get into this, this kind of, this sort of will to power thing that was happening. and, by the way, the VCs in those days, I think, were very similar. Tom Perkins, who's become re-famous again lately, you know is, is the same kind of character. He's, he's an ex-, he's a very, very extreme character and, and, and he always was. +(07:05) But that's what it took, you know, for him to do what he did in the 70s, and 80s in venture capital. So those were kinda the extreme days and then I think both VC and entrepreneurship, tech entrepreneurship, sort of professionalized, and so you had a lot of VCs then. And this includes great VCs, John Doerr, Mike Morris, Jim Breyer, you know, who are business people or investors first, and, and never ran companies. +(07:26) And then you have this kind of move through the 90s where you had this kind of default model where the one thing everybody knew was that founders couldn't possibly run their companies. And so you would have a founder and then you would basically promote or fire them to chairman or CTO and then you'd put in a professional CEO as fast as possible. +(07:41) And I think what happened is that model just got extreme. And i think by the late 90s in the Valley, we were mostly building companies that were kind of shells, or, you know, kind of like puff pastries of companies where, you know, they didn't really have, at the height of the bubble in '98, '99, the products that were getting built for the most part weren't very good. +(07:58) And these companies were kind of on this bomb run to get public as fast as possible, and you had all these catch phrases, like go big or go home. Or my other favorite one at the time which was, forget details, just do deals. And so you have this really kinda mercenary, hit and run approach to building companies. +(08:13) And then all those companies vaporized after the crash cuz it turned out they didn't have valuable products. They didn't have deep engineering capability. And then all the engineers who worked for those companies hated working for those companies. Cuz they were completely sales-driven, sales-led, these kind of mercenary kind of exercises. +(08:27) At, at the, at the height of, of, of how bad it got. Now I think you've got the exact opposite thing. I think the pendulum has swung all the way in the other direction, which is, now we all understand and take for granted, founder CEO, technical founder CEO is a good thing. You know, Mark Zuckerberg is kinda the apotheosis of kinda the, the idea that we have now. +(08:44) And so now what's been lost for a lot of the entrepreneurs. A lot of the entrepreneurs are engineers, but not business people. Now what's been lost is a lot of the actual art of building a business. and, in particular, what's been lost is the art of sales and marketing. And a lot of today's founders, one of the big issues we deal with is they're very technical. +(08:58) They're very product-centric. They're building great technology and they just don't have a clue about sales and marketing, and what's more is they almost have an aversion to learning about it. It's almost like a post traumatic stress kind of thing, you know, like 15 years after the crash. +(09:11) And so now the challenge for a lot of these companies is how to take what are actually fantastic products and fantastic technology and then integrate in top-end business thinking, top-end sales and marketing thinking, and top-end operational thinking. So I think we have actually collectively have a huge opportunity to put the pieces back together. +(09:25) And I think that's what the next five years are going to be about. >> Could you see the role of MBAs in terms of helping scale through that sales and marketing function? >> Yeah, so, yes, definitely and, and, in fact, in the abstract, there is kind of two models, that are both actually working quite well. +(09:39) The kind of reference model now is the Mark Zuckerberg, Sheryl Sandberg model. And I work with Sheryl at Facebook and I tease her all the time. She's lost control over her own name. It's now become a proper noun. >> [LAUGH]. >> You know every 24-year-old technical founder, you know, was like, I need a Sheryl. +(09:53) And I'm like, so do 400 other people. Unfortunately, human cloning is not quite at the stage yet where we can fulfill everybody's need. But basically the model of a very high-powered business person with deep capabilities in sales, marketing, and operations, who's able to partner as a number two, as a president or COO, with a technical founder, CEO, when you have somebody like a Mark Zuckerberg. +(10:14) So that's one model that works very well. And one of the interesting things about the last five or ten years is more and more of the top end business leaders in Silicon Valley have figured this out. And, like Sheryl, have chosen to partner not as the CEO, but as the president or COO with a great technical founder and build great companies. +(10:29) A recent example, Dennis Woodside, who's a top-end Google product or business executive, just left and became number two at Dropbox to Drew Houston, who's another one of these guys. And so that's one model and I think that's a very exciting model and I think it's working well. The other model is what you might call sort of the Bill Campbell, Scott Cook model. +(10:46) Or maybe the Dick Costolo, model as sort of the other example, which is, in the case where these companies don't have a founder who's capable of being CEO or who wants to be CEO, to have a business person, become the CEO, but with the sort of, with a much more advanced understanding of the role of founders and the role of product strategy and technology strategy than I think the professional CEOs got into in the 90s. +(11:08) So, and, and this is the, I, I describe this as the Bill Campbell-Scott Cook model because that's maybe the best example in the history of the Valley. Which is, you know, Bill Campbell, probably well known to the folks in the audience, you know, is not himself a technologist or a product person, but is an outstanding operator of businesses, has profoundly deep respect for founders, and has profoundly deep respect for products and, and for technology. +(11:29) And always makes it a point in, in his career, he's always made it a point to partner with the engineers as opposed to be threatened by them or feel like, you know, they have to be, you know, in the case of the technical founder, they have to be forced out. And, of course, Apple, Apple over the years has been a case study of this, and, of course, Bill came up for Apple and so he saw this. +(11:46) And so you kind of contrast the now legendary kind of John Scully-Steve Jobs model to to the Bill Campbell-Scott Cook model and you kind of see how, you know, kind of where that came from. And so that's a model that can also work very well. And so as, as the folks here think about as you build your careers, and think about these things, I think if you're gonna be in the tech industry, the really key question, you know, it might turn out either way, but the really key question is what's the partnership that +(12:07) you're gonna have with the technical visionary, in the company who will often be a technical founder? And I think if you can crack that code, I think there is just an enormous opportunity to, you know, to have one plus one equals three. >> So at, at Andreessen Horowitz, the, the VC fund you founded, you invest in many of these founder COs. +(12:25) They all want a share-all? >> Yep. >> And finding a share-all isn't, isn't necessarily that easy. And you've built up a, a very disruptive model within the venture capital industry where you provide a lot of value-added services including hiring and marketing, to portfolio companies. +(12:39) Could you talk a little about, how you came up with that disruptive model and what opportunities you see going forward to continue shaping the VC industry? >> Yes, so my partner and I came up as entrepreneurs kind of in the phase where the assumption was that you fire or demote the technical founder and you bring in the professional CEO and you become a sales, sort of a sales-driven company. +(12:57) And so, and we, we kind of, we have a lot of experience with that model. Like I said, sometimes it works sometimes it doesn't. But we thought, this is, we started our firm, we planned our firm in 2007-2008 and started it in '09 and our basic take was there was an opportunity. Many of the other venture capital firms had tilted hard in the direction of sort of professional sales-driven CEO, we decided to tilt hard in the direction of technical founder CEO. +(13:20) And so, we basically said, how would you build a venture capital firm optimized for a technical founder who wants to become a CEO? That let us, and by the way, not religious, that's not the only thing we do, but like, how would you center the culture of the firm around that idea? And I'll, I'll come back to the other part. +(13:35) So we kinda decided on two things that would come out of that. Which is, one is, if you have somebody running a company who has not run a company before or has not, maybe, necessarily been a manager in some cases before, or, in some cases, maybe has not held a job before, they become CEO of their own company. +(13:52) It really shines a very bright light on the, the background, and caliber of the general partner that you're going to propose to put on their board. And we just made the decision that, and there are many different kinds of successful VCs, but we just made the decision at our firm, the general partners will be people who have built technology startups before. +(14:08) And so I think at this point, seven of our eight GPs, I think I'm the only one who hasn't actually been the CEO. I think seven of the eight of our GPs have been a, a CEO and I think five or six of the eight now have been founders. And so, sort of by definition at our firm, you get somebody on the board who really knows, has been through the war. +(14:26) Really understands, you know, what things are like. And so, when something goes wrong, and, you know, things are just, like, horribly, like, crashing. You know, the key engineer quits, or the founders can't get along. Or the biggest customer dumps you. Or a competitor comes out with a much better product. +(14:39) And all these really horrible things that happen. You can't raise money, that we have somebody in the board seat who is a really good advisor and can say, I was in that situation before, and I can tell you what doesn't work, cuz I probably made all those mistakes and then I can tell you, you know, gives you some advice on what does work. +(14:53) So that's one, and then the other thing we said was we said, okay, well what's, what's the, you know, we sort of thought about what are the reasons why? You know, if, if a VC brings in a professional CEO and fires the founder, why do they do it? Part of it is likely experience. The other thing a professional CEO brings in, is very, very deep network. +(15:09) If you've been a, you know, VP or general manager, or CEO in Silica Valley for 20 years, you have this enormous network of executives who you can hire and engineers, and recruiters and you know all the reporters and all the editors. And you know all of the customers. You know CIO's, and CTO's and you know how to go sell the things and you know all the VC's you know how to raise money. +(15:30) So you just have the business people in the valley, who have been in the valley, have this just giant network of people and these technical founders often don't because they've often been heads down. You know, writing code most of their, most of their lives. And so basically, what we decided was, let's preconstruct the network that will basically, where we can take a technical founder, inject it straight in the network, and sort of give them super powers of a network that's comparable to what, John Chambers might have. +(15:51) And so, and, and that's been a very big effort on our part. We have about 60 full-time professionals now, across five operating teams in the firm. That are not GPs, but are full time professionals organized around the different areas of the network. So, sales, business development, corporate development, marketing and PR, executive talent, engineering talent. +(16:08) And so, as an example when it comes time to find a sharer as a consequence, we, waiting for, we are trying to build very deep relationships with all the sharer of both genders, throughout the valley just a sort of a normal part of sort of art network building exercise, and then when we have companies that are kind of maturing to the point when they need somebody like that, you know, as one example, we will know who those people are and we'll have kind of you know, very easy access to them. +(16:30) You kind of help them to sort of bridge the gap. >> No that's, that's very understand, helpful understanding how you lean in once you've identified the founders. >> That was good, that was good. >> Stop it. >> That was good. >> The how do you actually identify them initially? And so what do you think are the, the traits that founders have and I'd also love to hear about some of the best and worst pitches you ever heard. +(16:55) >> So the, the basic math is the, so there's a basic math component and then there's the, all the intangible. So the basic math component is there's about 4,000 start-ups a year that are founded in the technology industry that would like to raise venture capital. We can invest in about 20. +(17:10) So the falloff is significant. we, I like to say our day job is crushing entrepreneurs' hopes and dreams. We actually have focused very, very hard on being very good at saying no cuz that's mostly what we do. We see actually, we see 3,000 inbound referred opportunities a year. We narrow that down to a couple hundred that are taken particularly seriously. +(17:31) And I would say there's kind of this very interesting kind of process where there's you know, say the hard thing is deciding which one's we're going to invest in, because we can just invest in so few. The somewhat easier thing actually it turns out, this has been a surprise, it's actually after you have been in it for a while, the thing that's actually fairly easy to tell is, will this team and company be fundable by a top VC. +(17:49) Will it get funded by a top VC. It may be, it may be Sequoia, or Excel, or Greylock, or who, who knows who it is, but you know, does this company kind of clear the bar? And I think the way the math works, basically, is, you know, there's about 200 a year that are fundable by top VCs. That, that, that get funded. +(18:05) By the way, within the 200, about 15 of those will generate, you know, 95% plus of all the economic return. So just cuz it gets funded by a top VC doesn't mean, it, even the top VCs right tank, you know, generally, about half their deals. So even if you get funded by a top VC, it's not complete validation. +(18:21) So about 200 a year that are kinda fundable by top VCs. We can fund 20 and then 15 of them actually generate all the returns. And so, it's kind of a white knuckle thing when it gets right down to it, to try to make, you know, the picks. And if there's one thing that's frustrating in this job, that every VC deals with it's, you know, you miss most of the big winners, right? It's like, the thing all the top ventures have in common is they did not invest in most of the great successful +(18:43) technology companies, which is an incredibly frustrating thing. So, that's the basic dynamic, and that's the framework within which, you know, people come in and pitch to us. At the heart of it, there's two things which we look really, really hard for. I mean, there's the kind of surface level stuff you look for. +(18:59) So if you look for a huge market, you look for, you know, differentiating technology. And you look for you know, incredible people. I think in practice, I think that we collectively and certainly, we specifically and then we collectively, VCs, I think we probably, we, we spend a lot of time talking about markets and technology and we have lots of opinions and I'm not sure that those opinions are actually all that relevant, all that often. +(19:21) I think probably, the decision ultimately, is and should be around people. As like 90% of the decision. The two things we really zero in on on people are, you know, two things. They sound simple and they end up being very difficult: courage and genius. Courage is the one we talk about a lot because it's the one that people can learn. +(19:40) You know, courage which is to say not giving up in the face of adversity. You know, just being absolutely determined to succeed, you know, is something that, you can, you can like, force yourself to do. It can be very painful, you can force yourself to do it. The genius part is a little bit hard to force yourself to do. +(19:54) You know, courage without genius might not get you where you need to go but genius without courage almost certainly won't. And so, we're looking for some kind of magic combination of genius and courage. You know, there's there's one of my partners quotes, he quotes Nichie a lot on these, he says it's, it's will to power, it's, it's you know, it's people who simply will not stop. +(20:14) and, and by the way, right, there's always been this kind of thing at Silicon Valley of like, sort of this, like I call it the failure fetish, right? Failure is good, right? Failure, you guys have probably all been taught this. Heard about this from a lot of people like failure is a wonderful thing, failure teaches you all this stuff, and it's great to fail a lot. +(20:28) Like, and we don't like buy any of that. We think that's all complete, complete nonsense. We think failure sucks. [LAUGH] We think failure is a terribly, terribly depressing thing to go through. We think success, on the other hand, is wonderful. You know, you wouldn't think that this is something you have to actually say out loud. +(20:46) But we, we do find it to be clarifying when we point it out. And so we are strongly biased towards people who are so determined to succeed that they just, they never give up, they never quit, and I think that's a huge part of it. And that's something we really look deep, we look incredibly deeply for, you know that's the kind of thing that's not listed on a resume right. +(21:04) That's something that has to be deep in somebody's, fundamentally deep in somebody's character and you have to see it in their backgrounds. >> And what has been your, talking of courage, what has been your most courageous moment, and perhaps the moment of which you're most proud? >> Oh, the moment, actually this is actually a good day to ask that question because my partner Ben's book actually came out today. +(21:20) So if you haven't bought it yet, number one on Kindle in management, $14.44. $14.44. [LAUGH] Makes a, makes a great birthday present for all of your friends. [LAUGH] He actually tells the story in his book, it's actually in his book, and it's it's when our, it's actually it's it's when our, it's it's, our second company, Loud Cloud, when, when got just taken apart. +(21:42) We started our second company Loud Cloud in September 1999. And it was classic, you know, we were, I mean it was fantastic. It was incredibly high rapid growth rate off of a standing start. Straight into the, you know, the the last six months of the bubble. You know, unprecedented growth. Cover of Wired magazine. +(21:56) On and on and on. We took it public in 18 months, and then just the world caved in, our entire business caved in. burning, you know, an enormous amount of cash, cuz we were, we had, created a company for much, much higher growth. And and our stock, ultimately bottomed out at half of cash. So our shareholders, made the judgement, that not only were we so incompetent, that we were not capable of justifying the amount of cash we had in the bank, but that we were certain to burn at least half of it before we would call it quits and just give them the cash back. +(22:25) Those were probably the dark days. And then, you know, NASDAQ, you know, start sending the D-listing letters and they send about one a day saying if you don't get your, your, your stock back above a dollar, we're gonna D-list you and you'll be on the pink sheets and we came within days from that. +(22:38) And so guiding through that, and by the way, guiding through that kind of thing at the same time it looks like the entire world is ending, that it looks like you know the tech industry will never ever recover. I think we, the thing we are most proud of is, is, is actually working our way through that. >> Fair enough that is a tough thing to get in. +(22:53) The You spoke about Ben's, Ben's book and, and Ben talks a lot about these, these challenges, the whole notion in the book is dealing with the hard things and have there been hard moments in your relationship and can you talk a little bit about that relationship? I remember once reading that, he described you as the Beyoncé of the relationship and him as Kelly Rowland, so how do you feel about being Beyoncé? >> Yes, yes. +(23:19) I'm hoping, I'm hoping that wasn't a commentary of my figure, that's my, that's my main, that's my. [CROSSTALK]. >> Or your dances. >> Or my dance moves. So, I would, let me, maybe brought it up, so it goes to the nature of business partnership. So Ben and I have been partners for 18 years, so I first met Ben in 1995. +(23:35) He actually tells the story, in the book, of how we met, which is a whole story in and of itself. It involves a lot of curse words. You know, we kind of describe ours, I mean we, we, we love each other and we, we, we do everything together, everything in business together. We you know we describe ourselves a little bit as the old married couple. +(23:50) Yeah, you know that, you see like out on the park bench in the park in the middle of the afternoon, sitting on opposite ends of the bench kinda staring at each other. They're always there, but they're not talking, you know, and maybe they argue every now and then. But they're there this year and they'll be there in five years, they'll be there in ten years. +(24:04) So you, you know I would say at this point, at this point it's troubly hard to untangle how the partnership works other than just we've been working together for so long. And so we have the we have a deep level of trust that comes from understanding each other very deeply, and so I know exactly what he's good at and I know what to defer to him on and I know, he knows exactly what I'm good at and what to defer to me on. +(24:24) And then both of us trust the other, you know. So we, we both know that we'll make decisions in both of our best interests, and there's never anything that's you know, advantages one of us over the other, and so, as a consequence, each of us are very comfortable, you know, essentially caving to the other on any topic, which I think is, is actually very helpful. +(24:42) That said we argue about everything and we constantly argue. and, you know, we often come at things from very different points of view and, you know, in you know, I don't even know how to describe. We, we just have different backgrounds from before we have kind of different reference points for how we think about things. +(25:00) He is a far better operator than I am, so he's much better at running a business. So for example, a lot of things, when we, when we used to run companies together. A big thing we'd argue about is, you know, I would, I, I, sort of I think what he would say about me is that I'm sort of abstract, so I think about things like products and strategy and business in an abstract way. +(25:16) And so for me it's like, okay, what's the right answer? Like, what, what's the, you know, what should we do? And then the way he thinks about it is, from an organizational standpoint, from a management standpoint, is, what I, what are we capable of doing? And so I will often propose things that, where he's like, you're out of your mind. +(25:31) Like, the entire, you know, yes, in theory that might be a good idea, in practice, you'll destroy the entire company if we try that. And I'm like well, that's a pretty good point. And then, you know the argument in the other way is you know, look, I know that the organization is gonna get challenged by this and I know it's gonna be hard and we might lose people, but it's so important that we have to do this thing, that we have to really push it. +(25:49) And so I think a lot of the theories that he and I have developed over the years about how to run companies are kind of at that intersection point of what's kind of intellectually the correct thing to do or the optimal thing to do and then the actual practical reality, of what can be done. He talks a lot about one of his theories that we use a lot at the firm is uses his book is kind of all about this is we, we call it, there are no silver bullets, there are only lead bullets. +(26:13) There's this, there's this temptation especially when you get into crisis. When you get into real, real, real problem, there's this temptation to think there must be a magic answer. Like, there must be some stroke of genius, you know. It's almost, it's like what you, I don't know. +(26:23) if people watch the new Sherlock Holmes, you know, TV series, which I just love. It's like, you know, Sherlock is gonna have, you know, no matter how dire it gets and no matter how like evil Moriarty is, Sherlock's gonna have that stroke of genius that's gonna save the day. And there's this really strong tendency to kind of think, that that, that that's out there. +(26:35) And we see a lot of entrepreneurs that kind of cycle through different silver bullets, and then they don't work and they don't work and they don't work. Ben's point is always, it's probably the answer, it's probably firing a whole bunch of lead bullets. And so the answer probably is, you know, the engineers working, you know, later at night for, you know, six months and, you know, getting the next version of the product out. +(26:54) And the answer is probably for, you know, the sales reps to go call in twice as many customers and try to close some more deals and the answer is probably to, you know, your stock price is low. You know, go find the investors who are willing to invest when you're trading at half the cash cuz it turns out they actually do exist. +(27:06) And then your stock goes up a little bit. People start to regain confidence. And so, I would say, I've, you know, I've certainly come around to that point of view a lot. And so whenever we work with entrepreneurs we often have very similar advice cuz it's kind of, it'll be tempered through the very practical realities of what you have to do to get through a situation like that. +(27:23) >> At yeah, that, that seems a, a very healthy argument. Perhaps I can bring you on to a, an argument that, that's probably much harder to deal with. I mean, in general you enjoy an incredibly strong public image. And, last week, Carl Icahn, rather than using either silver or lead bullet, used the kind of badly trained Gatling gun. +(27:43) [LAUGH] And when, he a, he brought up, as an investor of eBay and that eBay should divest PayPal and allege the you had a conflict of interest and could you speak a little bit about how it feels dealing with those sorts of allegations in the press. >> I think that the and by the way it's not just, it's you know, Carl has become very active in a bunch of Tech companies lately. +(28:06) It's actually not just Carl, there's a firm called Elliot Associates, it's a top hedge fund that's become very active buying beaten down tech companies. There's a bunch of others. Actually this, this is part of my theory of we're not in the bubble, which is when activists become interested in a sector it's because the PEs are low, because the cash balance is high and the debt levels are low. +(28:25) And so, it, it, you know, although I'm not, not, I'm not getting that much of a thrill out of the, my current level of personal engagement. This stuff is, I view this, all this activist, and, activity as validation of, of my thesis, which is, we're not only not in a bubble, we're actually still in a bust. +(28:39) Especially the big tech companies are still in a bust. Multiples are very low, cash balances are very high. It's the kind of thing where I think time will cure that. Because in time, PEs will expand. In time companies will invest more of their cash in their own business or find other things to do with it and the activist will go back to harassing steel mills. +(28:59) >> Yep. Fair enough. >> And oil companies and airlines. >> Yeah, it was, it was interesting to see that in, as you, as you blogged in 2011, he was advocating exactly the, the sorts of board management that he's now seems to be criticizing it. >> Yeah, Carl Ichan in 2011 was extremely enthusiastic about board nominees with conflicts of interest that came from his organization. +(29:21) And so I, I posted from this morning extensive exerts from his communications from that time. And so, as far as I'm concerned he can now argue with himself. [LAUGH]. >> And it's very helpful for us as, as students to understand how you actually spend your day. And, and I'm sure it's a, an incredibly busy one. +(29:41) So, thank you for taking the, the time out today. But could you perhaps describe what yesterday looked like? [LAUGH] >> Oh, yesterday? Good lord, what did yesterday? What day was yesterday? >> Monday. >> Monday. Okay, good. So Monday is butts in seats day for us. So Monday is all day partner meeting. +(29:56) And all the VC firms kinda have this in common. And so, like one of the really critic, when you're starting a company, there's all these really critical issues, like what product you're gonna build. [UNKNOWN] are going to go into. And he start a VC from what's really important and what your conference table gonna look like, and how comfortable your chair is gonna be. +(30:11) Because you are gonna be in those chairs for along time. So it's basically, it's, you know, sometimes, yesterday was like eight hours, it can be as long as 12 hours straight, of just straight meetings. And so, it's basically two things happened on a Monday. Well three things happened on Monday. +(30:25) So we have a, breakfast every Monday. Then we alternate breakfasts. We have a general partner breakfast, and then we have a general partner plus senior operating staff breakfast and we alternate back and forth, and we kinda do all the firm related things. And then we have back to back, what we call the all GP pitch meetings. +(30:40) And so these are the companies that we are most likely to invest in. And so the signal if you're raising pension money, if you get invited on Monday, that's a good sign. If you don't get invited on Monday it's not a good sign. And so, and we have, and I think yesterday we had three of those back to back. +(30:54) Sometimes we have four or even five. also, by the way, pitch early in the day not late in the day. Just helpful advice. so, we do those, and then and then we have, at the end of the day we have what we call [UNKNOWN] review. so, we actually do that twice a week. Most firms do that once a week, but we wanna move a little faster, so we do it twice a week. +(31:13) Mondays and Thursdays, and so that's basically a complete, you know, sort of, you know, basically a complete pipeline report, maybe possibly interesting. So, so we actually run our deal process like a sale, like a, we actually use salesfirst.com and we actually run our entire process and the firm like a sales first runs a sales pipeline. +(31:28) And so we have comprehensive tracking of kind of the stage of every. All the way from all initial inbound deals, all the way through the ones that are being closed. And so we kind of review that entire pipeline, and we actually have a team that you know, a team that manages that. And a great, a great operating person who who manages all that. +(31:46) And so we, we go through all that. And that's where we have. That's basically a long argument and we go through it and we argue about every company. And we argue about everything else we can think to argue. And then we go home and collapse. >> In those in those deals for the meetings, are there any technologies? I know, for example, you've blogged recently about Bit Coin and how excited you are about Bit Coin and the future of the news industry. +(32:07) Are there any particular technologies or industries that you're, you're most, excited about? >> So there's a two part, two part answer to that. And so, the first answer I will answer the question in the second part, but the, the, the first answer is, we, this is another kind of theory we have at the firm. +(32:23) So, there are venture capital firms that are very top down and thinking about markets and technologies. So, if you go inside a particular sequoia and Excel, and Bessemer, and I think Kleiner Perkins used to, and may still. The way that they, they, they actually are, those firms are actually very explicit about how they think about Protestant markets, and so they actually will have I think in each case, they'll run an annual planning process, where they will actually get together, you know, at the beginning of +(32:50) each year, and they'll literally draw a map of what they think markets are gonna look like. And, so, you know, and, and it, it's basically a value chain map. And, so, it's an interesting exercise to think about. It's like, okay, like for example, for networking, is that, like, fiber optics and communication chips, the family of such things like routers, but, then we get into things like, you know, ISP's and then ultimately lead into things like, you know, and other wireless businesses, whatever. +(33:11) Kind of draw that entire thing out, as a map. And then you basically have boxes for each of the product, product categories. And then, you know, a VC firm can invest in one company per category. And then basically the goal for the year is to put a name in each box. And so they sort of consider it a success at the end of the year if they've invested in the best possible company they can in each box that they've identified. +(33:32) That's kind of one extreme, we decided to be more on the other extreme, which I think is a little bit more what I would call the benchmark approach or, in the old, in the prior generation, it would have been called maybe the Arthur Rock approach. Which is basically, and, and by the way, we have all the same theories, like we can't help ourselves. +(33:48) We just sit around and talk about this stuff all day. But, we need to climb more towards the other side, which is basically, the big breakthrough ideas. The, the entire art of venture capital in our view, is the big breakthrough ideas. The nature of the big breakthrough ideas is that they're not that predictable and in fact often upon first contact they seem nuts and it actually turns out to be the case, now all the crazy ones also seem nuts so it's a little bit of a you know they called Einstein crazy but they also called Charles Manson +(34:14) crazy you have to be cautious on this stuff. But, the really, really breakthrough ideas often seem nuts the first time, the first time you see them and, and it's the fact that they seem nuts, can be a very positive signal because number one, it, that, that can explain why that thing already isn't being done by an existing big company, cuz it's just considered too strange. +(34:31) And then number two, you know, if, if it works, like if, if the bit flips at some point and it goes from being nuts to being like, oh, that's a good idea. Like, then, you know, those are the companies that could just explode. Could become just gigantically huge. And most of the, big ideas, the PC seemed nuts at one point, the internet seemed nuts, BitCoin today seems nuts. +(34:48) And, Airbnb seemed nuts Uber seemed nuts in the beginning. And so you kinda wanna in, in our view we have this sort of approach, you wanna kinda tilt into the really radical ideas. But by their nature, you can't predict what they're gonna be. And so what you basically wanna do is have as prepared a mind as you possibly can. +(35:05) And learned as much as you can about as many things as you can. And then basically enter as close to a zen like blank slate kind of state at the beginning, you know kind of zen, you know set ideal of kinda perfect humility, which is hard for venture capitalist. You know, sort of perfect humility at the beginning of the meaning basically saying teach me. +(35:23) And then they either, you know, they either do or they don't. But then, you know, the, the hope is, you know, if Larry and Sergey walked in, and they're like, I know this is the 35th search engine but this will be the one that works you know, you're open-minded enough to say, you know, yeah, that might work, as opposed to, you idiots, don't you know that that's been tried and failed so many times before. +(35:39) And so we're way more on the side of we've gotten sort of opportunistic trying really hard not to let ourselves be educated by the really smart entrepreneurs. You asked about the best and worst pitch meetings. The worst pitch meetings by far, are the rip. And, I mean, we try really hard to not have these get to us but. +(35:54) You know, snap trap for dogs. Like, it, it, you know, it's the I, I. [UNKNOWN] This is not a startup thing. I was giving an example. You see it in Hollywood, right? It's like one volcano movie works and then there's like 400 volcano movies. It's like how many freaking volcanoes? It's like 35 of the top hundred games of the iOS app store and now are like Flappy Bird clones, like, and so and, and, and Paul Graham in, in, in the adventure community Paul Graham calls this the Hollywood approach to, to startups. +(36:18) Which is, it literally is, you know, it, you know, is Airbnb for parrots. It's just, it's these infinite variations of all the successful ones that you get. And by the way, the ones that sound silly also just ones vertical search engines. When Google worked, so search engines went very deeply out of style when Google worked. +(36:33) And then there were vertical search engines in every single category, and, except for travel, they all failed, right? Cuz it turns out there was just gonna be a search engine. There wasn't gonna be a search engine for health, there was just going to be a search engine. and, so, it's all the variations and clones and kind of, the mercenary kind of, kinda hit and run, you know, kind of stuff. +(36:48) So, we, we try really hard not to sit in those, because those are very painful. The ones that are the most exciting are the ones where it's a, it's a really, really bright founder who's done a tremendous amount of work and completely understands the domain. And walks in with a really crazy idea and then in the course of an hour, can basically walk you through. +(37:06) Where we have this sort of concept we use called the idea maze, which is the really bright founders with these really radical ideas tend to go through what they call the idea maze. So they tend to have worked for years. Working their way through the idea to try to figure out how to get from kind of the initial crazy starting point to at the end, something that will actually work in the real world. +(37:23) And the really great entrepreneurs can walk you through the idea maze and make you understand the flow of thinking that got, got them to the point where they actually came out the other end with what is a great idea. And what's interesting about that is those are generally not, there's this kind of you know, theory in venture capital that you want back coachable entrepreneurs. +(37:40) The entrepreneurs who really have the radical ideas are generally not in a way coachable, they generally react with hostility to being coached. And so one of the things we test for, is you know, basically say, have you thought about doing it this other way? And what we're not looking for is the, oh, that's a great idea. +(37:58) What we are looking for is the stare that's just like, you idiot. Right? You moron. You've been sitting here listening, you know, this is them to me, you've been sitting here listening to me for 20 minutes and I've been working on this for five years, and you think you understand this so well that you can make me a suggestion and not only are you an idiot for thinking you can do that, but I will now explain to you in detail why you're that big of an idiot. +(38:17) We love those. [LAUGH] Boy those are fantastic, those are outstanding. and, so, and it goes right back to the combination of genius and courage I was talking about. And so in particular I mean part it's of us in the word, but part of us it's just the look on the face. We love that look, it's like caviar. +(38:36) >> So I want to turn it over to Q and A, so perhaps you can just finish with one question, which is you had built two very, very successful companies. And then in 2009 you decided to not run a technology business, and instead found, what's become the fastest growing venture capital business. People were probably telling you, you're nuts, like you just said. +(38:59) >> Oh, yeah. >> Yeah. Well we went, we went, we went around to see, we went around to see, cuz we'd worked with, we were working with all, all the, all, most of the big, most of the big VCs, so we went around to see all of our friends in venture capital, and tell them what we were thinking, and they all told us we were nuts. +(39:10) Actually with two exceptions actually, two of them are incredibly helpful, Jim Breyer and Neil [INAUDIBLE] were both just tremendously, tremendously helpful. And we are very grateful to them for all their help, but the rest of them pretty much told us we were nuts. >> So what, what drove you? And, and perhaps you could answer in this context when we apply to the GSP we all get asked to write an essay on the following question, what matters most to you and why? So what today matters most to you and why? +(39:32) >> So we're really deep believers in the power of technology. Like we, we, we, we think that the technology industry has made the world a radically better place in the last 70 years. And we think it will make the world a radically better place yet in the next 30 years, and we think it's just starting. +(39:43) You know, the fact that we've just now after seven years in the computer industry gotten to the smart phone, which is the first computer that can get to everybody on the planet. That everybody on the planet is gonna have one of those. You know, that is a you, you know, we're just reaching the point now where we're able to apply technology to a, a lot of really fundamental problems in the world, a lot of fundamental problems and opportunities. +(40:02) And so it, it, we, we just all feel or I, I feel like we have spent, we, I spent my career and then even previous generations of entrepreneurs the value they have spent their career getting to the point where we can now do the things that we can do. And so it's an amazing, it's an amazing, amazing time in terms of what we can do. +(40:18) In terms of starting a venture capital firm there is an aspect to starting there are some serial entrepreneurs who like starting seven or eight companies in a row. Dave Duffield is probably the best example of this. I think [UNKNOWN] is company number 8 for Dave which is amazing. I've done it three times, Ben did it twice. +(40:34) Well, it depends how you count but twice, two and a half times. it, for some of us at a certain point it starts to be like, you know, it's like you climb all the way up the hill and then you end up back at the bottom of the hill and you have to start the climb again. And so at some point you start to think, maybe there's a way to contribute. +(40:50) That has to do with helping people climb the hill, as opposed to being the person out in front. And so I think we've reached that point. And we're very, you know, we're very happy doing that. >> Thank you, thank you so much. Now we can have some smart questions. >> We're actually hosting the Future of Media Conference here tomorrow. +(41:06) So I was interested to ask you about the news business. Of course you've been tweeting and writing and thinking a lot about that lately. And I guess if I could sum up your views, your optimistic about the prospects for journalism to thrive using a variety of business models. So I'm a former journalist and a lot of the issues that you are thinking about now have been being discussed within the industry for maybe eight or ten years, a lot of you know, soul searching and, and reflection and experimentation. +(41:37) So what I'm curious about is why did you get interested in it now? Like what prompted you to get excited about that space and to start thinking and talking about it? >> Also, less from an investment standpoint, cuz we don't really, content's not content, generally, so we don't really, we're not gonna be making a lot of investments in, in media production. +(41:56) Just cuz it's a, it's a different field. We, we can talk more about that. My interest in news, in media, is sort of twofold. One is, as a, is a gigantic consumer of it, and somebody who thinks it's very important, and then two, because everything I've worked on my whole life gets constantly blamed, for the decline and fall of journalism. +(42:13) so, the Internet. Everybody knows, right, the internet has completely destroyed the news business. The internet has destroyed journalism. The democracy is in peril, and it's all my fault. so, at a certain point, I, you know, I start to, I get, get, the, the, and, and, on behalf of all the people I, I worked with, to, to, to build, to build things like the web, it's like okay, maybe it's, maybe, maybe it's not the new technology's fault at a certain point. +(42:36) So my observation is, and that I think it's a very interesting topic, so I think the news, and I focused on the news business, cuz that's kinda the, the, the pointy end of the spear, in terms of, you know, the thing that people are most worried about. I look at it as, as, I, I look at it, I do what very few people have been willing to do, I think, which is look at it purely as a business and basically say, and, and, so here's my basic position is. +(42:56) Our view of what the news business is, and about journalism is, is an artifact of a specific period of time, from 1945 basically into, basically 1945 to 2005, and basically post-war post-war US. If you go back to the news business, before World War II, if you start in Colonial days, and if you extend all the way to the 1930's, the news business worked very differently. +(43:17) It was a very successful business and it was, a lot of people were in it, a lot of people made a lot of money. But like, as an example this whole idea of objectivity, the journalist now take as kind of this, kind of, you know, kind of, purer concept that has to be maintained. Like there was really no such thing. +(43:31) Like I always say like in in, sorry to say, like objectivity's an artifact of an era in which news businesses were monopolies or oligopolies. In the days when news businesses were fully competitive, subjectivity was out it went. And they always say, you know, you're a scumbag, how can you say that, and I'm like well Ben Franklin was a subjective journalist and so stop calling Ben Franklin a scumbag. +(43:48) You know, he actually knew what he was doing. He was actually a very successful journalist, a very successful publisher. There's a great book on the news business in the Colonial era in the US called Infamous Scribblers, which was what, it was a pejorative at the time for reporters, which could come back into fashion. +(44:03) and, it's a great articulation of how the news business actually grew up in the US. And then in the 20s and 30s, it got really interesting with figures like Hurst and Pulitzer. And so, you can kinda, you can actually study, kinda, the historical news business, kinda through, I think through those two time periods and then you kinda look at post World War II, and you say, well what happened. +(44:21) Well monopolies and oligopolies got established and so for and it, it was sort of this era of centralization in a lot of parts of the economy. But it was very clear in the news business, you had, you know, you'd have one major newspaper per metro area, because of the cost of distribution. You'd have, you know, three TV networks, nationally, because of the limited bandwidth for VHF TV. +(44:41) You'd have, you know, a handful of local radio stations. You'd have a handful you only had a handful of magazines. You could only really have three general news magazines on newsstands cuz you just couldn't afford to distribute more than that. You scale, scale economics kinda ruled the day. And so, so the news business went into this mode where they kinda said, okay, we're a monopoly. +(44:58) We're a monopoly or an oligopoly. And if you're a monopoly or an oligopoly, it's incredibly important to stay out of antitrust trouble. And the best way to stay out of antitrust trouble is to not, to not make anybody angry. And the best way not to make anybody angry, is not have any opinions. And so therefore let's be objective about everything. +(45:09) And then we can we can just basically say, First Amendment objectivity, don't break us up. And that worked really well, as long as the distribution was controlled, which it was. The distribution was locked down. And then the internet showed up. And then the internet introduced basically took the legs out from under all the distribution monopolies and then all of a sudden there are you know millions of voices. +(45:30) So my point as a business person is okay that, that's the past, right, that's, that's, that's the old days that's over. We need to look back to what happened when these things weren't monopolies and oligopolies. We need to look back to the thirties and twenties. And back to the colonial era, and we need to basically think about how to build news businesses, media businesses, that thrive in a competitive market. +(45:48) And that has to do with being, you know, incredibly aggressive. That has to do with in many cases having a very strong point of view. It has to with, not with the idea that you're gonna be the only point of view, but you'll be one of many and you have to argue things out. You have to have the right cost structure you have to think about market segmentation and you have to, you have to do all the things that people do when they actually build businesses. +(46:07) The issue in the news business is that a lot of the executives in the business did not grow up in a competitive market and and so they just don't know how to do that. So, now what's happening is, the new entrepreneurs like Jonah Peretti at Buzzfeed, or Sarah Lacy at Pando, or, you know, the people, the folks who built, you know, you see a lot of this in the tech industry, TechCrunch and all these things. +(46:23) You know, a, a lot of these new things, they're, or, you know, what Pierre Omidyar is doing with First Look is, you're getting, now, very smart people who are coming in from outside with very fresh points of view, building very exciting things. And all the traditional journalists are like, oh yeah that doesn't count, but healthy business, healthy journalism you have to get the healthy business before you get the healthy journalism. +(46:42) And so, so then you look at market size and you basically say, well how big is the market for all this stuff? And it turns out the market for news is gigantic and it's growing very fast because so many people are becoming part of the modern world and so many people are getting access to information for the first time. +(46:55) And so the global market for news is going to be five billion people within ten years and everybody needs to know what's going on. And so the market's gonna be large. And so I think there's huge opportunities for market growth, but it's gonna be from companies that are, that are able and that are willing to compete. +(47:09) >> I'm looking for an internship in High Tech, and there is a great quote that software is eating the world. And what I think is going to be next, and if you were an MBA, which university would you go to work to for the summer? Oh, that's a good question. So I think that, I mean, so the big one, the big one, the big one's clearly. +(47:31) The big one's clearly on deck are healthcare, education, and financial services I think are the next three, kind of, giant sectors where software can have a revolutionary impact. We're actually putting, we actually started out saying we're not going to do any healthcare, was one of our things we, we had a bunch of no fly zones when we started our firm. +(47:47) No rocket ships no no flying cars no space elevators no, no drugs either in the drug development sense or any other sense. And then and then and then we said, no biotech no healthcare because it's, you know, it's a different category, and if you're gonna make you know, pharmac-, you know, industrial biotech companies or medical device companies, FDA approval, it's a different thing. +(48:10) We've actually got much more involved lately in the cross-section of health care and IT. And there's all kinds of interesting things happening. At the intersection of healthcare and, and IT and software. And, actually, there's things happening, actually, on the medical side, and a lot of that, it's very interesting, things now happening around genomics. +(48:27) And big data applied to genomics. And then there's another whole set of things happening around healthcare information. and, you know, making, like, health marketplaces work better and, you know, direct access for consumers to doctors online and all these incredible kind of new services. So, I think healthcare is gonna be really interesting and we're, we're diving in much more aggressively there. +(48:45) education, you know, is a big one. That I talked about a lot in public. But I think education is right for, for transformation. Actually, Clay Christianson has been doing really amazing work on that th, it's worth reading. He's very passionate about that. And then financial services I mean we think it's go time for financial services. +(49:01) We needed a breakthrough software technology to be able to go after financial services. Financial services right. The big problem for startups and financial services is, it's regulated to death. And so and it's gotten worse, right, through, like we, we call Dodd Frank the big bank protection act of 2012 like it's, it's the largest wall in the world for a startup to have to climb over to really compete from scratch. +(49:21) But Bit Coin and Crypto currencies gave us that technology change that we needed to after financial services. And so we are now looking at a very broad cross section of, new kinds of, whether it is new kinds of lending, new kinds of insurance new kinds of derivatives new kinds of small business financing, new kinds of fundraising, crowdfunding, crowdsourcing. +(49:44) We think now there's enough technology change happening and both consumers and business are desperate for alternatives. There's some very interesting financial services companies to get built. And in the long run the two we would like to work on, we haven't gotten to yet, but in the long run law and government are the other two really big ones but those are probably more in the out here, talk about that more at a future date. +(50:07) >> How do you see about, how do you see sort of the path of affordable internet for the rest of the world? And when will Edgarson Crow It start investing in future rocket technologies? >> I'm sorry, what was the last question? >> When will Edgarson Crow It start investing in future rocket technologies? I just didn't catch the. +(50:22) >> Frontier, frontier mode. >> Frontier. >> Oh, oh, it's developed like, developing worlds. >> Yeah. >> Or whatever the, yes, yeah, the term is. [LAUGH] It's a, it's a, it's hard to keep up with all the terms. So we don't so we consider, I would say the following. +(50:38) We consider that the basically the, the, the amazing opening of the developing world or whatever terms you wanna use. The, the billions of people around the world who have not had access to what we would consider to be modern education, modern information, modern communication, modern politics access to markets. +(50:56) You know, most of the world has not, has not had access to those things historically. Most of the world is in the process of getting access to those things. And so we think the biggest thing happening in our time is the just tremendous flowering of, of basically the en-, the entire developing world. And it's all over the planet. +(51:09) And it's happening at different paces but it, it's really all over the planet. There's extraordinary stories now in almost every country on the planet of just amazing things happening. And we think it's a twin story, it's you know, it's a political and economic development story and a, and a markets development story, but then it's also a technology story. +(51:25) And, we think it sort of is the rise of the developing world with first the PC and now the smartphone. It's not an accident that it's happening now. and, you know, it's, it's on the heels of satellite TV and fax machines and so forth. And, and, and the internet, but now it's smartphones. +(51:38) And so we think a world in which everybody on the planet has a smartphone with internet access is a completely different world. Not just because they can play floppy birds but because they can, you know, very fundamental things. They can get up to date market information on pricing. Which is very important if your a farmer and you've never had that information before. +(51:56) Or they can get up to date health information, or they can educated, kids can get educated in ways, you know. A lot of countries around the world like they don't have textbooks much less like modern education systems so there's just an enormous ability to upgrade modern education across the entire world. +(52:10) And then of course huge political change and a lot of that is a consequence of people number one being able to see what they're missing and then number two being able to organize. And we think the potential for these technologies as a way for, for political organ, organization, political protest to happen. +(52:27) Everybody I ever talked to who has been through one of these kind of political protests in the last five years talks about the centrality of this technology. It's only Western media commentators that say it doesn't matter. It's the people on the ground that are all over all the new stuff. So, it's really, really fundamental. +(52:40) We as an investment firm, are not, we're a single office firm, we're a boutique firm, so we're just investing out of Silicon Valley and we're mostly investing in U.S. companies and maybe a few, a few companies in markets like Europe. We're not really set up to invest on the ground in the developing world. +(52:55) But a lot of the companies that we're investing in are building products that we think are going to be transformative on the ground. And so I'll just give you one example. Lyft is a company that we're involved in. Lyft actually is, is a ridesharing company. That's a, it's a, Lyft and Uber are kind of roughly in the same market. +(53:10) The difference with Lyft is that anybody can become a driver. So it's not professional drivers, it's ordinary people being drivers. Lyft is actually based on the founder's experience on the ground in Zimbabwe, when he was working on a development project where he saw in, like, poor villages in rural areas all over the world. +(53:26) You know, you have you know, there might be a couple of cars and if somebody's going to take a drive into town, you know, anybody who wants to get in, in, you know, is gonna like chip in for gas, and everybody's gonna get to go on that ride. So ridesharing is kind of a thing that happens when not everybody has a car. +(53:40) So, Lyft is basically gonna take the concept of ridesharing global and gonna make it an information system, so you can have a much more optimal, right, spread of drivers and cars and rides. And so you can have transportation work much better, all throughout the world, including in the poorest parts of the world. +(53:54) Right, all through the smartphone. Which could be a huge boost to quality of life. Air BnB, same thing. Air BnB, you know, it's, you know, it comes across as a way for kids to travel around and stay in other people's houses. The founder of Air BnB, and by the way this is a classic case, the founder of Air BnB Brian Chesky went around and tried to raise venture capital everyone said you're crazy that's this, you know, nobody will ever stay in somebody else's house that's nuts. +(54:14) He was getting really depressed. He went home for Christmas and talked to his grandfather, and his grandfather says, oh yeah, yeah, yeah back in the 30s and 40s, that's what we used to do. Like, if you were going to go to another city, you would find out a friend of a friend, and you would say, hey can I, you know, use your spare bedroom, cuz like we couldn't afford hotels. +(54:29) And so the opportunity to take real estate globally, and make it much more accessible. And make it much more cost effective for people to be able to stay, you know, be able to stay whenever their on the road, whenever their traveling. To make it possible for everybody who owns a house or owns any kind of property to be able to open it up for be able to make extra money. +(54:47) You know we think many of these ideas can scale, basically all the way up and all the way down. We think that these can be very broad-based ideas and so we can kind of come at things, ridesharing, real estate sharing, education, financial services, all these things, and we can make them very, very broad. So that is what we are trying to do. +(55:05) >> Mark we're sadly out of time, so thank you so, so much. >> Thank you everybody. >> Thank you. [SOUND] [MUSIC] + +## It's time to build +Every Western institution was unprepared for the coronavirus pandemic, despite many prior warnings. This monumental failure of institutional effectiveness will reverberate for the rest of the decade, but it’s not too early to ask why, and what we need to do about it. + +Many of us would like to pin the cause on one political party or another, on one government or another. But the harsh reality is that it all failed—no Western country, or state, or city was prepared—and despite hard work and often extraordinary sacrifice by many people within these institutions. So the problem runs deeper than your favorite political opponent or your home nation. + +Part of the problem is clearly foresight, a failure of imagination. But the other part of the problem is what we didn’t *do* in advance, and what we’re failing to do now. And that is a failure of action, and specifically our widespread inability to *build*. + +Want more a16z? +Email Address * +Subscribe +MANAGE MY SUBSCRIPTIONS +By clicking the Subscribe button, you agree to the Privacy Policy +We see this today with the things we urgently need but don’t have. We don’t have enough coronavirus tests, or test materials—including, amazingly, cotton swabs and common reagents. We don’t have enough ventilators, negative pressure rooms, and ICU beds. And we don’t have enough surgical masks, eye shields, and medical gowns—as I write this, New York City has put out a desperate call for rain ponchos to be used as medical gowns. Rain ponchos! In 2020! In America! + +We also don’t have therapies or a vaccine—despite, again, years of advance warning about bat-borne coronaviruses. Our scientists will hopefully invent therapies and a vaccine, but then we may not have the manufacturing factories required to scale their production. And even then, we’ll see if we can deploy therapies or a vaccine fast enough to matter—it took scientists 5 years to get regulatory testing approval for the new Ebola vaccine after that scourge’s 2014 outbreak, at the cost of many lives. + +In the U.S., we don’t even have the ability to get federal bailout money to the people and businesses that need it. Tens of millions of laid off workers and their families, and many millions of small businesses, are in serious trouble *right now*, and we have no direct method to transfer them money without potentially disastrous delays. A government that collects money from all its citizens and businesses each year has never built a system to distribute money to us when it’s needed most. + +Why do we not have these things? Medical equipment and financial conduits involve no rocket science whatsoever. At least therapies and vaccines are hard! Making masks and transferring money are not hard. We could have these things but we chose not to—specifically we chose not to have the mechanisms, the factories, the systems to make these things. We chose not to *build*. + +You don’t just see this smug complacency, this satisfaction with the status quo and the unwillingness to build, in the pandemic, or in healthcare generally. You see it throughout Western life, and specifically throughout American life. + +You see it in housing and the physical footprint of our cities. We can’t build nearly enough housing in our cities with surging economic potential—which results in crazily skyrocketing housing prices in places like San Francisco, making it nearly impossible for regular people to move in and take the jobs of the future. We also can’t build the cities themselves anymore. When the producers of HBO’s “Westworld” wanted to portray the American city of the future, they didn’t film in Seattle or Los Angeles or Austin—they went to Singapore. We should have gleaming skyscrapers and spectacular living environments in all our best cities at levels way beyond what we have now; where are they? + +You see it in education. We have top-end universities, yes, but with the capacity to teach only a microscopic percentage of the 4 million new 18 year olds in the U.S. each year, or the 120 million new 18 year olds in the world each year. Why not educate every 18 year old? Isn’t that the most important thing we can possibly do? Why not build a far larger number of universities, or scale the ones we have way up? The last major innovation in K-12 education was Montessori, which traces back to the 1960s; we’ve been doing education research that’s never reached practical deployment for 50 years since; why not build a lot more great K-12 schools using everything we now know? We know one-to-one tutoring can reliably increase education outcomes by two standard deviations (the Bloom two-sigma effect); we have the internet; why haven’t we built systems to match every young learner with an older tutor to dramatically improve student success? + +You see it in manufacturing. Contrary to conventional wisdom, American manufacturing output is higher than ever, but why has so much manufacturing been offshored to places with cheaper manual labor? We know how to build highly automated factories. We know the enormous number of higher paying jobs we would create to design and build and operate those factories. We know—and we’re experiencing right now!—the strategic problem of relying on offshore manufacturing of key goods. Why aren’t we building Elon Musk’s “alien dreadnoughts”—giant, gleaming, state of the art factories producing every conceivable kind of product, at the highest possible quality and lowest possible cost—all throughout our country? + +You see it in transportation. Where are the supersonic aircraft? Where are the millions of delivery drones? Where are the high speed trains, the soaring monorails, the hyperloops, and yes, the flying cars? + +Is the problem money? That seems hard to believe when we have the money to wage endless wars in the Middle East and repeatedly bail out incumbent banks, airlines, and carmakers. The federal government just passed a $2 trillion coronavirus rescue package in two weeks! Is the problem capitalism? I’m with Nicholas Stern when he says that capitalism is how we take care of people we don’t know—all of these fields are highly lucrative already and should be prime stomping grounds for capitalist investment, good both for the investor and the customers who are served. Is the problem technical competence? Clearly not, or we wouldn’t have the homes and skyscrapers, schools and hospitals, cars and trains, computers and smartphones, that we already have. + +The problem is desire. We need to *want* these things. The problem is inertia. We need to want these things more than we want to prevent these things. The problem is regulatory capture. We need to want new companies to build these things, even if incumbents don’t like it, even if only to force the incumbents to build these things. And the problem is will. We need to build these things. + +And we need to separate the imperative to build these things from ideology and politics. Both sides need to contribute to building. + +The right starts out in a more natural, albeit compromised, place. The right is generally pro production, but is too often corrupted by forces that hold back market-based competition and the building of things. The right must fight hard against crony capitalism, regulatory capture, ossified oligopolies, risk-inducing offshoring, and investor-friendly buybacks in lieu of customer-friendly (and, over a longer period of time, even more investor-friendly) innovation. + +It’s time for full-throated, unapologetic, uncompromised political support from the right for aggressive investment in new products, in new industries, in new factories, in new science, in big leaps forward. + +The left starts out with a stronger bias toward the public sector in many of these areas. To which I say, prove the superior model! Demonstrate that the public sector can build better hospitals, better schools, better transportation, better cities, better housing. Stop trying to protect the old, the entrenched, the irrelevant; commit the public sector fully to the future. Milton Friedman once said the great public sector mistake is to judge policies and programs by their intentions rather than their results. Instead of taking that as an insult, take it as a challenge—build new things and show the results! + +Show that new models of public sector healthcare can be inexpensive and effective—how about starting with the VA? When the next coronavirus comes along, blow us away! Even private universities like Harvard are lavished with public funding; why can’t 100,000 or 1 million students a year attend Harvard? Why shouldn’t regulators and taxpayers demand that Harvard build? Solve the climate crisis by building—energy experts say that all carbon-based electrical power generation on the planet could be replaced by a few thousand new zero-emission nuclear reactors, so let’s build those. Maybe we can start with 10 new reactors? Then 100? Then the rest? + +In fact, I think building is how we reboot the American dream. The things we build in huge quantities, like computers and TVs, drop rapidly in price. The things we don’t, like housing, schools, and hospitals, skyrocket in price. What’s the American dream? The opportunity to have a home of your own, and a family you can provide for. We need to break the rapidly escalating price curves for housing, education, and healthcare, to make sure that every American can realize the dream, and the only way to do that is to build. + +Building isn’t easy, or we’d already be doing all this. We need to demand more of our political leaders, of our CEOs, our entrepreneurs, our investors. We need to demand more of our culture, of our society. And we need to demand more from one another. We’re all necessary, and we can all contribute, to building. + +Every step of the way, to everyone around us, we should be asking the question, what are you building? What are you building directly, or helping other people to build, or teaching other people to build, or taking care of people who are building? If the work you’re doing isn’t either leading to something being built or taking care of people directly, we’ve failed you, and we need to get you into a position, an occupation, a career where you can contribute to building. There are always outstanding people in even the most broken systems—we need to get all the talent we can on the biggest problems we have, and on building the answers to those problems. + +I expect this essay to be the target of criticism. Here’s a modest proposal to my critics. Instead of attacking my ideas of what to build, conceive your own! What do you think we should build? There’s an excellent chance I’ll agree with you. + +Our nation and our civilization were built on production, on building. Our forefathers and foremothers built roads and trains, farms and factories, then the computer, the microchip, the smartphone, and uncounted thousands of other things that we now take for granted, that are all around us, that define our lives and provide for our well-being. There is only one way to honor their legacy and to create the future we want for our own children and grandchildren, and that’s to build. + +## Why You Should Be Optimistic About the Future +https://www.youtube.com/watch?v=UnU5Dikdr2U + +Transcript: +(00:00) it feels like the Internet's impact on culture is just beginning in the sense of like a world in which culture is based on the Internet which is what I think is happening it's just at the very start right because it had to get universal before it could set the culture but that's actually happening [Music] good afternoon thank you mark for answering some questions I have a bunch of questions which I hope that we can talk about and these all have to do about the future where we're going I want to start with a question about the +(00:43) past you know a generation ago a lot of smart people didn't think the internet was going to work and therefore they were unprepared for its benefits what are we smart people and today not prepared for yeah so you may remember actually generate it wasn't even a lot of was even just that a lot of people thought that the internet was gonna work a lot of smart people didn't think that in fact the inventor I can't resist I can't resist on the story they actually the inventor of Ethernet which is a +(01:10) foundational trait of the Internet spent the 90s actually predicting the internet would crash would collapse and what he called it would be the giggle apps would take down the internet by like 1986 1987 he wrote a column at the time for a magazine called info world and he said that if he was wrong by a the internet hadn't collapsed by 1997 he would eat his column and to his enormous credit in 1988 he actually went on stage of the conference he actually took ripped his column out of the magazine he put it in +(01:38) a blender with water he blended it up and he drank it on stage so it's one of the one of the one of the more shining examples of intellectual honesty I've never seen as it turns out he was wrong it turns out the internet did work so I think the big thing I've been thinking about this a lot you know if it feels to a lot of people like things are getting strange there maybe I'm the only one who feels that way but if you read the news they're just track things up in the world as things feel things feel kind of +(02:03) weird and different over the last few years I actually think there's like there is an actual generational thing that's happening and you alluded to the generational component like it did take 25 years to get everybody online and like we're not quite there yet but we're getting very close mm-hmm like I think the most exciting thing happen in the world right now is Mukesh Ambani is the richest man in India has this program called geo where he is literally providing internet access to the 500 million lowest income Indians like from +(02:27) screw it's like but like literally it's like free for six months and then it's like $1 a month it's like the most amazing thing and it's like it's working incredibly well and so we are very very close to every every at least every adult in the planet being internet connected but it took 25 years to get there and so so that so for me it's like okay so so then what and one interpretation of that is okay we're done we did it the other interpretation of that is actually okay that's just the beginning point right +(02:50) right and it's like the beginning point of what right and I think it's beginning point of like okay like what if you actually interconnected everybody on the planet like what you know there's like the metaphor of the global the global mind the global mind with a global brain like what if you actually connected everybody together and let everybody find out what everybody else was thinking it's one of those things that people think sounds good then they encounter it face to face and they're +(03:08) like I don't know right that was like during my time that wired people were kind of concerned about the digital divide and I said the digital divide is going to cure itself the thing you should be worried about is what happens when everybody is online so you think they were not prepared for what will happen when everybody is online no well I think I think we're not we're not prepared and then I think it's going to be very exciting I mean I think we're already seeing that in many ways I think +(03:32) that there's and then I think we've kind of figured out collectively that it's gonna be different and so the initial impulse to sit-ins to say things are gonna get much worse I don't think that's right I think it's things are gonna get very different I think these will be yep much more positive and we'll talk a lot about that today hopefully but things are definitely going to be different um I think one lens that I've been trying to put on lately is kind of think about through a cultural lens sort +(03:54) of what happens to culture because culture you know Ben Ben just you know wrote this book about culture being kind of the foundation of behavior and I think that's really true certainly in companies but I think it's also true in countries and globally and it feels like the Internet's impact on culture is just beginning in the sense of like a world in which culture is based on the Internet which is what I think is happening is just at the very start right because it had to get universal before it could set the culture but +(04:17) that's actually happening now okay at the same time a generation ago well there was a few people who actually did think the internet was going to work but they were also like myself expecting VR and conversational AI to happen tomorrow so what are we expecting to happen now that it's not going to happen yep so I object to the question so this is one of those things in our business that you deal with a lot which is you know you find yourself you know these the Antron errs come in and they pitch an idea and you kind of feel like you +(04:50) should draw a judgment or whether the idea is going to work or not if it's something I'm really leery of doing anymore and the reason for that and think you know this from all if you're reading every successful technology that I'm aware of you know the things that are like all of a sudden like the next big thing like the iPhone a 2007 or just as an example they all have this like incredible 25 or 40 or 50 year backstory to them that you sometimes have to go back and excavate right because you +(05:14) haven't heard a lot of the back story because the previous efforts failed right but if you go back and look like there's often this often a multi-generational run-up and sawdust give you a few my favorite examples so iPhone you know hit big in 2007 I bi out four years went around say well iBM is there was a 20-year project IBM shipped the first smartphone in 1997 called the Simon I thought that was true it actually turns out it's not true I found the other day RadioShack had a smartphone in 1982 with +(05:40) their Tia they literally have a phone version of their trs-80 mini computer they sold about four of them but it was a thing right so that that had a 25-year fuse on it video conferencing you know video conferencing goes back at least to the mid 60's to the World's Fair yep the telegraph was invented in the 1870s and then sat on a shelf for a hundred years before the Japanese turned it into an industry and then my favorite is fiber optics nominal your you can't stretch you could say fiber optics were invented +(06:09) in 1840s Paris had a optical Telegraph network under under the city you could actually do you could actually under out you get flogged Rafi in the 1840s and in Paris it was literally they were shining flashes of light through glass tubes so there's this like this incredibly rich backstory to all these things and so as a consequence it's actually less a question of like what's the new idea it turns out the idea is probably already out there someone right okay and then it's less the question of like is it +(06:34) going to work it's more the question of like when is it going to right and I pushed it so far people in our office have heard this I push the push all the way to the point where I just think we should assume that whatever we're being pitched is going to work it's just a question of timing and of course timing turns out to be the hard part but right at least focuses the conversation right right so it's the same idea of kind of looking at the history of things one wonders who really made all the +(06:57) money when electricity came along it probably wasn't the people necessarily generating electricity who do you think is going to make the money when a eye comes along who is is it the yeah I providers is that the AI a service is that the out we're the writers who's gonna be making money in AI yeah so we think that there's two obvious business models and probably others but the true obvious one is to be a sort of a horizontal platform provider and for social provider you know for AI kind of analogous to the +(07:27) operating system or the database or the cloud you know the other opportunities kind of in a in the verticals and so the applications of AI and there's certainly a lot of those so that's the general answer I think that the deeper answer is there's an underlying question that I think is an even bigger question about AI that that reflects directly on this which is is AI a feature or an architecture is AI feature we see this we see this with pitches we get now which is this like we get the pitch and +(07:53) it's like here are the five things my product does right and both points one two three four five and then oh yeah number six is AI right and so you guys it's always number six right because it's the bullet that was added after they created the rest of the deck and so it's like okay if AI is a feature then that's actually correct which is like ever every basically everything is just gonna kind of have a I sprinkled on it a I features kind of every product that's possible we are more believers in the +(08:17) other scenario that AI as a platform and as an architecture if in the same sense that like the mainframe was architecture or the many computers in architecture the PC the Internet the cloud type in architectures we think is very good as a eyes the next one of those and if that's the case then it means that basically when there's an architecture shift in our business that means basically everything above the architecture gets rebuilt from scratch because the fundamental assumptions about what you're building change right and so +(08:43) you're no longer building a website or not like you're building a mobile app you don't like a building any of those things you're building instead an AI engine that it's just like in the ideal case it's just giving you the answer to whatever the question is and if that's the case then basically all applications will change along with that all infrastructure will change basically the entire industry will turn over again the same way that it did with the internet and the same way did with mobile and +(09:03) cloud and so if that's the case then it's just it's going to be like an absolutely explosive here to growth rhythm for this entire industry because it means then that all the incumbents supposed incumbents really aren't incumbent at all yeah they just the products just won't be relevant more I mean I just give you example there are lots and lots of sort of you know business applications there's your business apps as an example there's lots of business apps you know were you basically you're type data into +(09:24) a form and then they'd stores the data and later on your run reports against the data charts and that's been the model of business software for you know 50 years and in different versions you know what if that's just not needed anymore like what if what if in the future what you'll do is you'll just give the AI in your business access to all you know email all phone calls all everything all business records all financials in the company and just let AI give you you know give you the answer to whatever the +(09:44) motion was and you just don't go through any of the other steps Google's a good example that's like they're pushing hard on this like the consumer version of this right as search right so search has been that you know it's been the 10 blue links you know for 25 years now you know what Google's they talk about this publicly what they're what they're pushing towards this is just like no it'd just be that answer right which is what they're trying to do with their with their voice you eyes and so that +(10:06) that concept might really generalize out right and then everything gets rebuilt right so so one of the new interfaces to AI that people are talking about is voice as the new interface what are we likely to get wrong about voice yeah so I think the thing that if we're gonna get right something wrong about voice I think it's gonna be that it would be a one-to-one replacement for existing user interaction models so that it would be like a replacement for keyboard there to be replacement for the mouse or for +(10:33) touch probably not because it's a different modality right it's it's it's not you know weird like we know exactly what they keep you know after all this time we know what to give words for we know it touches for and for voice to displace those seems like a stretch on the other hand to the previous to the previous question there has been this turning point reached it feels like in a I applied to language and from there to voice write to text into speech which is it filled it feels to us in the +(11:00) technology like the natural language processing methods that people been working on for again for 50 years computer scientists been working on getting computers to understand basically speech and and what we're seeing now is in the technology is that that now has started to work in the same way that machine vision started to work about seven years ago and so if that's the case then all of a sudden the conversational UIs are about to get much better again you and then you couple that with okay what are you actually +(11:24) trying to achieve when you talk to a click when you talk to a computer are you actually trying to like you know are you trying to write a document or you're trying to read an email or you're trying to like do all these other things you do today or are you fundamentally gonna be doing something different because the machines gonna be so much smarter and I think that's a very interesting open question when I think about the AR mirror world I find how very hard to imagine without it having a voice component where we can understand +(11:46) what you're saying besides what you're looking at that is an essential part of the AR world yeah I think actually I'd go so far as to say it may be the case that voice actually is the key to the air like voice may be the thing voice may actually be the foundation of the whole thing you know for this is kind of a cliche at this point but like you know the Apple earpods I think we're a fundamental breakthrough like we is there's even one of these funny things where it's like okay wireless headphones okay cool like +(12:09) where else headphones where there's you know there's not even a wire connecting the two things cool okay it seems like more of the same but you know that if you want the experience you can have now is like you can wear one of these things basically all day and you can you can talk to it all day and you know they're getting you know the new versions are getting better you know that and you know Siri and Google now Cortana all these things are getting really good really fast and so it may be that we +(12:29) have just this constant ongoing running dialogue this is kind of you know basically the machine not talking to our ear and then you know the visual the visual overlay of AR will obviously be important and valuable but it might be it might visual overlay might be supportive on top of the voice experience and we could very quickly have universal language translation if I speaking of years and I think people underestimate the change that would bring about in the world you'd have millions of people who were highly +(12:56) skilled and everything I stepped the skill of English now being able to participate in a global economy we were talking about unexpected unexpected things biology which is a million times as complicated as digital we're now talking about a biotech revolution are we misunderstanding what biotechnology actually is yeah so that's the big bet that we've made with our with our with our bio effort that we started a few years back we think biological sciences at a turning point at the scientific level and we think it's at a turning +(13:29) point from basically being a process of discovery of how biology works to being able to engineer biology right up to it including literally being able to program biology right being able to actually built it basically be able to use electrical engineering a computer science and these mechanical engineering and these kind of fields for engineering things and be able to apply those kinds of skill to biology if we're right about that then the whole concept of kind of how bio and biotech work might be on the +(13:56) verge of really changing the most obvious Apple application that would be in pharmaceuticals you know there's this concept of drug discovery really right and sorry is the word discovery and discovery it's always like you know discovery sounds great it's like it's optimistic it's like ooh this is something you like discovering things that's fantastic the problem is right discovering like just leave a callback is like they literally have to like run all these experiments and try to discover the drug that works like trying +(14:17) to kind of you know reverse engineer back from nature and and the problem is like sometimes they discover it sometimes they don't right and so even so the example we always give us we talk about with computers right we've been on this kind of a 50-year track of what's called Moore's Law right working where chips make any faster and cheaper every year for a long time in biology and drug discovery there's what they call a Rooms law which is more spelled backwards and it's the cost of discovering a new drug +(14:44) and it's exactly the wrong direction that's right it's something to the right you know billions of dollars now and so if you could actually engineer biology right then all of a sudden you can start to apply this like just you know these decades of skills that we built up and how to engineer things and be able to do things like engineer new pharmaceuticals from scratch and it all runs on basically ultimately Moore's Law Moore's law has been foundational to this here it's almost hard to imagine anything we +(15:08) have in the modern world today without Moore's Law do you think Moore's law has a another 30 years run is it limited as finite we'll go on forever will define Moore's law and a broadest sense of computers getting cheaper by half every couple of years so what's your take on Moore's law yes the traditional definition is is computer in the form of the chip like in a specifically a chip right so Moore's law has always been expressed it's kind of unit 1 of chip and that could be right that could be a +(15:38) CPU or it could be a graphics card or it could be a graphics chip or a memory chip and then and then specifically what you were doing was you were able to put more transistors on that chip right for the same cost and actually for a long time as you did that you're actually able to reduce the power requirement for per transistor which was this kind of added at an added benefit and so chips kind of kind of got you know simultaneously they got faster they got cheaper and they got more power efficient and that was kind of a +(16:00) cornucopia effect agenda as you said most of most of what you see today in the computer industry so the bad news is that that in in that form seems to be coming to something of in hand which is we have we're to get out we've hit basically we being the semiconductor industry broadly the tech industry have kind of hit the limits of fundamental physics like we're now down at the sort of deep atomic level mm-hmm and it's becoming much harder to make there's still progress because much harder to make progress at the per chip +(16:25) level the good news is that the industry starting 10 or 15 years ago the computer industry broadly refocused off of what you do with a chip to what you do with a large number of chips right so kind of the old model of a chip was you make the chip more powerful because you're trying to scale up what you can doing the chip the new model is you use thousands of chips in parallel and you have this kind of approach to scaling out and of course the full form of that is what's now known as the cloud and so we now have a +(16:50) few no 15-year had a steam going to basically be able to get good at using lots of chips to do things and that's why you see the continued ability to write to accelerate you know a lot you know many many things that you deal with they're getting you know still much faster as if they're still on the Moore's Law the experiences you're having or getting faster so we think number one like the rise of scale on architectures is a really big deal like you know in modern clouds as a developer you don't really care about +(17:12) what the power of any particular chip is you just like light up some more of them and they don't cost much so there's that the other thing as chips are now specializing in particularly you've got the rise of these new dedicated chips for things like neural networks where there's another level another level of opportunity optimize and then the other kicker is the programmers software people like me get to step up in the old days when computers were expensive programmers are really good at optimizing every single step of a +(17:39) software program programmers got out of that habit probably starting thirty years ago where it didn't matter as much anymore because Moore's law was working so well and so software today is just like massively inefficient there's actually I forget the name there something called birth flaw which is that it was written at the time if it still holds but it was somebody did benchmarks of you take Microsoft Office 2008 PC from 2000 and you take Microsoft Office 2007 on a PC from 2007 and every function you could do you could not +(18:07) doing twice the time right so like literally like the old the old adage in in in tech in the 90s was when Andy Grove was wanting Intel and Bill Gates or any Microsoft it was Andy Gibb Moore's law and then bill taketh away in the form of in the form of software bleh and so and worth lot literally as a mathematical proof of that and so like it's become prime time again for software programmers to get really good at optimization which is like what's happening in AI world and also in the cryptocurrency world and so with those +(18:37) different approaches it feels like we've got you know it feels like decades of advances ahead that aren't purely dependent on classic Moore's Law and because if we take the long term like the thinking of a hundred year span to have prosperity like we've seen we'd kind of require that computer power sort of get cheaper every year because of a dent that's hard to imagine a world like this so is your confidence that we could kind of keep this going based on just sort of human ingenuity or +(19:05) is your do you think that there's some basic principles of science that suggests that we're just at the beginning of what we can discover well so weird Gordon Moore who invented Moore's Law is co-founder of Intel you know we said Mars lives was interpret as a prophecy and he said it was not a prophecy it was this it was a goal right right and it was a goal basically what you could do if you focused intensely if you focus entire industry intensely on a set of engineering optimizations right over a long period of time and so he +(19:32) used to say it's just like there's nothing a little inevitable about it it's a consequence of thousands and then tens of thousands and then millions of engineers like working to actually deliver on these kind of sets and my arbitrary goals and so I think the I think the answer that is we have many many areas of improvement as I said the problem is we don't have the one that we have which is this transistor doubling kind of effect but we've got many men I mean looks I mean there's far more +(19:54) engineers working on all this stuff today that were working on it in 1965 when he invented Moore's law or in nineteen you know 95 when everybody bought a PC like we've got we have a lot of a lot of mind power going into this we've got a lot of different technological options we've got a lot of you know incredibly impressive work happening all over the world the other thing is you can't you know you never like you know some of those things like the transistor was not obvious and then they invented that and then this +(20:21) integrated microchip was like not obvious and then they event than that and so you don't quite know you know there are lots of technical proposals for how to get to the next level Moore's law you know there's all kinds of theories running optical computing and then in the long run biological computer computer watching computing exactly and so over the course of the next like 20 years like look for this way this is one of the world's largest prizes right if you write if you're the engineer who figures out how to reseller +(20:43) at Moore's Law or how to shift computing onto a new substrate like biology that is the thing to do and so that that's the prize yeah and that historically it's been pretty motivating me so taking this kind of theme of marching forward progressively we have 4G we're talking about 5g so far 5tu seems to be faster 4G with a lot of hype added to it there's a technical specification for 5g which is really awesome you know 100 gigabytes to millisecond latency almost impossible do you are you counting on +(21:15) that for the next decade that we're gonna have actual what they promise with 5g yes I think there's pretty good as we will and the reason is because 5g has become a national geo political battle like it's actually a very interesting twist it's become actually a primary like you know if the cold war between the US and USSR was like defined by the space race like at least the sort of nascent cold war with China is actually a lot of it is around 5g interestingly enough I mean it could have been around +(21:42) a lot of things but it happens that it's around 5g and so you now have nation-states that very very badly need to win and two big nation-states in particular and so I think there's going to be a lot of you know so I start with the payoff from the space race it's like all the products they got you know spun off from them satellites and GPS and everything else and the other thing on 5g you know sometimes so if I vo lead to applications say I've even thought of yet and I think that's kind of true but +(22:07) I look at it a little bit differently which is a little bit like the Moore's Law conversation we're having which is look a little bit as a math kind of question which is there's there's sort of three classic rules for how networks scale and how networks scaling turns into value or usefulness and there's sort of historically there's what it's called Sarnoff's law which was based on broadcast TV which is the value of a network is equivalent to the number of nodes right so it's it's it scales with +(22:30) n right so a TV network with 10 million viewers is twice as valuable as a TV network of 5 million viewers so that's kind of the obvious one then there was Metcalfe's law which is basically the value to network is on number of kin between two points and that's like how email works right it's emails a person a person and that's correlates to end square it's the value of the network Rises exponentially with N squared and then there's this thing called reads law which is called the group forming law which is the value of +(22:56) network is proportional the number of groups and subgroups that conform inside the network which turns out to be due to the ambient yeah and if you want to have fun in your plane flights know them it's like oh just go on Excel and like chart you know and N squared and 2 to the enth right and 2 to the ants just goes like straight vertical like you can't even put them on the same chart and and 2 to the N tha's like what what's not happening with like social networks right so like Facebook groups and all +(23:17) these all these all these other things people whatsapp groups noise every things people do with social networks and games and all these other things and so so those are like the three ways in which network growth pays off and like all three are working now based on you know broadband wire a broadband they're all working you know you see it happening very much with mobile you know the introduction of 5g the way I think about it is it's going to turbocharge those three networks in particular that last one or though you know those last +(23:39) two and so it's gonna add a lot more and there's just be a lot more devices on the network there gonna be a lot more things that those devices can do they're gonna be a lot more point-to-point connections that make sense to have there's going to be a lot more groups that form a lot more economic activity that happens something that was again we're expecting to happen but dint was in the world of what's sometimes called the sharing economy there was a after Airbnb and uber there was a stampede of +(24:04) companies that were going to be uber for X and then X was everything in the world very few of them have succeeded again there was an expectation we'd see more of them but we haven't so is that whole idea kind of at a dead end is that is it just we're in a very slow disruption this is gonna take a while like the generational requirements we were talking about technology earlier or is something else so what's what do you think happened there and what are we looking at once again I object to the question okay +(24:36) throw the gavel so I I look at a little bit differently which is the this is something we try it hard to do it at our place it is very tempting and we do have this conversation all the time at our place so it's like okay what about the trend what about the theme right what about the variations on the same kind of as you said and this is something happen where something wins back you always get this kind of you know we've described it as kind of a Hollywood model of you know it's like you know what you know what's your new +(25:00) movie about a pretty woman Meets The Rock right or you know whatever and so in the valley it's you know super fracture most recently super human 4x which I'm very excited about it's one of the one of the big new trends after another one of our companies so but I don't think it's really that that's not how the great ideas arrived they don't look like that they look like very specific they look at very specific theories not general theories so they tend to be very specific to the details of the market +(25:28) involved one of the things that I think we've learned about ride-sharing why ride-sharing works so well I'm it worked well for many reasons one of the reasons that were so as an idea is because as long as the driver is good as long as I rated as a certain level it doesn't really matter who the driver is so like one of the classic examples was uber for cleaning your house for your apartment and it just it turns out you just don't want a different person over every week to clean your house like it's a problem +(25:49) and so so there's a lot of these kinds of I would say simple you know sort of the simple applications the idea they don't necessarily work now by trying all those ideas you kind of map the idea space and you start to get a better sense of like what the overall structure is and and I think what's happening now is you're starting to see another set of companies coming out the other end that have kind of fully internalized that lesson and had figured out new models that work and so my favorite example +(26:10) might one of my companies called honor so honors you might think loosely might think of it as kind of a burr for senior care for for in-home care for seniors it's it's a loose my wife she turns out it's a very loose model for a couple reasons one is it's really deeply not a fungible service like if you have an aging parent you actually very much don't want somebody different to show up all the time you want you want the same person and so in that case for example honor actually is actually has full-time +(26:37) employment relationship salaried employment relationships with with with the workers right which of course is very different than the uber and lyft model it actually turns out the matching problem is much more complicated right because you've when you're mashing human beings in somebody's home there's like 20 variables that you need to match on so that everybody's comfortable with the experience as an example you know in some cases you literally need people with the physical strength to be able to +(26:57) live people when you're caring for them you do want to be able to do this kind of multi-dimensional mapping and so that and that models really working and so I think we're gonna see a whole set of these like I think there's a big kind of Vista of exploration that's gonna happen from okay and I would I would suspect there'll be dozens if not hundreds and new models that people figure out so speaking of new bottles do you ever think about new models for the VC industry itself and how you would apply +(27:18) the principles of innovation and disruption to what you do in general so as you look out 30 years what kinds of innovations would you expect in the basic business that you're in yeah so there's something very timeless about venture which is there's actually a new book out called literally called VC actually tells the story that has been kind of hard to get out for a long time in a really clear way which is then the modern venture model is actually one of the one of the historical precedents for it was actually how whaling +(27:46) expeditions got financed in the 1600s so coming up on five hundred years ago so whaling of wedges it was literally like okay you're gonna have like a ship with a captain and a crew that's gonna go out and try to like bring back a whale right and it's like a problem number one it's like only two-thirds of the ships are gonna come back right so like high failure rate to is like okay like whoo you know what the ship has really matters who the captain is really matters how do you know who could captain is and then you know what's +(28:12) what's a good crew and like you know the crew are they gonna be willing to follow the captain and then there's all these like strategy questions like do you want the captain who knows where all the whales have been caught recently so they go there or do you want the captain that says no that's here it's gonna be over fish you want to go someplace else and so literally all the whaling voyages like in the in the in the in the colonies 500 years ago got funny asked with basically angel syndicates +(28:35) basically a venture capital effectively and then literally that the term carry which is sort of how feces get paid the so called carried interest which is like the 20 or 25 percent that you make that you have the profits that you share the term carry actually was the percentage of the whale that the ship carried it was literally physical carry it was literally that part of the whale like that's where that term came from and so there's a timelessness to the art of trying to figure out how to finance these kind of expeditions and to be +(29:02) unknown that is you know this likely to endure the big question for me is how will the shape of the companies or they'll say the ventures themselves change right and so you know today there's like a well known understood template for kind of prototypical Silicon Valley venture investment and it's like a company in a certain place it's a c-corporation domiciled in the US it's financed a certain way and and certain types of employees a certain relationship with its employees and so forth you know 30 years from now you know or +(29:30) what you know are these are we gonna be financing companies here or you know anywhere or you know in two places 50 places 500 places or the company's still gonna have physical place or are they gonna be fully virtual are they gonna be companies or are they all gonna be blockchains right are they gonna be right are they gonna have actual employment relationships or are they gonna have you know basically developers incentive through cryptocurrency that's a real model and so I think the big question is like we don't even know what the shape +(29:55) of companies is gonna look like their ventures is gonna look like in 30 years so if I could figure that out then I could answer with venture it looks like without that I think it's hard to say okay so we were tempted to do a little bit of long-term thinking and learn long-term thinking is sort of rare and often ignored whereas civilizations demanded as being necessary so do you have any suggestions about how long-term thinking could be applied in Silicon Valley and and whether you have even any suggestions to +(30:27) the people in this room about how they could use long-term thinking yeah so the thing I've always found about long time having long term thing is of course Central especially one of the things about the valley that I find outsiders miss the most which is it you feel it feels like it's all moving so fast and yet like any of the important companies and the important products take like a decade or more to build and so it's like everything important basically takes a long time and so it actually a lot of it +(30:51) actually feels quite slow and so I have long term orientation is absolutely necessary and I think we probably all agree there's not enough of it in the world the thing about long term thinking I found is like it's really easy if you know the thing is gonna work right like a boy that's it's completely straightforward let's go on a 10-year journey to a place where we know it's gonna be great right the problem is this long-term thinking crossed with uncertainty right and quite possibly fatality like the thing they just simply +(31:17) not work for any of a thousand reasons and so that's the issue and so I think the issue is less around long-term thinking I think the issue is more about how to deal with risk and how to deal with uncertainty and how to make really big consequential decisions in the face of like you know literally an unknowable you know future landscape and for their I mean this is kind of kind of secret weapon adventure it's like it's like ventures the worst of all asset classes in a lot of ways and that it's like it's a liquid and it's like +(31:42) incredibly volatile and it's like hitting you know it's hit or miss in this kind of crazy way the one thing that venture really has going for it is an asset class is we have the concept of the portfolio kind of wired into the model in which you just kind of assume in top adventure you just kind of assume fundamentally it's half the company's gonna work after the mark right and then the classic right the class the cliche is like the ones that work then you don't have to pay you have to work enough so that they pay for +(32:04) the ones that don't to make to make the whole enterprise work and so if you can adapt yourself from the mentality of will this thing work right - will this will this portfolio of things basically pay off right well enough things work that they'll actually pay for the portfolio then at that point you can start to make risk somewhat tractable thing to contemplate it's still hard to divorce yourself emotionally from it right because it's just like it's still like absolutely no fun you know it's just terrible and when +(32:29) any of the individual things don't work but at least you have a conception of frame rate for able to it'll make ten long-run bets and being able to get to the other side now the response don't forget - that is oh that's great if you're a VC the problem is your portfolio you know you're you're a founder or a CEO like you don't you don't get that right you have to you have the much harder version of the problem which is you're on a one-way journey like you're the captain of the +(32:49) whaling ship yeah there's all those other captains over there but like you know they're on their own you're on your own even there though you know the best run companies tend to run experiments they tend to run right multiple experiments against against against their goals and they and they certainly want this experiment sequentially as they kind of you know try to figure out what works in a lot of cases they want experiments in parallel is they're trying to test different things and so I also think that's kind of mentality is +(33:12) sort of portfolio risk also applies to how you run a company which is you want to basically you you want to have a great deal of conviction about retrying ahead but you want to have a lot of flexibility inherent and how you're gonna get there right and what the tactics are and then you wanna be able to run a lot of experiments against that and you can kind of diversify your risk of any one theory by doing that and that's what governments are in some senses they have a portfolio of different kind of prospects about the +(33:35) future bets and in some senses so they think that's an optimistic view yeah exactly like what governments do yeah so I mean that's what they're and they're adverse to risk unfortunately so well the problem is the problem governments from government is like the end of one right so there's only one government per weight right we're really good run you know I mean next federalism which has been a huge huge advantage I think for the u.s. +(33:58) but like well you know the u.s. national government only gets to run one scenario right running experiments in the population is not necessarily well-received because they can't tolerate failure yeah right failure has real consequences so there's there's currently not only introspection about government but also about capitalism and capitalism so far has depended on growth and growth is something that VC's pay attention to but we're now wondering if what's the minimum amount of growth that you might +(34:28) need to have prosperity can you have prosperity with low growth can you have prosperity with fixed growth do you have any insights about about that at the civilizational scale yes I think in actually I don't even say that the issue is even more intense these days because there's now very prominent people in public life arguing that growth is bad right and in fact it's it that it that it in fact is ruinous and destructive and that the right goal might actually be they have no growth or to actually go +(34:53) into negative growth then especially in a very common view in the environmental movement so I'm a very strong proponent a very strong believer that growth is absolutely necessary and I'll come back to environmental thing in a second because it's a very interesting case of this I think growth is absolutely necessary and I think the reason growth is absolutely necessary is because you can fundamentally have two different mindset views of how the world works right one is positive some which is you know +(35:15) rising tide lifts all boats we can all do better together and the other is is is a zero-sum right where for me to win somebody else must lose and vice-versa and reason I think economic growth is so important at cores because if there is fast economic growth then we have positive some politics and we start to have all these discussions about all these things that we can do as a society and if we have zero some grow if we have if we have a flat growth or no growth or negative growth all of a sudden the politics become sharply zero-sum and in +(35:46) the most in the most the most you just kind of see this if you kind of track you know kind of the political climate you just basically it's the wake of every recession right it's just in the wake of every economic recession the politics just go like seriously negative on in terms of thinking about the world's is zero-sum and and then when you get a zero-sum outlook Impala six that's when you get like anti-immigration that's when you get anti trade that's when you get anti tech if the world's not growing then all +(36:08) that's left to do is to fight over what we what we already have and so my view is like you need to have economic growth you need to have economic growth for all of the reasons that I would say right wingers like economic growth which is you want to have higher levels material prosperity more opportunity more job creation all those things you want to have economic growth for the purpose of having like sane politics like a productive political conversation and then I think the kicker is you also want economic growth actually for many of the +(36:33) things that left-wing people want one of the best books this year new books this year is a guy Andrew McAfee I was reading a book called I think more from less it's actually a story of a really remarkable thing that a lot of people are missing about what's happening with the environment which is globally carbon emissions are rising and resource utilization is rising in the u.s. +(36:55) carbon emissions and resource utilization are actually falling and so in the u.s. we have figured out to grow our economy while reducing our use of natural resources which is a completely unexpected twist right to the plot of what kind of if you lose environmentalists in six years of seventies like nobody predicted that and it turns out he talks about this in the book but it turns out basically what happens is economies when economies advanced to a certain point they get really really good at doing more with less right they get really really good +(37:19) efficiency and they get really good at energy efficiency they get really isn't about you use environmental resources they really go to recycling in lots of different ways and then they get really good at what's called dematerialization which is what is happening with digital technology right which is basically taking things that used to require atoms and turning them into bits weight which inherently consumes consumes less resources and so what you actually want like my view unlike the environmental +(37:42) issues is like you've got a global problem which is you have too many people in too many countries stuck in kind of mid amid the Industrial Revolution they've got to grow to get to the point where they're in a fully digital economy like we are precisely so that they can start to have declining resource utilization right right I mean the classic example energy like you know the big problem of the energy emissions global a huge problem of emissions and with health from emissions is literally people burning wood like in their houses +(38:08) right to be able to eat and cook and what you want to do is you want to go to like hyper efficient solar or ideally nuclear right you want to go to these like super advanced forms of technology so actually it so you want that and by the way if you want like a big social safety net you know and all the social programs you want to pay for that stuff you also want economic growth because that generates taxes of pace of that stuff and so like growth is the single kind of biggest form of magic that we have right to be able to like actually +(38:31) make progress and hold the whole thing together and you know to your point about the developing countries I think the idea of leapfrog and technology is a myth it doesn't really work you actually have to if you want to have a high-tech infrastructure you actually need the intermediate roads clean water you can't skip over that and so they all need to be built out in order to have that prosperity at the end so you know the simplest you know seems like you don't worry about much I don't worry about +(38:57) much but one thing I do worry about this cyber conflict cyber war partly because I think we have no consensus about what's allowable does this worry you at all so I think there there's a lot of unknown as to it I think people are trying to figure this out but it's it's a complication to grapple with I will make an optimistic argument which is going to sound a little strange if you kind of project forward what's happening with with generally cyber with information you know operations of different kinds but also with drones +(39:28) you know UAVs and then also with you know unmanned you know unmanned fighter jets right um and you know ships increasingly being built it'll be unmad submarines at some point if you projected stuff forward you start to get this very interesting potential world in which basically the way I think about it it's like all human conflict between peoples are between nation-states up until now has been basically throwing people at each other right throwing soldiers at each other and like letting them make the decision +(39:56) of who to shoot and like hoping they don't get shot like with very serious repercussions of all those individual human decisions you do have the prospect of basically a new world of both offense and defense it's like completely motorised completely mechanized completely software driven and technology driven and a lot of people it's just immediately like oh my god that's horrible you know Terminator like you know Skynet like you know this is just the worst thing ever there's a novel called kill +(40:17) decision if you wanted to snow Pinsky okay there's a novel called kill decision by daniel suarez dinosaurus then extrapolates the the drones forward and a little it'll keep you up late at night but the optimistic view would be like boy isn't it good that there aren't beings involved isn't it good like if the machines are shooting at each other like isn't that good isn't that better than if they're shooting at us by the way and by the way yeah I would go so far as to say like I don't know that I'm +(40:40) in favor of like the machines making like kill decisions like decisions on me to shoot but like the one thing I know it's humans do that very badly like very very very badly I'm the opposite of pearl war I don't want to see any of this stuff actually play out but if it has to play out there maybe having it be software machines it's gonna be actually better outcome right I mean this kind of weird that we don't allow we don't want machines to kill humans we want other humans to kill you but we want 18 year +(41:02) olds we want to take 18 year olds out of their homes right we want to put a gun in their hand and send them someplace and tell me decide who to shoot like it that that is gonna go down to history's haven't been a good idea okay it just strikes me as like unlikely so we have only time for one last question which is I'm usually I claim to be the most optimistic person in the room but with you sitting across to me I don't think that may be true what is your optimism based on so my optimism okay so get +(41:30) cosmic for a second why not I guess we're here it's the last question last question so the science fiction author science fiction science fiction authors always talking about was good they called the singularity this constant singularity answer it's a singularities basic what happens when the machines get so smart than all of a sudden everything goes into exponential mode and all of a sudden you know the entire world changes so I am I reading history is actually we actually were in the singularity already +(41:53) and that it actually started 300 years ago mm-hmm and if you look at basically if you look at basically any chart of human welfare over time and you can look at no child mortality is an obvious one but like there's you know there's many many many others and you just look at progress on that metric so your telomere tality as an example and it's just basically flat flat flat flat flat flat flat for only fifty thousand years right is everything and you know if this is the family offices at Thomas Hobbes you +(42:17) know life is you know nasty brutish and short right it was just like the thing like everything was terrible everywhere all the time forever the end until 300 years ago when all of a sudden there's this me and the curve and then all the indicators of human welfare not uniformly across the planet but in societies that we're making progress the societies weren't making progress first all of a sudden all those indicators of human welfare went up into the right right I don't know of course bonded by +(42:42) the way to economic growth but it was also right it was the Enlightenment it was the rise of democracy it was the rise of markets was the rise of rationality of the scientific method by the way human rights free speech free thought right and they all kind of catalyzed right around around around 300 years ago and and they've been making their way into the world you know in sort of increasing concentric circles kind of ever since and so we have you know I would argue like we have the answer it's like we actually don't need +(43:07) new we don't need new discoveries to have the future be much better we actually know how to do it is to apply basically those systems and and and basically contra the sort of constant temptation from all kinds of people to try to you know compromise on these things or subvert these things you know basically double down on these systems that we know work right so double down economic growth double down on human rights double down on markets on capitalism double down on the scientific method fix science like we got as far as +(43:35) we did with science actually being pretty seriously screwed up right now with the replication crisis like so we should fix that and then science will all of a sudden start to work much better technology right used to yeah use of technological tools so we should we literally have the systems like we know how to do this we know how to make the planet much better in every respect and so what we just need to do is is keep doing that and then what I try to do when I read the news is notwithstanding everything's going on is basically try +(44:00) to look through whatever's happen at the moment try to look underneath and kind of say okay are those fundamental systems actually still working like is the world getting more democratic or less right this is free speech spreading or receiving right or markets expanding or falling right are more more people able to participate in a modern market economy or not and you know those indicators generally are all or all still up into the right mm-hmm so let's go out and make the world better yeah thank you +(44:24) thanks everybody [Applause] diff --git "a/investor/\360\237\227\243\357\270\217belief/PT.md" "b/investor/\360\237\227\243\357\270\217belief/PT.md" new file mode 100644 index 0000000..ea60bb8 --- /dev/null +++ "b/investor/\360\237\227\243\357\270\217belief/PT.md" @@ -0,0 +1,104 @@ +## Successful Businesses are Based on Secrets | WIRED + +Video URL: https://www.youtube.com/watch?v=yODORwGmHqo + +so I'm David Ryan editor of Wy and it's an amazing honor to introduce Peter who has a fabulous new bestseller out um 0o to one and Peter is a busy man he has a number of lives that all run at the same time he is um managing partner of the founders fund president of the of clarium capital co-founder of PayPal of palente here um makes good Investments like $500,000 for 10.2% of a new company called Facebook in 2004 and when he tweets he only ever tweets once he gets 60,000 followers pretty much straight away now we have big expectations for this evening because everywhere Peter has spoken in the last week or so about his new book um he's generated headlines and the headlines include you know Twitter's horribly mismanaged there's a lot of pot smoking there Apple must innovate more technology stalled in 1970 The Wall Street Journal competition is for losers and I think my favorite one from Fortune Peter teal disagrees with you so I'm expecting Peter to tear apart some of my questions but first please could we welcome him and give him the stage to talk about 0 to1 thank you very much well um after after that introduction I'm I'm actually quite nervous about saying anything at all here since uh feel like I'm on being recorded and uh it may come back to haunt me in all kinds of strange ways in in in writing a book about entrepreneurship um one of the one of the uh this and teaching it came out of a class that I taught at Stanford in 2012 and one of the challenges in teaching entrepreneurship or writing about it is I think people tend to sort of write books U business books that make one of two kinds of mistakes there's one where it's sort of uh you tell all these War Stories and this be like this is what I did at PayPal in 99 this how we combined email and money and this is you know how this all worked and that's not terribly relevant to people uh who are interested in starting new businesses or or learning from that and then I think the other end of the spectrum you have uh these uh business books which I think are sort of pseudo scientific and it's like this formula and you do follow these five steps and you will have a successful company and I I think that approach also does not work for a similar reason which is that I think every moment in the history of business and the history of Technology happens only once the next Bill Gates will not be starting an operating system company the next Larry Page won't start a search engine the next Mark Zuckerberg would not be starting a social network if you are copying these people you're not learning from them uh and um um and so the the the theme of my book is that uh what's actually important is the things that make these companies singular and and different and um and you know this chapter entitled one you know this opening line from Anna coren and um it says all happy families are alike all unhappy families are unhappy in their own special way uh and I think the opposite is true of business all um all happy companies are different all um all unhappy companies are alike and that they've failed to escape the uh sameness that is competition and um and this was this the title of the chap of one of the chapters is all happy companies are different it got when the Wall Street Journal ex excerted it and uh and published it uh they uh they retitled that to competition is for losers which is a somewhat punchier way of putting that idea and um and you know I try to get at this uniqueness question through these various contrarian questions I think are are good to ask one the the business one I always is like what great company is nobody building the the more intellectual uh version of this question that I think makes for always a terrific interview question is tell me something that's true that almost nobody agrees with you on and this is a shockingly hard question for at least two different reasons uh first reason is that we've been taught that truth is conventional that truth is just simply what people agree on uh and so it automatically sounds like you have to be really brilliant and it's really hard to discover some new truth that uh is hit herto unsuspected um but it's it's also very difficult because of just the social context in which uh the these questions get asked so if uh if you're answering it and the interviewer says yes I believe that all along that's uh uh that then then that's by definition a bad answer and a good answer is one the person asking it does not agree with or does not want to hear and um and you know and it requires certain amount of courage and we live in a world which I think courage is an even shorter Supply than than genius so what I want to uh what I want to try to do in my brief comments tonight is I'm just going to give uh a few few answers my my book is just a whole series of answers to that question of things I believe to be true that uh most people don't agree with me on um but I'll I'll give three um three uh quick answers to it tonight in my in my comments and we'll uh go to the questions um so first one um uh and this is this sort of flows out of this idea that you want um all happy companies to be different and unique um most people think that capitalism and competition are synonyms I believe they are antonyms um you know a capitalist is someone who accumulates capital a world of competition is a world where all the profits are competed away and so if you're an entrepreneur or a Founder you always want to be building a monopoly you do not want to be uh doing something where you're in Cutthroat for ferocious competition a restaurant is a terrible business to go into it's super competitive and extremely non- capitalistic people never make money opening restaurants uh and uh I sort of give Google as the example of a company at the other end of the spectrum where you have enormous profits um and you've had no sort of serious competition in 12 years ever since they definitively distanced themselves from uh Yahoo and Microsoft back in 2002 and um and I think one of the things that's that these breakthrough zero to1 companies have is that they uh they they do things where uh when you when you do something new you um you are in the sort of happy place where um you're not sort of directly competing with too many people and then you have a product that you are offering the world that does not yet exist and uh and that's that's that's what makes these franchises uh so so extremely valuable um I think this question of competition and Monopoly there's sort of many different reasons it's it's it's sort of I think very underappreciated uh you know sort of there's always this Nuance where the people who have monopolies generally don't talk about it and the people um who are in perfect competition pretend that they have something special so you know if you if you're if you're running Google um your talking points are you know we're a we're not a search company but we're a technology company and we're competing ferociously with apple on the phone and with Facebook on social and Microsoft on office and we're building cars and we're competing with the car companies and we're just there's like competition everywhere and uh we're not the Monopoly the government is looking for um and then um and then if you were say trying to open a restaurant in London you would say uh you know you're trying to get investors to invest in your restaurant you will um you will say something like well this is a completely uh and and then the investors will say well I don't want to invest in a restaurant it's a bad business it'll lose money and you'll say no this is a completely different restaurant it's oneof a kind it's the only French nepes Fusion cuisine in um in London and in like uh you know South Paddington or wherever in some small area of London and um and so you will again artificially Define what you're doing in a way that's very different and so so I think this question always gets obscured by the people inside these businesses and that's why it's this very important idea that's uh much bigger than than we think it is but I think it's also um we also um get trapped in uh competitive for sort of a psychological reason it's often um we often find it reassuring to be in crowds to do things that lots of other people are doing um and um and uh you know the word ape already in the time of Shakespeare meant both primate and to imitate and there's something about human nature that's disturbingly lemong like sheeplike aplike her like um and uh and this is this is always a dangerous sort of behavior pattern when you're starting a business business because when you're in a crowd it's the crowd of people you're you're competing against uh um and you know you know I've been a sort of very big critic of uh of the uh higher education system uh because I think that sort of encapsulates this ideology of competition in its a purest form where you have to compete on you know all these sort of tests and jumping through all these hoops you get to the same sort of uh short list of elite universities in the US then you go into the same short of elite jobs and um and a lot of it gets uh gets driven by by people not really asking uh what are they doing why is it important but you know when you're focused on competition you're always focused on beating the person next to you and you do get better at the thing you're competing on like if you're on a high school swimming team you'll swim a little bit faster than you otherwise would have swam but uh you often lose sight of what's truly important and truly valuable and there sort of this autobiographical part of this book where if I had to give advice to my younger self uh um you know I was I was hypert tracked I was I was you know in 8th grade um Junior High School one of my friends wrote in my yearbook that I would surely get into Stanford in four or five years time and that's what happened and then I went to Stanford law school then I sort of got tracked into sort of a big law firm in um in Manhattan uh where from the outside everybody was trying to get in on the inside everybody was trying to leave um and um you know uh after I I left after seven months and three days uh one of the uh one of the uh one of the people down down the hall for me told me it was very reassuring to see me leave he had no idea that it was possible to escape from Alcatraz which um was of course all you had to do was go out the front door but uh but it was uh um it was again the so much of people's identity was wrapped up in winning these various competitive dynamics that they could not think about this so if I gave advice to my give advice to my younger self it would be to you know really think hard why was I doing these things was I engaging in um was I doing these things because I genuinely wanted to do them or was it simply a sort of a Prestige status game that I was playing uh which um and I think if I was honest that there was way more of that going on and there's I think a lot of that going on with a lot of us that we should always uh think hard about um about doing less of second uh second contrarian idea I'll throw out is I think the sort of a conventional view is that there are um that uh that the are not that many answers left to this question what's true that uh nobody agrees with you on we believe that all these truths have already been discovered in the past um and you know and maybe there are still some things but they're just about impossible to figure out so there are conventions that we understand there are Mysteries that nobody can figure out um but I I by contrast think there still a lot of things left on intermediate level there are a lot of things that I call Secrets which are truths that are hard but possible to to discover um and um and I think I think there is always a secret at the core of every great business there's some um some sort of research program that uh or some some area that that people are really really focused on they think about really hard and it sort of advances um their thinking to the point where they get an understanding about something that other people do not yet have you know at PayPal we were very interested in uh um um um sort of the cryptocurrency and um encryption technology and currencies and could they be intersected could you build a new digital currency this was a question that you know animated us tremendously we didn't quite succeed in building one at PayPal even though we had t-shirts that said that we're going to be the new world currency we didn't quite succeed in that in that goal but uh but that sort of a in-depth substantive Focus actually did help us think really hard about how do you architect a new payment system how do you do certain things differently and it it was was a key part of inspiring us and I think there are sort of many Secrets left and this is something um this is something we generally do not understand and it's it's it's not obvious where one should look you know if you're if you were living in the 17th or 18th century you could look at a map and there were empty spaces left on the map and you could become an Explorer and go and discover those places so there sort of a natural geographical sense in which there were Secrets left or um in the 19th century uh there were still you know places in the periodic table of elements that were empty and you could sort of do some basic chemistry and find some secrets and and there's sort of a sense that maybe geography or chemistry are fields that are you know basic chemistry and geography sort of fields that are closed but I think most fields are not like that I think most fields are still ones where there's tremendous amount of innovation possible certainly this is true in um in the uh computer field where we've seen uh massive innovation in in recent decades in the world of bits computers internet mobile internet that whole ensemble um I I think we've seen less innovation in the world of atoms um uh you know Transportation um energy clean energy biomed biotech um space travel all the kinds of things people thought about in the 50s and 60s but I think it's um I think it's not because there's some law of nature that it's hard to innovate or impossible to innovate in these areas it's just that there's sort of been this cultural change where we haven't tried as much and and we you know it's it's it's sort of there's there's a lot of this has a self-fulfilling character if if you think that you can't find a secret then you're not going to try and you will not look and you will not be a person who ever finds one and uh and uh and and uh and so failure pessimism can have a self-fulfilling character as uh conversely if you think there's a lot to be discovered um progress can accelerate and more more things can happen so second sort of uh contrarian truth that I believe to be true is that there actually are many Secrets left to be discovered third um third uh third idea and this one's maybe a little bit bigger picture I I think that uh in a uh successful 21st century this century is going to be a a a great and peaceful Century I think there are two big trends that have to continue there's a trend of globalization and a trend of technology and I think it's always important to think of these two things as very different globalization involves copying things that work horizontal growth extensive progress going from one to n um and I always draw globalization on the on the x-axis technology involves doing new things going from zero to one vertical or intensive progress I always draw it on a y- axis and I I always set up that contrast because I think there's a tendency for people to um to Define technology and globalization as synonymous and I think it's worth thinking of these as very different vectors of of of progress we um um the U the uh and if we sort of think about the last 200 years there have been many periods where we've had one or the other or both and which again underscores how they're different the 19th century you had tremendous amounts of both globalization and technological progress you had both in the 19th century 1914 after World War I starts globalization goes into reverse um and uh there's less trade the world becomes less connected you know Iron Curtain goes down parts of the world more or less are completely cut off and it doesn't really resume till uh Kissinger goes to China in 1971 and we've had the last four decades have been ferocious globalization technology has been we had a lot of progress 1914 to 71 since 71 I would argue it's been more limited and um and with a stress on on the on the computer side and so we've had a era of globalization without technology and um sorry uh technology without globalization and then globalization without technology that's characterized the last 100 years in in contrast to the century that came before um it's this ort of shift from the 50s and 60s today is in in in the way we talk about the world in the 50s or 60s you would have said that the world was divided into the first world and the third world the first world was the place where accelerating change was happening the third world was the place that was sort of permanently screwed up and broken today we speak of the developing and the developed world the developing world is that part of the world that will copy the developed world and um and it's sort of this Convergence Theory of globalization China is the epitome of it has a very straightforward plan for the next 20 years it's just going to copy everything that works in the west it'll skip a few steps but it has a very straightforward uh plan ahead of it but um this developing developed dichotomy while sort of prog globalization is also implicitly anti-technological because it tells us that we are living in the part of the world that is developed that is done that is finished where nothing new is going to happen that's stagnant where we can expect nothing to change uh where we can expect the the younger generation to um do do no better than their parents maybe worse um and um and I think this is a conception of our world that we should resist very strongly uh and um and I think we should not even accept the idea that we're living in the developed world that we're living in some sort of end of Technology era um and so I think we should um always return to this question how do we go about developing the developed world thank you very much I was going to ask you which important truth to very few people agree with you on but I think um you answer my question there's a big theme in the book a number of interviewers uh actually still ask me that even after after I gave a talk similar to this one so you paid you paid way more attention I stopped there's so much we could focus on in the book but um there's a theme that comes up repeatedly which is the value of the Monopoly business you talk about the history of progress is a history of better Monopoly businesses replacing incumbents yet if you look over the last century monopolies haven't always acted in the most enlightened way they haven't always benefited the consumer as governments would have hoped and governments have often had to come in and regulate isn't this a bit of a pure view of the value of a monopoly well um Al so there's there's always a you have to always distinguish looking at this from the inside and the outside so from the point of view of someone starting a business um from the point of view of a Founder an entrepreneur you want to always build a monopoly and it's it's it's always uh great to have one from uh from from the inside um and it gives you way more cushion you can do all sorts of things much better you can be you I think you can be a more ethical company when Google says uh don't be evil on on one level it's a marketing Ploy but on another level you can you can um you don't have to care about nothing but money in a company where you're making so much of it so um so um on the inside I I think you always want to want to have a monopoly from from the outside from the point of view of society the question are are monopolies good or bad and I think they are I agree with the critique of monopolies in a stagnant world where it's like the Parker Brothers board game and the monopolist is just a rent collector or a toll collector um but I think that if if you have a dynamic world where people are inventing and creating new things um um these are actually positive so when Apple builds a smartphone that just works it has a monopoly on the iPhone for many years and I think it still has somewhat of a monopoly in various ways today um and that's a good Monopoly because it created something that did not exist before and the sort of the Common Sense way to differentiate good from bad monopolies lines are a little bit fuzzy but the common sense way is just do consumers think it's a good thing or do they think it's a bad thing so they might think the post office is a bad government structured Monopoly uh whereas Apple uh would be would be seen as a as a good one but is it healthy for the wider economy if a single company like Facebook owns so much data on people's preferences that real-time knowledge based on what we're doing mobile and they can set the rules they're more powerful than governments well I um you know I certainly much prefer Facebook tone the data than than the government I think that um I think that uh I think I think I think I think you have to draw all these distinctions um you know it's it's uh there there there's an important debate about privacy there are important questions about um about um about how one draw these lines but I think the uh you know Facebook product has has created a lot of value in the world it's made the world more connected it's helped people stay in touch with their friends um and I think it's been a it's been enormous net plus to to people um people all over the place um you know I think um so yeah every startup in its pitch talks about it's going to come and disrupt a sector um you say in the book that disruption is a self- congratulatory buzzword for anything trendy and new uh well I I S certainly go after all these different uh all these different uh categories uh I'm I'm always skeptical of anything that gets you know used uh this much non-stop by people so uh you know the uh disruptive person in elementary school get sent to the principal's office disruptors are people who look for trouble they typically find it um you know um uh um Napster was a classically disruptive business that sought to take on the M disrupt the music industry um and you know it's uh had this disruptive name you know you what do you what do you nap you nap some music you nap a kid you it's it's uh and um and um and so there's always a sense where um um disruption already starts by thinking of itself in competitive terms where you're defining yourself against somebody and I I think you don't want to take your bearings from breaking things but rather from building things so we have to go for an entirely New Market I I I think those are the the best companies are ones that go for for uh broadly uh new markets uh you know it's certainly if you if you go after an existing Market that's where uh that's where you're going to get the most ferocious resistance the most intense competition um you know PayPal went after online payments you know visa and MasterCard were not altogether happy because it was a new market they thought they perhaps should have owned but um but it was not like we were taking away business directly from them and in some ways we were actually expanding the universal payments that they enabled and so at the end of the day they didn't come after us nearly as much as they would have had we say tried to disrupt their business in some direct way so why didn't PayPal become the default payment mechanism why didn't PayPal become Bitcoin well I think payment mechanisms and currencies are are are related but quite different uh sorts of things um so I think yeah uh PayPal set out to build a new world currency we did build a rather powerful Internet payment system um Bitcoin I think at this point is is somewhat the opposite uh it has actually created a new currency at least on the level of speculation um and the payment system part of it actually rather lacking where it's hard to act in Bitcoins people will do it for things that are illegal and they can't buy otherwise but for legal payments it's actually quite hard to do and so I think the sort of the opposite of uh set of things from from from that which PayPal had it's it's it's quite hard you know in general I think a currency is a much harder thing to build than a payment system because it requires a much larger Network effect you know it's much harder uh to buy into it um you know the uh you know when I was um I don't have it on me but you know the the prop that I would always use when I was uh a CEO at PayPal was you'd hold up a $100 bill in front of an audience and it would be like hypnotic everybody would want it and there's sort of this network effect to money it's very difficult to create a new currency because you're competing with um with an existing very large Network already there's a lot of people in the valley though putting big money into Bitcoin businesses thinking this is going to be the one are you not excited I am I am still a little bit more skeptical because they've not yet gotten the payment system part of it to work but uh but I think that's the you know if you said what would make me more bullish on bitcoin it's if there were more um more legal payments happening on it so when you meet entrepreneurs what personal characteristics do you define in those that you think are going to be the big success stories well I think um I think these um these uh zero to1 companies they're they're generally not solo efforts it's always a bit of a team effort of people working together and so you you know you want a great business idea you want a great technology you want very talented people but one of the very critical question is how well do these people actually work uh work together and um because I think I think the most common cause for failure is the people just don't get along and sort of blows up uh it blows up internally uh way before it gets destroyed externally by by comp external competition so people internally don't get along um the uh and so I question I always like to ask is the prehistory question you know how did you meet how long have you known each other all those sorts of questions bad answers are things like we met a week ago at a uh entrepreneurial networking event we both wanted to start companies and um and that's sort of like saying we got married at the slot machines in Vegas uh you might hit the jackpot probably very bad idea um the the the good answers are things like we met and we've been working for a number of years we've been friends for many years you know there's some complimentarity maybe you know one of us is focused on the technology side the other person's focused more on the business side uh so that's that's that's a good answer I don't know if you notice a correlation between very successful Tech businesses particularly coming from the valley and people with let's say limited social skills there is a spectrum are these the sorts of true correlations that you see there is there is a curious correlation where some sort of mild form of Asbergers uh seems to be linked with with a lot of these companies and I I think we should always flip that around and um think of this as U like really an indictment of our society uh and ask the question what is it about our society where people who are uh socially well adapted are um subtly talked out of all their original ideas before they even have them because they pick up on all these social cues and they can sense oh this is a little bit too weird this is a little bit too strange it's not quite respectable what do people think of this and uh you are talked out of your truly original and creative ideas before they are even fully formed you're picking up on all these social cues and they um and they they lead you to to give up on them so I think I think there is some sort of dynamic like this um they've done this study at Harvard Business School of which which you can think of as a very non- asger like place where you have people who are super well well adapted socially they're very extroverted uh they don't have very strong convictions generally um and they are sort of in this hot house environment for two years where they talk to one another about what they will do for the rest of their lives and uh the the sort of conclusion at the end of the two years is that the largest cohort generally goes into uh trying to catch the last wave which is always somewhat of a mistake in 1989 everybody wanted to work for Michael milin it was one one or two years before he went to jail for all the junk bomb Trading people were never interested in technology or Silicon Valley except in 99 2000 at the top of the dotc Mania when they they all moved to Silicon Valley 2005 to 07 it was all real estate and and private equity and um and so I do think uh I do think there's some very odd Dynamic like that so what mistakes are we in the tech world making now you know there are there are probably well there many categories of mistakes that that we're making you know certainly one one one broad category that flows from my My Monopoly thesis is that um you often want to look at small markets if you want to get to be to be a monopoly you have to get to large markets chair so you want to go after small initial markets you know Facebook initially went after the 10,000 people at Harvard it went from 0 to 60% market share in 10 days and that was a very auspicious start and then you sort of grew out in concentric circles PayPal uh our initial Market was eBay power sellers there were about 20,000 of them and we got to about 30% market share within about 3 months again a a very good start um you know the the opposite end of the spectrum um you know one of the sort of one of the big red flags in any presentation I get is where the first slide says we have a big market and this was this was a this was a mistake that all the clean tech companies in the last decade made where um the first slide was we have a we're in the energy Market it's measured in hundreds of billions or trillions of dollars uh and we are a tiny minnow in a vast ocean and uh what that means is that you have to not just beat the other um nine thin film solar panel companies but then you have to also beat the other 90 other types of solar panel companies and then you have to beat the wind companies and the fracking companies and the U and then you have Chinese manufacturing come out of nowhere and so you're sort of minnow in a vast ocean not you want to be I think the phrase you use is most clean tech Founders would have been better off opening a new vegan restaurant in Palo Alto but you do talking they're both trillion dollar Market Restaurant Market is also trillion dollar trillion dollar market but then it does come down to timing because there's a lot of very effective clean tech businesses now one of them is Solar City Elon musk's business and his cousins which you have invested in um and is it about choosing the timing well I think it's you have to get a number of things right so I think you have to get the Monopoly question right you have you have to get the timing question right you have to get you know there's be I I think you want to have some secret that you understand that other people don't you want to get the team right so there a whole a whole set of these things that uh that that uh that come together um you know I I I I I think that um you know I think Tesla is a very impressive company that that that has that that has has has worked uh incredibly well um and in some sense they actually started with a with a what was sort of a relatively small Market electric sports cars were a actually a smaller Market than the energy Market as a whole it was still a pretty big Market it was still non-trivial to to raise the roughly half billion dollars in capital it it took but um but it was about as they to did about as um as as it was a small Market that was about as big as you could possibly be Elon Musk your co-founder at PayPal as well as producing cool electric cars and solar panels is sending space Rockets up eventually to Mars yes just the space station the shortterm how do we get more Elon musks you know it's it's um I I I I you know it's always I don't I don't know if there's a simple formula for this we had um um it's uh you know the the PayPal business was actually quite successful in producing a series of these entrepreneurs there been uh about uh of the 220 people who were involved with uh PayPal in um 2002 in in Silicon Valley uh there had been about7 billion doll plus companies started by them so elon's Tesla and SpaceX I was involved in starting paler friend Reed Hoffman started LinkedIn um YouTube YouTube uh Yammer Yelp um and um and so it's always a question what was it about PayPal that was so conducive to this and it's it's always hard to know because whenever you have just like a sample size of one it was one company sort of happened you know it's always hard to to Really um be too scientific about it but but one of the variables that I think may have been very important was that the lesson people learned inside PayPal was that it was hard but possible to build a great business so we had a lot of ups and downs there were a lot of challenges but at the end it was a very successful exit to to eBay um I think the lesson people learn in most of the companies they're involved in is very different they're either in companies where things fail and the lesson they learn is that it's impossible to build a great business so if you work in a failed company you will typically the next time around try to do something uh less ambitious you may succeed but it will be a lesser business uh because you will aim for something smaller um or if you're in a super successful company where everything just worked seem seamlessly from day one a Microsoft or a Google um you will learn the lesson that it's easy um and I think uh to build a great business which I think is just as toxic as the lesson that it's impossible and so I think most people end up finding themselves in contexts where they learn that it's easy or impossible to build great bu businesses and these are equally mistaken lessons I think the the intermediate one that uh and the intermediate mindset that it's hard but possible is a is a critical component uh to it pretty much every major city around the world now is trying to become the local Silicon Valley Consultants are making a fortune teaching Innovation to governments um how reproducible is the success story of the valley you're in London now we have a lot of very very successful companies do you take London seriously as a tech center well I'm I'm always skeptical of copying things so I I think just like you shouldn't try to copy Microsoft Facebook um copying Silicon Valley is probably also the wrong idea um you know one one reason it's the wrong idea is we don't even actually know what makes Silicon Valley work maybe it's the weather maybe it's non-compete Agreements are not enforced in California maybe it is there are these crazy Network effects there sort of all kinds of um you know uh different explanations one can give and so um even if we wanted to I'm not sure we even know what what makes it work and then I think more fundament Al when you're when you're copying something you already are setting yourself up to be defined in a lesser way you know if you're the if you're the Oxford of Iceland that's not quite Oxford and and and and so in a similar way um uh calling things um you know there's some way which you don't want to you don't want to Define your way yourself in a way that's that's derivative the something of somewhere is often the nothing of nowhere um I think that um I think instead what one should try to do is think really hard what are the kinds of things that uh could be done say in London better than anywhere else in the world um you know I think one of the things that uh that so I think for example you know we're we're investors in transfer wise and I think I think think is doing great as a as a company uh and I think uh I think there's perhaps a lot of innovation in the area of financial technology that could take place in in London because um this is a finance center it's a natural thing that London can do much better than Silicon Valley Silicon Valley is not a Global Financial Center um and uh it's London I think is also better than New York by the way since um uh people in New York uh New York's very self-hating people hate the finance industry in New York in a way in which they don't hate it as much in London so if you start a financial technology company in London this would be a cool thing to do it would not be considered as cool a thing to do in New York you invested I think in Deep Mind an artificial intelligence company and you have sort of you have sort of eccentric British Tech technologies that get developed too um and uh this is a company that Google bought for 400 million pound and they hadn't released a product um so it's it's the uh yeah it was sort of a Manhattan project for AI that was the that was the uh uh that that was and Remains the the goal of deep minds they assembled a phenomenal uh phenomenally talented team and and I think uh so I think are there are things like that that that that work uh that work quite well but but you know there's no reason that innovation in general has to happen only in Silicon Valley you know there Silicon Valley does have some advantages but it probably also has disadvantages the network effects that help Silicon Valley um I think also in some cases probably um lead to more leming like and crowd likee Behavior which I think is is so problematic do you worry about what artificial intelligence will do to us to our ability to earn a living I think I think I I'm of the view that strong AI is still quite a long long ways off um you know if we ever were to get artificial intelligence if we were ever to build computers that are as smart as human beings in every way uh this would be an a momentous event this would be you know it would be as as significant as extraterrestrials landing on this on this planet um and you know if aliens landed the first question would not be about the economy and what does it mean for your job the first question would be political are they friendly are they unfriendly and so um so I think uh I think uh to even frame it as a question about jobs is to understate the importance or or or um seismic nature AI would represent um I think short of strong AI however I I think people um are way too worried about computers in this economic sense I think this we have we live in a we live in a society that's generally um is is we live in a financial and a capitalistic age I I sort of argued elsewhere that I think we do not live in a scientific or technological Age and and uh and most people in the US Western Europe really don't like science they don't like technology they're biased against it and all sorts of ways it's true of people it's true of the politicians it's true of the governments it's true of the culture the easy way to see this is you just look at all the Hollywood movies that basically show technology that doesn't work that kills people it's destructive you can choose whether it's the Matrix or Terminator or Avatar or elisium I watched the Gravity movie the other day you'd never want to go to Mars even to outer space you'd be much happy happy happier being back on some muddy Island somewhere um on on this world and that's that reflects the sensibility of most people that the future is something to be feared and that should uh that we should try to that we should try to prevent and I um and that's why I think uh people who are involved in the scientific or technological worlds are the counterculture in our society today uh and um and it is sort of a very unusual perspective to think that the future something we hope for so roundabout answer um so I think this this idea that um today's computers are uh replacing people um is just another one of these technological unst narratives we have I think it's not true I think computers and people are fundamentally different uh they're good at really different things they're fundamentally complimentary not substitutes the you know the the much bigger challenge for the middle class and the developed world comes from globalization because people in India people in China are actually not that different from us they can substitute for our labor and that's where that's where the substitution is taking place I don't think we should stop globalization but we should you know it has some problematic aspects I think technology has far fewer you also think that technology can help fight aging how much can we solve this problem in our lifetimes well we we've made a we've made a lot of progress over the last 150 years you know the the maximum life expectancy in 1840 was 46 years among Swedish women today it's uh something like 87 88 years life expectancy has gone up by about 2 and a half years um a decade so every day that you survive your life expectancy goes up by five or six hours um I think that um um and so there's always been this debate between the uh you know the mathematicians who sort of extrapolated these curves and the biologists who always argued that um you you weren't going to make any more progress you're about to hit the wall um and I I would I would tend tode with the mathematicians over the biologists and I think I think we can do we can do a lot more um I I don't think we're going to solve you know um we're going to find sort of a single magic pill that um that makes people live forever I think it will be a lot of specific blocking and tackling but I think you know I think we could make a lot of progress on cancer I think there's no reason that we could not cure Alzheimer's or dementia you know one out of three people at age 85 has dementia I think we could be doing a lot more and it's it's really um you know it's always again I think it's outrageous that our culture is so complacent about these things and uh again I think there's room for um a lot of progress in these areas so when you hear of supporters of the singularity people like rers well talking about being able to defeat aging the body can live forever what do you think well I think I think it's you know I think this has been this has been a core um animating idea of the entire modern scientific project uh you know the uh the search for U the Water of Life that the Alchemists had was as great as for the thing that would trans mute everything into gold they were actually the Alchemists were more interested in living forever than in finding uh the the thing that substituted everything for gold it was you know Francis Bacon the new Atlantis this is this is the whole the whole project of uh of modern science and so uh so you know it's it's hard to know how much you know how much we can you know change the world from a place where life is nasty brutish and short but um but I think uh I think the the moral thing is to is to push it as far as we possibly can in the other direction you know I know there are all sorts of arguments on the other side people say that you know death is a natural part of life uh and I to which I always respond that I think um it is at least as natural for us to be fighting death and uh and I think that you know we have this modality where we are either we either accept it or deny it and I think most people are sort of in the schizophrenic mode where they um accept death and deny it at the same time um and what acceptance and denial however have in common is they're both um modes of thinking that stop you from doing anything it's going to happen or there's nothing you can do about it's not going to happen or it's going to happen and there's nothing you can do about it both those modes tell you that you shouldn't do anything and I think we should U be spending more time fighting it I'm quite upset that we're not having an argument actually I feel I'm missing out on the Peter tail experience you're not shouting me down um you're a man with I never shout people down you're a man with political views um you have strong views about education and other things let me just imagine with you that you're suddenly in the white house it's not going to happen let's just imagine I I always think thought I always think fictional thought experiments are very treacherous things though but let's just say you change does it mean does it mean that I'm mayor of the United States I'm kissing babies all day long or mean that I'm dictator how far can we play this thought experiment what would you want to change in how America is run well you know uh my my my uh you know my my single issue is that uh we have to make more progress in science and technology and uh you know I I don't think the government can do that much on this because it's so hostile to these things reflects a population that's hostile think get best we can hope that the government um gets out of the way of of stopping this sort of progress there was a time in the past When government was able to coordinate things and do things I I don't think we're living in that sort of a society anymore you know um the US was able to build a nuclear bomb in three and a half years in a project organized by the government or you could put someone on the moon in the Apollo program the 1960s um these sorts of things would be utterly inconceivable today you know a letter from Einstein to the White House would get lost in the mail room and people would say you know who is this crazy scientist who thinks you can do this uh this completely different sort of a thing um and so I think um I think the the the best role for government in our our world is is is somewhat more limited uh because you know it's I I looked at this the other day there you know 535 congressmen and senators in the US and by a generous count maybe 35 of them have a background in science or engineering um and the rest of them are like in the Middle Ages you know they don't know that solar panels don't work at night or windmills don't work when the wind is not blowing and um and so uh and so I think we we can't have that great a hope for how much Innovation can happen from government if that's if that's how it's constit now again if we can change everybody in the government that's now we're again playing with a very crazy thought experiment though governments are trying to map the brain though Obama's got a big project the Europeans have a big project look there's there's there's a lot of funding that goes to various kinds of ideas but um the the politicization of the funding I think has uh has often had this counterproductive effect you know in much of science I I I worry that there's been a sort of gresham's law at work where the people who are Nimble in the art of writing Grant applications to governments have replaced The Eccentric people who are the truly great scientists and so the uh the um The Eccentric University Professor is a species that I think is going extinct very fast um uh because of the sort of Gram's law Dynamics yes there's there's a lot of money going to projects the projects that get funded are consensus projects the experiments that get done are experiments that everybody thinks will work and experiments that everyone thinks will work are ones that I I think rarely Advanced science at all the interesting experiments are ones that no one thinks work and that actually end up working but those those are very hard to fund politically so you've been back backing a seasteading organization to try and create a new kind of nation how's that going uh that still has a long ways to go um it is it is um it is a it's a small project I got involved in it it it always generates enormous interest um because I think I think people do have a sense that the frontier is somewhat closed and there's something odd and that there's no longer a place you can go to in this world where you could start a new community or new society and so um so even talking about doing something as speculative as building uh communities on on platforms on the ocean um naturally um generates an enormous amount of of conversation uh you know I think I think it you I think figuring out ways to reopen the frontier is is very important it's not clear we can do it geographically because you know the land surface is fully covered outer space is still pretty far away um but uh but I think you know the frontier is this place where new things happen there's sort of a certain logic to California being the place where so much Innovation happened it was the place where the GE graphical Frontier ended you went West to go to the frontier the frontier ended in California and even though it's geographically you know are finished exploring it um it uh it is still a place where a lot of new things happen so on the founders fund website you famously say that we wanted flying cars instead we got 140 characters what have you got against Twitter you have to always put this in some some context so I think that um so so it's it's a I I think that Twitter is a has a great business model um you know it's it's uh it's in some ways so such a good business model that um and when you have a monopoly like Twitter has on on the service they they built out uh you often don't have to run your business uh in quite as operationally tight a way as uh you as you would if you were say opening a restaurant or something like that so so I think uh Twitter is a great business but uh there is always a sense that it's not enough to um take our civilization to the next level and this is always sort of the idea that we need to do things not just in the world of bits we should also be trying to do more things in in the world of Adams the narratives we always tell in our our media are ones of specific failure linked to General failure or specific success linked to General success and so it would be Twitter is a great business and therefore look at how much is happening in technology um and I think we we need to consider that uh it's it's a specific success that's perhaps obscuring a lot of General failures you know the the uh the cell phones that uh um that distract us from our surroundings also distract us from the fact that our surroundings are strangely old that we live in these old cities where transportation systems are maybe from the 19th or early 20th century where you a lot of stuff has not changed in a very long time and so so it's it's sort of so yes I think Twitter itself is a great company um it's not enough to take us to the next level now I I now it's it's always this often gets framed as you know a critique of the Twitter Founders why couldn't they have built something like flying cars and I think that's that's very unfair because they did build a great company and we shouldn't blame them it really the blame is on all of us for not working on these other things there is a startup in Slovakia that we just featured in wide called erom mobile that is making a flying car by the way what do you think when you see Uber value at 17 billion and WhatsApp bought for $19 billion and Snapchat worth somewhere above1 billion what do you think about the the valuations well uh there there is sort of always this recurrent question whether we have a bubble in technology like we had in the in the late 9s uh I tend to think that there is no bubble in in Tech because uh these bubbles are psychosocial phenomena in which you get the entire public uh involved and um and the public has largely been absent this time around the IPOs are happening very very late they're far fewer of them than there were in the late 90s and so you do not have the sort of public frenzy around technology that you had in 99 uh 2000 uh you know if I had to identify a bubble today we had a bubble in Tech in the '90s we had a Housing Finance one in the last decade the obvious place where the bubble exists today is in uh government and in you know government's printing money it's in uh it's in government bonds probably which is what what a government's are sort of buying up with all the money they're Printing and so I think that the place that are bubble and you know in some ways the interest rates that are set by the government touch everything but uh the center of the bubble are probably things that are most like government bonds it's government bonds corporate bonds maybe it's uh parts of the housing market that are very linked to fixed income maybe it's parts of the equity Market that behave like fixed income so it's stocks that pay high dividends or it's uh or it's these sort of value stocks where you have very predictable cash flows um whereas the tech stocks are primarily valued on growth which is a very different variable I personally have about three quarters of my net worth tied up in a liquid tech stocks in Silicon Valley and elsewhere and um I I actually think that it's one of the best places to hide from this massive government bubble that exists everywhere else so I I think there is a bubble it's not in Tech it's in government and I I invest in Tech in order to hide from the bubble in government what is the one that you wish you'd invested in that you missed well there always are there always are um many uh misses uh that one one has um uh but you know I think sort of the the misses that really that count as misses are ones where you spend some time talking the company you have a real conversation you really think about it so I missed out on you know the series aound at YouTube I missed out on the series a round at zingga I had you know lots of conversations with Mark Pinkus and known him for many years and was very close to doing it but far bigger Miss than those two uh was not doing the series B round at Facebook um and um and I think it's sort of very instructive uh sort of set of mistakes that happened um you know on the on the on the um and we done this I was the seed investor did the series a round um August of 2004 it had sort of a post money valuation of 5.6 5.7 million and know and Facebook was just growing very fast sort of eight months later um you know they sort of um got this financing Round The Washington Post was offering 50 million and then you know Excel uh Venture fund Silicon Valley came in much higher valuation at 85 million um it was 12 times the price per share of the of the uh of the first round the middle of delution but so with much higher valuation in just eight months and I this this conversation with uh with Mark and was like you know why don't you do the this whole round kind of like you and said it's like gone up so much and and it felt like we had really maximized the valuation um and and and part of the psychology was that on the inside it didn't feel like things had changed that much you had this horrible graffiti art on the wall wall of the office there were sort of it was still only eight or nine people so every day that you went to the office uh you it actually felt like it was the same company you had been at the day before um it was true there were these uh these uh charts where things were moving in a very happy North northeasterly Direction but um but they were very these were abstractions that people I think didn't fully fully get in away and so um so um and so whenever you have these massive upgs in companies led by um by smart investors um almost as a rule of thumb uh they are undervalued and I've done this back testing on this where the steeper the UPR round the greater the undervaluation and it's because when these companies have momentum people underestimate the momentum they underestimate when things change they underestimate how much things have changed and it's not just true of the outside investors it's also true of the people on the inside you know I I course corrected on this in the series C round in Facebook which was at 525 million pre just one year later and so it's was like 5x again and uh and it was sort of this question what did what had I done wrong and and I think I think it was like okay this this makes no sense at all this is a really crazy valuation but I'm I'm going to still invest because um there's something about the momentum that people are missing and that that was that was actually a um round that I'm very glad to have participated in how did it work out well so it Facebook is worth worth more than 550 million which is the post money on that round today just last question for me before I revoke My Monopoly position and invite the room to ask some questions what do Facebook become long term you know I have to always be a little bit careful what I say about Facebook since I'm on the board of directors there but I think I think you know I think that I I think it is um it is a founder-led company uh which means that um it will uh it will it's very focused on innovation on continuing to develop a lot of new things broadly in this in this social networking space um I'm I'm generally the most bullish on the founder-led companies in Silicon Valley you know Amazon Google Facebook uh jobs did a phenomenal job when he came back to Apple and there is a certain amount of Freedom that Founders have that the um politician CEO people that often get brought in to run these companies simply uh do not have so we have two rules for the questions now um the first rule is it must have a question mark at the end and the second rule is it must be a genuine question about Peter not some of your own projects and anybody who disobeys gets put on a floating country somewhere of California um if we can have the lights up a little and there are people with microphones and if you would like to ask Peter a question raise your hand and there's a hand that's just shut up over there and we can have another microphone somewhere at the back over there cheers uh I'm quite interested uh because you obviously started speaking and the book's mainly about zero to one about making you know a completely new market in which to fill I wanted to get your opinion on kind of the business model behind rocket internet because of course they're the complete opposite of we just completely copy and imitate at speed well I don't I don't think it's a technology company I mean it's a it's a globalization play it's it's about copying things that work and uh and sort of um um copying them elsewhere in the world um and so uh it it may work uh it uh it uh you know it's uh you have I think there are some tough questions about how is rocket different from ideal lab which was sort of had had this incubation model that that blew up famously in the late '90s I I think the the argument rocket would make that is different is that it has a formula that is just repeating mechanically um and I think it's a it's a play on globalization I I I'm not interested that much in globalization I think it's important but I'm much more interested in technology but I think investors have to evaluated as as a as a you know as as a globalization place like investing in McDonald's and China or something like that question over here and then another hand on this side I've got a mik here if that's okay my name's Katie I'm from nine others and thank you very much Peter I really appreciate your time speaking to us I think we've all been inspired by your book and by your story and what I'd like to know is what inspires you where do you go to to find inspiration from what you read or who you meet or places you visit and that sort of thing well it's um I'm always really bad at answering these sorts of personal questions but but I I think it's I I find tremendous inspiration from just uh all the terrific people that I I get to work with uh you know so there's something there's a dynamism to that that's uh that's incredibly uh incredibly motivating and inspiring and uh you know I think there are all these ways in which uh one can be pessimistic about certain Trends and certain things in our societies but uh but I I think there there is sort of a very powerful an antidote in the sort of indomitability of the human spirit that is sort of a an enormous part of the of the tech scene in Silicon Valley and elsewhere is there a microphone with somebody here hi Peter uh one question why you are so sure that the next Innovation Way will be in the states if necessity is the model of innovation I think there is a very interesting story in Japan also the flash already tra for 20 years of good infrastructure in robotics Fukushima tragedy geopolitical problems so what about Japan well I I think Japan is is very interesting there's there's definitely a logic I I would look I I generally think the place to look is the developed world because that's that's the place where people need to do new things the developing World they can just copy stuff um so I I I do think you look at us Scandinavia you know northern Europe UK Israel um and then Australia Canada New Zealand um and and then I think certainly Japan is is a very interesting one it's um it's you know it is um it's culturally not a place where people typically start companies so that's it's sort of so uh the idea of doing something new in a small company is is sort of a is is a is a modality that does not exist um so I think I think there may be lot of innovation in Japan but it may strangely happen in in more in larger companies than in smaller companies that would be my guess but I think I think it's very interesting and it's uh it's definitely um it's still the third largest economy in the world and it's weirdly uh very off the radar how much of your investing are you doing outside the US the bulk is the bulk is in the US it's always you know it's always we it's it always ends up uh there's always sort of some connection we have to the people where they become strongly recommended in one way or another but we we've done you know we've done some things here in the UK we've done a few things in the rest of Europe we've done some things in Israel we've done some things in Australia New Zealand so we've done some things all over somebody have a microphone on this side um I was going to say you said that you know be asperges is the key to Innovation and Facebook is taking all this data and curating a social experience and it's it's almost rewarding people to kind of complete continue to socialize so how does that they take all that data and make people innovate again I mean you've got a huge Community listening to you and you're kind of curating their social experience but are you necessarily helping that population innovate for themselves um and how do you sort of take their data forward and do something more exciting like maybe what Google's doing um and can you kind of innovate Society rather than reward it for sort of gaming social psychology well I I um you know I I think I think there are sort of all these different different parts going on I I think I I certainly do think that um that if you if you are constantly getting feedback on on your ideas and discouraged from having unconventional ideas that that is a big problem I'm not I'm not convinced that that's fundamentally what happens on on on Facebook but uh but it is I think it's a problem that we have um in society in general uh and it's something that we all need to work on overcoming um and it's it's it's very unfair to blame it on any single company I would say I don't think Facebook's doing it but um and it's unfair because I think this is this is a problem that's endemic to society we live in society we're sort of social beings on some level and and so we we we always uh take our cues from the people around us and often get talked out of our good ideas from the people around us so it's it's a very deep problem prob space for two more questions and there was somebody over here um hi there I was just wondering if you could um actually describe your first meeting with Mark Zuckerberg you know he was 19 years old uh he was uh he was very um introverted and rather quiet at the time he still is somewhat introverted uh he um you know it seemed like the we we done a lot um my friend Reed Hoffman and I uh had done a lot of homework on the whole social networking space um you know Reed had started a company called social net all the way back in 1997 seven years earlier uh and so we had thought about this a lot and um and so it actually didn't depend that much on what happened in the meeting at all we were going to invest just about no matter what so we had the meeting for an hour and then we wrote the check but uh we had we done our homework uh beforehand um I I think the the odd question about Facebook that's uh very hard to get a handle on is uh why nobody in Boston invested between February of 04 and August 04 when I met them in Silicon Valley um and I have I have this sort of crazy theory that uh that um there's there's always sort of there always are these competitive Dynamics people get caught up and that're they're very unhealthy and I think one one version of it is that older people um often don't want younger people to do better than themselves and uh and so if you're I think there were these Venture caps in Boston that were personally annoyed at how well Zuckerberg was going to do and it sort of stopped them in a way from making the investment um what I told Zuckerberg when I when I made the investment was that I I sincerely hoped that he would be far wealthier and Far more successful than I had ever been and I'm very glad that happened I interviewed Reed Huffman um who was given the chance to lead the first round of investment he was working with Shawn Parker Shan Parker said hey I've got these guys working at this place called the facebook.com and and Reed nobly said well I think there might be a conflict of interest for me but I'll introduce you to my friend Peter so I asked Reed um does he regret that how does he feel now and he said well the money I don't regret but I regret not having the chance to work with an entrepreneur like Mark um last question Peter fascinated by your comments on education um and uh love loved some of the points you made around um secrets and uh really you know how we're going to try and take civilizations to the next level my question is really um what's your view on mukes um being the facilitator of that opening up the new Pioneer um so it's always it's always a little bit uh tricky if one has uh too many companies of a given category so uh you know you don't want to be the fourth online pet food company or the 10th thin film solar panel company uh and and so I I I think there is this challenge of how are they differentiated from one another What will what will really work and so I always think that if we describe it at the level of mukes we maybe haven't uh defined the problem uh properly um but I I do think you know I do think we're at a point where uh the universities are going to change radically it is a it is an extremely um you know it's an extremely uh corrupt system we have at this point there's we have an education bubble in the US we have a trillion dollars of of student debt uh to a first approximation this has gone to pay for a trillion dollars worth of lies about the value of the education people have received um and um and I it's not at all obvious yet though what's going to what's going to replace this or how it's going to change um my um you know the the uh you know the I'm I'm I'm somewhat skeptical that it will be replaced by any sort of single unitary system you know I have this Fellowship for uh young people to start companies um and you know it's not my claim is not that everybody should do this I don't think everyone should become an entrepreneur and I think I think there is no one siiz fits-all I so I think the future will be much more heterogeneous much more diverse in terms of what people do um and uh what's what's really anomalous is the sort of unitary tracking where you have to go to an elite College you go to Yale or you go to jail there's nothing else you can do you know um and so the the uh I think the universities are perhaps in um perhaps in the same place as the Catholic church was in 150014 if we go back 500 years where uh you had sort of this monolithic way you this Universal uh way of uh body of knowledge of teaching things the difference between between the Yale and the Harvard political science faculties are probably no greater than the differences between the Dominicans and the franciscans we've all kinds of small debates within this context um uh we are we have a system of indulgences uh that's costing more and more to support this Priestly or professorial class of people uh we are told that uh it's the only way to Salvation um you must get a diploma to be saved if you do not get a diploma then you will go to hell um and I think um and I think the uh I think the the the message that I have that's uh like the 16th century reformers that's a somewhat troubling message is that you have to work out your salvation on your own you have to save yourself and uh and and that's that's I I believe that is the truth but it's a it's it's a somewhat uncomfortable one so Peter has agreed to sign some books kind of at the back outside here um what you must then do is read the book because it's a great tightly written book and then you need to start a monopoly because otherwise tonight's wasted please can we thank Peter teal thank you + +## Competition is for losers + +all right good afternoon uh today's speaker is Peter teal Peter was the founder of PayPal and paler and Founders fund and has invested in uh most of the tech companies in in silica Valley and he's going to talk about strategy and competition thank you for coming Peter awesome thanks uh Sam thanks for inviting me thanks for for having me uh I I sort of have a I have a single eay FS that I'm completely obsessed with in um on on the business side which is that uh if you're starting a company if you're the founder entrepreneur starting a company you always want to aim for Monopoly and um and that uh and you want to always avoid competition and so uh hence uh competition is for losers uh something we'll be talking about today I'd like to um I'd like to start by saying something about um the the basic idea of uh when you start one the companies um how you go about uh creating value and there's this question what makes a business valuable and I want I want to suggest that there's basically a very simple uh very simple formula that um um you you have a valuable company if two things are true uh number one that it creates X dollars of value for the world and number two that you capture y% of X and and the critical thing that uh that I think people always miss in the sort of analysis is that X and Y are completely independent variables and so um X can be very Big Y can be very small X can be of intermediate size and if Y is reasonably big you can still get a very big business so to create a valuable company you have to basically uh both create something of value and capture some fraction of the value of what you've created and sort of just to just to illustrate this as a as a contrast um there's if you sort of compare the US airline industry with a company like Google on search um if you sort of measure by the size of these industries you could you could say that airlines are still more important than search if you just measure it say by revenues there's 195 billion in uh domestic revenues in 201 uh 2012 Google had uh just north of 50 billion um and and so and certainly sort of on some intuitive level if you said uh if you were given a choice and said well do you want to get rid of a all air travel or do you want to get rid of your ability to use search engines the intuition would be that air travel is something that's more important than search and this is of course just the domestic numbers if you looked at this globally um airlines are much much bigger than um than uh than than than than search or than Google is but uh but the profit margins are quite a bit less uh you know they were marginally profitable in 2012 12 uh I think the entire 100-year history of the airline industry the cumulative profits in the US have been approximately zero know companies make money they episodically go bankrupt they get recapitalized and you sort of cycle and and repeat and this is reflected in you know the the combined market capitalization of the of the airline Industries maybe uh something um of the US airline industry something like a quarter that of Google so so uh you have you have a search engine much much smaller than than air travel but much more valuable and I think this this reflects these very different uh valuations on X and Y so um you know if we look at perfect competition um you know there are sort of there's some pros and cons to the world of perfect competition um on a high level uh uh it's always um this is what you study in econ one it's always it's easy to model which I think is why econ professors like talking about perfect competition um it somehow is efficient especially in a world where things are static because you have all the consumer surplus gets captured by everybody and uh and politically it's uh What uh what we're what we're told is good in our society that you you want to have competition and this is somehow a good thing um of course there are a lot of negatives uh it's it's generally not that good if you're you're um you're involved in anything that's hyper competitive um because you often don't make money I'll come back to this a little bit later so uh so I think at one end of the spectrum you have uh industries that are perfectly competitive and at the other end of the spectrum um you have things that um I would say are monopolies and um and they're you know they're much stable longer term businesses you have more Capital uh and um and if you get a creative Monopoly for inventing something new I think it's symptomatic of having created something something really valuable um and so I do think this you know the the the sort of the the the extreme binary view of the world I I always articulate is that there are exactly two kinds of businesses in this world there are businesses that are perfectly competitive and there are businesses that are monopolies and um there's shockingly little that is in between and uh this dichotomy is not understood very well because uh people uh are constantly lying about the nature of the businesses they're in um and this is why this is in my mind this is the most important is not necessarily the most important thing in business but I think it's the most important business idea that people don't understand that there are just the two kinds of businesses and so let me say a little bit about the lies that people tell and so you basically um the basic uh if you sort of imagine that there was a spectrum of companies from perfect competition to Monopoly um the um the apparent differences are quite small because the people who have monopolies pretend not to they will basically say uh you know and it's because you don't want to get regulated by the government you don't want the government to come after you so you will never say that you have Monopoly so anyone who has a monopoly will pretend that they're in incredible competition and on the other end of the spectrum if you are incredibly competitive um and if you're in some sort of business where you will never make any money um you will be tempted to tell a lie that goes in the other direction where you will say that you're doing something unique um that um is is somehow uh less competitive than it looks because um because you want to you will want to differentiate you want to try to trct Capital or something like that so if the monopolists pretend not to have monopolies the non- monopolists pretend to have monopolies the apparent difference is very small whereas the real difference I I would submit is is actually quite big and so there's this Distortion that happens because of the lies people tell about their businesses and the lies are sort of in these these opposite direction let me let me drill a little bit down further on the uh the way these lies work and so um you know the the the basic uh lie you tell as a non-m monopoly is that we're in a very small Market the basic lie you tell as a monopoly is that the market you're in is much bigger than it looks and so um and so typically if you want to think of this in sort of set theoretic terms you could say that a monopoly tells um a a lie where you describe your business as the union of these vastly different markets and the non- monopolist describes it as the intersection so that uh in effect um if you're if you're a non- monopolist you will rhetorically describe your Market as super small you're the only person in that market if you have monopo you will describe it as super big and um and there's lots of competition in it so uh some examples of how this how this works in practice uh so I always use restaurants as the example of a terrible business this always you know sort of my idea is you know capitalism and competition are antonyms um capitalist is someone who accumulates Capital world of perfect competition is a world where all the capital gets competed away so uh you're opening a restaurant business no one wants to invest because you just lose money so you have to tell some idiosyncratic narrative and you will say something like well we're the only British food restaurant in paloalto so it's British paloalto and uh and of course that's too small a market because people may be able to drive all the way to Mountain View or even Meno Park um and there probably are no people who eat nothing but British food at least no people who are still alive and so so that is um that's that's a sort of a fictitiously narrow Market um there's there's sort of a Hollywood version of this where uh the way movies always get pitched is you know okay it's like a college football star you know uh joins an elite group of hackers to um to catch the shark that killed his friend um sorry and so that's now that is a movie that has not yet been made but um but but the question is is is that the right category or is the correct category it's just another movie in which case you know there are lots of those it's super competitive incredibly hard to make money no one ever makes money in Hollywood uh doing movies or it's really really hard and so you always have this question about does the intersection does is it real does it make sense does it have value that one should ask and of course there are startup versions of this where you and the the sort of the bad really bad versions you just take a whole series of buzzword sharing mobile social apps you combine them and you have some kind of uh narrative and whether or not that's a real business or not uh is is uh um is it's generally a bad sign so it's it's almost this pattern recognition when you have this rhetoric of the sort of intersections um it it it generally does not work the something of somewhere is really mostly just the nothing of nowhere and it's like the Stanford of North Dakota uh one of a kind but it's not Stanford um so let's look at the opposite the opposite law is um if you are uh let's say uh the uh the search uh company that's down the street from here and has about a happy 66% market share um and uh is you know is completely complely dominant in the search Market um Google has not almost never describes itself as a search engine these days um and instead it uh it describes itself in all these different ways so it sometimes says it's an advertising company so if it was search you'd say wow this this like it's it's it has this huge market share that's really really crazy it's it's like a incredible Monopoly it's much bigger than it's much a much more robust Monopoly than Microsoft ever had in the 90s maybe that's why it's making so much money um but if you uh if you say it's an advertising Market you could say well there's search advertising is 17 billion and that's part of uh online advertising which is much bigger and then you know all us advertising is bigger and then by the time you get to Global advertising that's close to 500 billion and so you're talking about 3 and a half% so um a tiny part of uh of this much larger market um or uh if you don't want to be an advertising company you can always say that you're a technology company um and so um sorry let me see um and so the and and so um and and the technology Market is something like a one trillion Doll Market and the narrative that you tell as Google in the in the technology Market is um well we're competing with all the car companies with our self-driving cars we're competing with Apple on TVs and iPhones uh we're competing with Facebook we're competing with Microsoft on um on Office Products we're competing with Amazon on cloud services and so we are in this giant technology Market where there's competition in every direction uh you you look and uh no we're not the Monopoly the government's looking for and we should not get regulated in any way whatsoever and so I think one has to always be super aware that there are these uh these very powerful incentives to to uh to distort uh the nature of these markets one way or or the other so um you know the the the evidence of narrow markets in the uh in the tech industry is um is if you basically just uh if you look at sort of the some of the big tech companies Apple Google Microsoft Amazon um they just um they've just been building up cash for um year after year and you have these incredibly High profit margins and I would I would say that the the one of the reasons the tech industry in the US has been uh has been so successful financially is because it's it's prone to creating all these Monopoly like businesses and that's that's um and it's reflected uh by the fact that these companies just accumulate so much cash they don't even know what what to do with it Beyond a certain point um and so so let me say um let me say a few things about uh about how to how to build a monopoly and I think uh I think the one of the sort of very counterintuitive ideas that comes out of this Monopoly uh thread is that um you want to go after small markets if you're a startup um you know you want to get to a monopoly you're starting a new company you want to get to Monopoly um Monopoly is you have a large share of a market how do you get to a large share of a market you start with a really small market and you take over that whole market and then uh and then over time you find ways to expand that market in in concentric circles and uh the thing that's always a big mistake is going after a giant market on on day one because that's typically evidence that um that you somehow haven't defined the categories correctly that and it's it normally means that there's going to be too much competition in one way or another and so I think almost all the successful companies uh in Silicon Valley had some model of starting with small markets and expanding and you know if you take Amazon you start with you start with you know just um a bookstore we have all the books in the world so it's it's a it's a it's a better bookstore than anybody else has in the world when it starts in the 90s it's online there's things you can do you can't do before and then you gradually expand into all sorts of different forms of e-commerce and other things beyond that um you know eBay you start with Pez dispensers you move on to beanie babies and eventually uh it's it's all these different um auctions for all these sorts of different Goods um and uh and what was very counterintuitive about what's very counterintuitive about many of these companies is they often start with markets that are so small that people don't think um they don't think that they're uh valuable at all when when you get started um the the PayPal version of this was uh was you know we started with uh with power sellers on eBay which was about 20,000 people when when we first saw this happening in December of 99 January 2000 right after we launched uh there was a sense that uh that these were all um it was such a small Market it was terrible we thought these were terrible customers to have it's just people selling junk on the internet why in the world do we want to be going after this Market but um but you you know you there was a way to get a product that was much better for everybody in that market you could um and we got to something like 25 30% you know Market penetration in 2 or 3 months and you got some Walkin you got brand recognition and you're able to to build the business from there so um so I always think these um these these very small markets are are quite underrated uh the Facebook version of this I always give is that uh you know the initial Market at Facebook was 10,000 people at Harvard it went from 0 to 60% market share in 10 days that was a very auspicious start um the way this gets analyzed in Business Schools is always um that's ridiculous it's such a small Market it can't have any value at all and so I think the business school analysis of Facebook early on or of PayPal early on or of eBay early on is that the markets were perhaps so small as to have uh almost no value uh and they they would have had little value had they stayed small but it turned out there were ways to then grow them concentrically and that's what made them uh that's what made them so valuable um now I think the opposite version of this is always where you have super big markets and um and I there's so much so many different things that went wrong with all the clean tech companies in the last decade but uh but one one theme that ran through almost all of them was they all started with massive markets and every clean tech PowerPoint presentation that one saw in the Years 2005 to 2008 which was sort of the clean tech bubble in in Silicon Valley started with we're in the energy Market we're in a market that's measured in hundreds of billions or trillions of dollars and um and then you know once you're sort of a a minnow in a vast ocean um that's not a good place to be that means that you have tons of competitors and you don't even know who all the competitors are and so you want to be you know you want to be a one-of-a-kind company where it's the only one in a small ecosystem you don't want to be the fourth online pet food company you don't want to be the 10th thin film solar panel company you don't want to be the 100th restaurant in paloalto um you know restaurant industry is a trillion dollar industry so if you do a market size analysis youd include restaurants are a fantastic business to go into and it's often large markets large existing markets typically mean that you have uh tons of competition very very hard to uh to differentiate so the first very counterintuitive int uh idea is is to go after small markets often markets that are so small people don't even notice them they don't think they make sense that's where you get a foothold and then um and then if those markets are able to expand you can scale into a big monopo business um you know um a second uh sort of there's sort of several different uh characteristics of these Monopoly businesses um that I like to um focus on and U there's probably no no sort of single formula to it and I I always think that uh that in technology there's always a sense that you know the history of technology such that every every moment happens only once and so you know the next Mark Zuckerberg won't build a social network the next uh uh the next Larry Page won't be building a search engine the next uh Bill Gates won't be building an operating system and if you're copying these people you're not learning from them but it's it's and so um there is always um these very unique businesses that are doing something that's not been done before end up um end up having the potential to be a monopoly if you're you know the the the opening the opening line in um Anna Ken is that all happy companies sorry all happy families all happy families are alike all unhappy families are unhappy in their own special way and the opposite is true in business where I think all happy companies are different because they're doing something very unique all unhappy companies are alike because they fail to escape the essential sameness that is competition and so so one one sort of characteristic of a monopoly technology company is some sort of proprietary technology um my sort of crazy somewhat arbitrary rule of thumb is you want to have a technology that's an order of magnitude better than the next best thing so Amazon had over 10 times as many books I it's maybe not that Hightech but you figure out a way to sell 10 times as many books in an efficient online way you know PayPal the alternative for PayPal was using um was using uh uh checks to uh send money on eBay took 7 to 10 days to clear PayPal could do it than 10 times as fast so you want to have some sort of very uh very powerful Improvement in some um on in some order maybe an order of magnitude Improvement on some key Dimension um of course you know if you if you actually come with something totally new um it's it's it's just like an infinite Improvement so I would say the the iPhone was the first smartphone that worked and so that's you know that's like maybe maybe not infinite but it's sort of definitely an order of magnitude or more of an improvement so I think uh the the technology is designed to give you a massive Delta over over the next the next best thing I think um I think there often are network effects that can kick in that really help the thing that's very um and these these lead to monopolies over time the thing that's very tricky about Network effects is uh they're often uh they're often very hard to get started and so um so even though everyone understands how valuable they are uh there's always this incredible tricky question why is it valuable to the first person who's doing something um economies of scale uh if you have something that with very high fixed costs very low marginal costs uh that's typically a monopoly like business and then um then there's this thing uh of of branding uh which is sort of like just uh this idea that gets lodged in people's brains I I never quite understand how branding Works uh so I never invest in companies where it's just about branding but it is I think a real phenomenon that uh that creates uh that creates real value I think one of the things I'm going to come back to this a little bit towards the end but one of the things that's very striking is that software businesses are often um are for some reason uh very good at some of these things they're especially good at the economies of scale part because the marginal cost of software is zero and so if you get something that works in software um it's often significantly better than the existing solution and then you have these tremendous economies of scale and you can scale fairly quickly so even if the market start small um you can grow your business quickly enough to uh stay um stay at the same size as the growing market and uh and maintain the sort of Monopoly uh Power now the critical thing about these monopolies is um is it's it's not enough to have a monopoly for just a moment the critical thing is to have one that lasts over time um and so you know in Silicon value is the sort of idea that you want to be the first mover and I I always think it's it's in some ways um the better framing is you want to be the last mover you want to be the last company in a category those are the ones that are really valuable Microsoft was the last operating system at least for many decades uh Google is the last search engine Facebook will be valuable if it turns out to be the last social networking site and um and one way to one way to think of this uh last mover uh value is this idea that most to the value in these companies exists far in the future um if you do sort of a discounted cash flow analysis of a business you look at you have sort of all these profit streams you have a growth rate the growth rate's much higher than the discount rate and so most of the value exists far in the future I did I did this exercise at PayPal in March of 2001 we had been in business for about 27 months and um and we sort of had you know the growth rate was 100% a year we were discounting future cash flows by about 30% and it turned out that about 3/4 of the value of the business as of 2001 came from cash flows in years 2011 and Beyond and um and whenever you do the math on any of these tech companies you get to an answer that's something like that so if you are trying to analyze any of the tech companies in Silicon Valley Airbnb Twitter uh Facebook um any emerging internet companies all the ones in y combinator um the math tells you that three qus 80 85% of the value is coming from cash flows in years 2024 and Beyond it's very very far in the future and uh and so one of the things that uh we always overvalue in Silicon Valley is growth rates and we undervalue durability because uh growth is something you can measure in the here and now and you can always track that very precisely um the question of whether a company's still going to be around a decade from now that's actually what what dominates the value equation and that sort of is a much more uh qualitative sort of a thing and so if if we um if we went back to this idea of these characteristics of Monopoly uh proprietary technology Network effects economies of scale um um you can think of these these characteristics as ones that exist at a moment in time where you capture a market and take it over but you also want to think about are these things going to last over time and so there's a Time Dimension to all these characteristics so Network effects often have a great time element where as the network scales the network effects actually get more robust and so if you have a network effect business that's often one that uh um can become a um a bigger and stronger Monopoly over time a proprietary technology is always a little bit of a tricky one so you want something that's an order of magnitude better than uh the state-of-the-art in the world today and that's how you get people's attention that's how you initially break through but then um you don't want to be superseded by somebody else and so there are all these areas of innovation where there was tremendous Innovation but no one made any money so uh you know dis Drive Manufacturing in the 1980s um you could you could do a better dis build a better dis Drive than anybody else you could take over the whole world and two years later someone else would come along and replace yours and in the course of 15 years you got vastly improved disc drives so it had great benefit to Consumers but um it didn't actually help the people who started these compan compies and so there's always this question about having a huge breakthrough in technology but then also being able to say explain why uh yours will be the last breakthrough uh or at least the last breakthrough for a long time or will you make a breakthrough and then you can keep improving on it at a quick enough Pace that no one can ever catch up so if you have a structure of um a structure of the future where there's a lot of innovation and other people will come up with new things in the thing you're working on um that's great for society it's um it's actually not that good for your business typically um and then um economies of scale uh where I talked about so so I think anyway so I think this last mover thing is is very critical I'm always tempted you know I don't want to overdo the chess analogies but you know the first mover in chess is someone who plays white white is about a one-third of a pawn Advantage so there's a small advantage to uh going first you want to be the last mover um who who wins the game and so so was the Kappa Blanca uh world champion uh uh chess champion Kappa Blanca line you must begin by studying the end game and and I do think that's um well I wouldn't say that's the only thing you should study I think this uh the sort of perspective of asking these questions why will this still be the leading company 10 15 20 years from now is a uh is a really critical one to to try to think through let me um let me sort of uh I want to sort of go in two slightly other directions with this Monopoly versus competition idea and I think um so I think this is the the central idea uh in my mind for for business for starting business for thinking about them and U and there are some some very um interesting perspectives I think it gives on the whole you know on the whole history of innovation and technology and science because um you know we we've lived through um we've lived through um you know 250 300 years of incredible technological progress in you know many many different domains uh you know steam engine to Railways to telephones Refrigeration household appliances um you know the computer Revolution Aviation all sorts of different areas of technological innovation and then there's sort of analogous thing that one can say about science where uh we've lived through centuries of of enormous amounts of innovation in in in science as well and um and the the thing that I think um people always miss when they think about these things is um is that um because X and Y are independent variables um some of these things can be extremely valuable Innovations but uh the people who invent them who come up with them do not get rewarded for this and uh and certainly if you go back to um you need to create X dollars in value you capture y% of X I would suggest that the history of science has generally been one where y 0% across the board the scientists never make any money um they're always duded into thinking that they live in a just universe that will reward them for their work and for their inventions and this is probably the fundamental delusion that uh that scientists tend to suffer from in our in our society um and and even in technology there are sort of many different areas of Technology where um where there were great innovations that created tremendous value for society but uh but people did not uh did not actually capture uh that much of the of the value and so I think there is this sort of whole uh history of um science and technology that can be told from the perspective of how much value was actually captured and um and certainly there are entire sectors where people didn't capture anything so you you're the smartest physicist of the 20th century you come up with special relativity you come up with general relativity you don't get to be a billionaire you don't even get to be a millionaire um it just it just somehow doesn't work that way um the railroads incredibly valuable most of them just went bankrupt because it was too much competition um right Brothers um you fly the first plane you don't make any money and so I think there is sort of the structure to these industries that's uh that's very important um and I think the uh the thing that's actually rare are the success cases most the so it's actually you really think about the history in this in this 250 years sweep um it's unus Y is almost always 0% it's always zero in science it's almost always in in technology and so it's very rare where people made money you know the early uh the late uh 18th early 19th century the first Industrial Revolution was the textile mills you had the steam engine you sort of automated things and you had these Relentless improvements that people improved efficiency of textile factories of manufacturing generally at a clip of 5 to 7% every year year after year decade after decade you had 60 70 years of tremendous improvement from 1780 to 1850 um but even in 1850 most of the wealth in Britain was still held by the landed aristocracy uh the workers didn't you know the workers didn't make that much the capitalists didn't make that much either it was all competed away there were hundreds of people running textile factories it was an industry that just uh um the structure of the competition prevented people uh from from making any money um and so I I think there are in my mind there probably are only two broad categories in the entire history of the last 250 years where people have actually uh come up with new things and made money doing so um one is uh these sort of vertically integrated complex monopolies which people uh did build in the Second Industrial Revolution at the end of the 19th and start of the 20th century and so this was like Ford it was the vertically integrated oil companies like Standard Oil um and what these vertically integrated monopolies uh typically required was this very complex coordination you got a lot of pieces to fit together in just the right way uh when you assembled it you had a tremendous Advantage this is actually uh done surprisingly little today and so I think this is sort of a business form that um when people can pull it off is very valuable it's typically fairly Capital intensive uh we live sort of in a in a in a culture where it's very hard to get people to buy into anything that's super complicated and takes very long to build uh but I you know when I sort of think about my colleague Elon Musk from PayPal success with Tesla and SpaceX uh I think the key to these companies was the complex vertically integrated Monopoly structure they had so if you sort of look at Tesla or SpaceX if you ask you know was there sort of a single breakthrough I mean they certainly innovated on a LW of Dimensions I don't think there was a single 10x breakthrough in battery storage or you know maybe working on some things on rocketry but they hadn't there was no sort of single massive breakthrough but what was really impressive was integrating all these pieces together and um and doing it in a way that was more vertically integrated than most of their competitors so Tesla you also integrated The Car Distributors so they wouldn't uh steal all the money as has happened with the rest of the car industry in the US or SpaceX um you basically uh pulled in all the subcontractors um uh where most of the large uh Aerospace companies have single Source subcontractors that are able to sort of charge Monopoly profits and make it very hard for the integrated aerospace companies to make money um and so uh vertical integration I think is sort of a a very underexplored modality of of technological progress that people uh would uh would do well to look at more and then I think there is there is something about software itself that's very very powerful um software has these incredible economies of scale these low margin costs and there is something about the world of bits as opposed to the world of atoms where you can often get very fast adoption and and the fast adoption is critical to capturing and taking over markets because even if you have a small Market if the adoption rate is too slow there'll be enough time for other people to enter that market and compete with you whereas if you have a small to midsized Market and have a fast adoption rate you can uh take over this market and so and so I think this is one of the reasons Silicon Valley has done so well and why software has been of this phenomenal industry and what I what I would suggest what I would want to leave you with is there are sort of these different rationalizations people give for why certain things work and why certain things don't work and I think these rationalizations always obscure this question of um creating X Dollar in value and capturing y% of X so the science rationalization we're always told is that the scientists aren't interested in making money they're doing it for charitable reasons and that not a good scientist if you're motivated by money and I'm not even saying people should always be motivated by money or something like this but I I think we should we should be a little bit more critical of this as a rationalization we should ask is this a rationalization um uh to obscure the fact that y equals 0% and the scientists are operating in this uh in this sort of world where all the uh all the Innovation is effectively competed away and they can't capture any of it directly and then the uh the software Distortion that often happens is because people are making such vast Fortunes in software we infer that this is the most valuable thing um in the world being done full stop and so if people at Twitter make uh billions of dollars it must be that Twitter is worth far more than anything Einstein did um and um and uh and what that sort of rationalization tends to obscure is again that X and Y are independent variables and there are these businesses where you capture a lot of X and there are others where you don't and so uh and so I do think um I do think the history of innovation has been this uh this history where uh the the the the microeconomics the structure of these industries has mattered a tremendous amount and when um and um and and and there is sort of this the story where some people have made vast fortunes because they were in Industries with the right structure and other people uh made uh nothing at all because um because they were in these sort of very competitive things and we shouldn't just rationalize that way I think it's worth understanding this better and then finally let me come back to this this uh this sort of overarching theme for this talk this competition is for losers idea which um is always this provocative way to to title things because we always think of the losers as the people who are not good at competing we think of the losers as the people who are um slow on the sports on the track team in high school or who do a little bit less well on the standardized tests um and don't get into the right schools and so we always think of losers as people who can't compete um and I want us to really rethink and and revalue this and consider whether it's possible that competition itself um is off that we we we're sort of it's not just the case that we don't understand this Monopoly competition dichotomy intellectually so sort of been talking about why why you wouldn't understand it intellectually because um people lie about it it's distorted we have all these uh the history of innovation rationalizes what happen in all these very very strange ways but I think it's more than just an intellectual blind spot I think it's also a psychological blind spot where we find ourselves you know very very attracted to competition in in one form or another um we find it reassuring if other people do things the word ape already in the time of Shakespeare meant both primate and imitate uh and there is something about human nature that's deeply mimetic imitative apik sheeplike leming like cd-like um and it's this very very problem uh thing that we need to always think through and try to overcome and and there is always this question about um competition um as as a form of validation where we we go for things that lots of other people are going for and um it's not that there is wisdom in crowds it's not when lots of people are trying to do something that that's proof of uh it being valuable I think it's when lots of people are trying to do something that is often um that is often proof of insanity there 20,000 people a year who move to Los Angeles to become movie stars about 20 of them make it um I think the Olympics are a little bit better because you have a you know um you can sort of figure out pretty quickly whether you're good or not so it's there's a little bit less of a dead weight loss to society um you know um um you know your your the sort of educational experience at a place uh the the the pre- Stanford educational experience um there's always sort of a non-competitive characterization I think most of the people in this room had machine guns they were competing with people with bows and arrows so um it wasn't exactly a parallel competition when you were in junior high school and high school um there's always a question does the tournament make sense as you keep going and this is uh and so um there is always this question if people go on to grad school or post post-doctoral educations does the intensity of the competition really makes sense there's the uh the you know classic uh Henry Kissinger line that uh um describing his fellow faculty at Harvard that the uh um the battles were so ferocious because the stakes were so small describing sort of Academia and um and and you sort of think on one level this is a description of insanity you know why would people fight like crazy when the stakes are so small but it's also I think simply a function of the logic of the situation when it's imp really hard to differentiate yourself from other people when the differences are when the objective differences really are small then uh you have to uh compete ferociously to maintain uh a difference of one sort uh or another um that's often more imaginary than real there's always sort of a personal uh version of this that I I tell where um you know I was sort of hyper hypert tracked I you know my e8th grade Junior High School yearbook one of my friends wrote in you know I know you'll get into Stanford in four years as a sophomore I sort of went into went to Stanford four years later uh at the end of High School uh went to Stanford Law School uh you know ended up um at a big law firm in uh New York uh where from the outside everybody wanted to get in on the inside everybody wanted to leave um and and you had um and it was this very strange Dynamic where after I uh sort of realized this was maybe not the best idea um and left after seven months and 3 days you know one of the people down the hall from me uh told me um it's really reassuring to see you leave Peter I had no idea that it was possible to escape from alcatra which of course all you had to do was go out the front door and not come back but um but so much of people's identities got wrapped up in um in winning uh these competitions that uh they somehow lost sight of what was important what was valuable uh you know competition does make you better at whatever it is that you're competing on because when you're competing you're um comparing yourself with the people around you you're figuring out how do I beat the people next to me how do I do somewhat better at whatever it is they're doing and you will get better at that thing I'm not I'm not questioning that I'm not denying that but um but it often comes at this tremendous price that uh you stop asking some bigger questions about what's truly important and truly Valu and so I would I would say that don't always go through the tiny little door that everyone's trying to rush through maybe go around the corner and go through the vast gate that no one's taking thank you very much I guess we time for you want to take a few questions or sorry oh yeah people want to take I'll take a few questions we have a few minutes time yeah go ahead um since yeah as you mentioned earlier often monopolies and competition often look similar because the narratives people tell the narratives we tell ourselves do you have any ways to easily determine the difference when you're looking at an idea or evaluating your own idea well I'd say the question I'm I always try to focus on is what is the actual market so not what's The Narrative of the market because you can always tell a fictional story about a market that's much bigger or much smaller but what is the what is the real objective market so it's always yeah you always try to figure it out and you realize people have incentive to distort things yeah so which of the aspects of monop that you mentioned would you say s like Google um well they have uh they have Network effects with the the ad Network they had proprietary technology that gave them the initial lead because they had the the page rank algorithm which was uh sort of an order of magnitude better than any other search search engine you have economies of scale uh because of the need to store you know all these different uh sites and at this point you have brand so Google has all four maybe maybe the proprietary technology is somewhat weaker at this point but definitely it had all four and maybe three and a half out of four now yeah how does this apply to paler and second what's you like second is what what with the iPhone uh head oh this is that's that's a that's a there sort of a set of companies that are doing different copycat payment systems on on mobile phones there's square there's PayPal sort of they have just they just have sort of different shapes that's how they differentiate themselves one is a triangle one is a square um and so you know um maybe at some point the Apes will run out of shapes or something like that but um but I think um no palent here we we started with a focus on on um the um intelligence Community which is small submarket um you had a proprietary technology that used a very very different approach um uh where it was focused on the human um computer uh synthesis rather than the uh uh sub substitution which I think is the dominant Paradigm so there's a whole set of things I would say on the on the market approach and the the proprietary technology uh yes um we have design thinking methodology and uh lean uh startup thinking um which is used to mitigate Risk by not creating things that people don't want but how do young innovators uh have inspiration to create complex systems that last through time can you repeat the question yeah so the question is um what do I think about lean startups uh um iterative thinking where you get uh feedback from people uh versus uh complexity that may not work so I I am personally quite skeptical of all the uh Lean Startup methodology I think the the the really great companies um did something was sort of somewhat more of a Quantum Improvement that really differentiated them from everybody else um they they typically did not do massive you know customer surveys the people who ran these companies uh sometimes not always suffered from mild forms of Asbergers so they we not actually that influenced not that easily deterred by what other people thought or told them to do um so I I do think we're we're way too focused on um iteration as a modality and not enough on trying to um have um you know um a virtual ESP link with the public and figuring it out ourselves um I I would say that uh let me see um I would say that uh the um I I'm not quite the risk question I think is always a very tricky one because there are um you know there there it's it's not it's often I think it's often the case that you don't have enough time to really mitigate risks if you if you're going to take enough time to figure out what people want um you often will have missed the boat by then um and um and then of course there's always the risk of of doing something that's uh that's not that uh significant or meaningful so you know you you could say a track in um in law school is a lowrisk track from one perspective it may still be a very highrisk track in the sense that maybe you not um have a high risk of not doing something meaningful with your life so we have to think about risk in these uh in these very complicated way I think risk is sort of this very uh complicated concept yes you talking about the last move Advantage but then doesn't that imply that there's already competition to begin with chest piece on the chest board um yeah so there's always this uh terminology thing so I I would say that uh there are uh there are categories in which people sort of are bundled together I would say the Monopoly businesses were in in effect they they really were a big first mover in some sense you could say you could say Google was not the first search engine there were other search engines before but on one dimension they were dramatically better than everybody else so they were the first one with page rank with with sort of a automated approach um Facebook was not the first uh social networking site my friend Reed Hoffman started one in 1997 they called it social net so they already had the name social networking uh in the name of their company seven years before Facebook uh their idea was that it was going to be this virtual cyers space where I'd be a dog and you'd be a cat and we'd have all these different rules about how we'd interact with each other in this virtual alternate reality Facebook was the first one to get real identity so it was so I'd say I hope Facebook will be the last social networking site it was the first one in a very important Dimension people often would not think of it as the first because they'd sort of lump all these things together I have one more question okay one more question let's take one here uh if theoretically someone who uh worked at Goldman Sachs out of college and left out 6 months and is now studying computer science at Stanford uh how would you recommend rethinking that um you know I don't I don't have a I don't have a great um I'm not great at the Psychotherapy stuff so I don't I don't quite know how to I don't quite know how to uh how to solve this that there are these um you know there are these very odd stud they've done on people who go to um business school there one they've done at Harvard Business School where um it's sort of the anti- asberg um personality where you have people who are super extroverted uh generally have low convictions uh few ideas and you have sort of a hot house environment you put all these people in for two years and at the end of it uh they systematically end up the largest cohort systematically ends up doing the wrong thing they try to catch the last wave you know uh 1989 everyone at Harvard tried to work for Mike milin it was one or two years before he went to jail for all the junk bond stuff they were never interested in Silicon Valley OR tech except for 99 2000 when they timed the bubble peing perfectly um they did uh and then you know 05 to7 was housing uh private Equity stuff like this so so I do think um I do think this uh tendency for us to see competition as validation is um is very deep um I don't think there's some any sort of easy psychological formula to uh to avoid it so I don't I I don't quite know how to uh what sort of therapy to to recommend but um but my my my first my first starting point which is only like it's maybe 10% of the way is to never underestimate how big a problem it is we always think this is something that afflicts other people so it's easy for me to point to people in Business Schools or people at Harvard or people on Wall Street I think it actually does afflict all of us to a very profound degree we always think of advertising is things that work on other people how who are all these stupid people who fall for All Those ads on TV they obviously work to some extent and they work uh to a disturbing extent on all of us and it's something we we all should work to overcome thank you very much + +## You are not a lottery ticket +Peter Thiel: You Are Not a Lottery Ticket | Interactive 2013 | SXSW - YouTube +https://www.youtube.com/watch?v=iZM_JmZdqCw + +Transcript: +(00:02) I thought I'd sort of take a somewhat different tack today and know try to talk a little bit about what I think the biggest single philosophical question in all these startups maybe even in life generally is that people have and it's um it's whether is it all just a matter of luck or how much of this is luck and how much is not luck when you start one of these businesses and and do do one of these things and it's sort of a big question that underlies a lot of a lot of these different ventures that people +(00:38) do and I want to I want to try to tackle that question at least indirectly today I want to talk a little bit about you know a little bit about the question of luck why it's very hard to answer I want to suggest to you that we live in a society where people are incredibly biased to thinking that things are dominated by luck and I want to at least suggest that there are some alternate ways of thinking about the future that it's it's worth for us to explore um let me let me start with this is probably +(01:12) sort of my my general might the general slide that I always have on the nature of progress the nature of the future and when you think about how the 21st century is going to unfold you can think that there are basically two axes to the 21st century there's a technology axis and a globalization axis they're they're very different people always use these words interchangeably but globalization is basically about copying things that already work it's the story of China and the emerging world there's still you +(01:51) know much the world six billion out of seven billion people are still incredibly poor and what they mainly have to do in the next few decades is just copy things that work there's some things where they can avoid copying bad ideas from from the developed nations but a lot of it is the sort of horizontal or extensive progress and sort of going from 1 to N but then there's also technology doing new things vertical or intensive progress and it's sort of where you're the first person or business or inventor or artist in the +(02:29) history of the world to do to do something new it's going from from zero to one and what I want to suggest is that there's there is something sort of very different about going from zero to one instead of 1 to N 1 to n is there sort of is this law of large numbers and you can sort of run the experiment many times and sort of see how things work but you when you go from zero to one there's a sense in which you know every every sort of event in the history of progress in the history of technology or +(03:04) science has something singular and non-repeatable about it and so and so if we ask you know is a given invention a given startup a given artistic achievement was this something that would have happened anyway is it something was a total fluke it's actually a really hard question to answer because you only have sort of a sample size of one to go on and you can sort of you can sort of say that that with a sample size of one you know the variance becomes infinite so it's almost impossible to know whether you were +(03:41) lucky or not and I think that's probably this sort of important starting point to have is that it's completely unclear whether or not it's luck or not and you can have certain biases on this but in anything we're going from zero to one it's very very hard to say now um there certainly is some very mild anecdotal evidence that that you can give that that's certain that the anti luck argument is that there are certain people who succeeded in doing various businesses on a repeat basis you know +(04:23) there's probably perhaps most famously still achieved Steve Jobs with you know and Pixar you know Jack Dorsey with square Twitter Mike my colleague Elon from PayPal who went on to start SpaceX and Tesla was heavily involved in Solar City of course you know there's always a counter-narrative with these things you can say well this person just each of these people had just one big breakthrough and then everything else was somehow leveraged off of that and so you know whenever you sort of drill down on this question was it luck +(04:54) was it not luck it's actually um it's actually really hard it's really hard to say it is striking however how much the way we talk about this has changed and so if you go you know if you go back in time the classical way people used to talk about it was that luck was something to be overcome or to be mastered so you know thomas jefferson i'm a great believer in luck and i find the harder I work the more I have of it which again suggests that it's it's this it's this thing to to overcome or to you +(05:34) know that's uh that doesn't dominate things or you know even simpler Samuel Goldwyn the harder I work the luckier I get I'm very much in contrast to that you have something like today's dominant view where success stems from sort of this whole context the context is random you know maybe you were a member of the lucky sperm Club the lucky egg club you know you were lucky where you were born and stuff like that and that sort of is what drives everything and there's a course you know a version of this that +(06:09) applies to startups that you know the successful ones were accidental you know it's pretty clear how big a role luck plays I agree with Paul Graham on an awful lot of things but but again I I think this is sort of a place where it's just automatically channeling our default bias as a society and it's worth asking how much of this is true and how much is is not true what I want to focus on today is um is not so much in those are two different directions you can go with this question of luck one is sort of the +(06:48) past orientation which is you know how did I get here you know what were the factors that contributed to to success but I want to focus instead on instead I want to focus on the future question of of what you know what you can do next and where you go where you go from here is this is there sort of is is and is the future is that something that's fundamentally dominated by chance or is somehow thinking of this is being dominant by chance somewhat of a somewhat of a wrong way or an incomplete way to think about the future and want +(07:34) to offer the alternate and alternate perspective on that now I want to say a little bit about the structure of the future and how people can think about the future and and you can think of it I think most simply as being determinate or indeterminate and in particular I have the sort of two-by-two matrix for the future it's sort of a consulting type type idea but you can basically on most simplistically you can say that the future is either optimistic or pessimistic and it's either determinate or indeterminate an optimistic future is +(08:15) one where you believe things will be better than the present on whatever axis defines better a pessimistic future is one where you think things will generally be worse and then a determinant future is one where you can sort of map out reasonable amounts of it and and plan against it and then an indeterminate one is a is where you have absolutely no clue and it's or just a random walk all the way certainly depending on which of these quadrants you believe is basically correct out the future it tells you some very +(08:56) different at least some very different approaches that you would pursue in terms of how you think about your life or the kinds of things you're doing and I want to sort of try to develop this framework a little bit more as we as we as we think about this so for example most simply on a determinant versus indeterminate axis you know if you believe the future is determinate you will act with with some degree of conviction you'll have specific ideas and you'll have some confidence to to to work towards those ideas if it is +(09:39) indeterminate on the number one rule is to diversify because you have no idea what's going to work and you should just try lots of different things and should have some sort of a portfolio approach to the future and so I think one of the things that's always very interesting when you think of this determinate versus indeterminate thing is that all these things that ultimately become self-fulfilling so if you if you think it's determinate and then you focus on doing one thing extremely well that sort +(10:07) of leads to conviction and then maybe that becomes self-fulfilling if you think that it's fundamentally indeterminate you end up with a portfolio approach it's diversified and it has maybe that itself becomes self-fulfilling and becomes somewhat indeterminate and it becomes more indeterminate in a way um there's of course on you know on the optimistic versus pessimistic axis the simple one is just you know are you fundamentally afraid of it do you think it's fundamentally something that's +(10:37) going to be better and these sorts of perspectives also lead to very different ways to act if you want to if you want to put this in sort of a historical context I think that I'm going to try to develop explain why I put these different zones in these quadrants but I think the u.s. +(10:57) in the 50s and 60s and maybe even before then was fundamentally optimistic and determinate the future was clearly going to be better people thought that every generation was going to be better off the generation came before and it was for the most part in very specific ways there was a there was a determinate way in which the frontier was going to be developed there was a way that you know cars would get faster planes would get faster rockets would get faster there was sort of all these very specific ways that the future would get better from year to year and +(11:31) decade to decade I think is sort of a very different paradigm that the US had for a quarter century from 1982 to 2007 where the future was going to be better that was the official religion it was still sort of very optimistic but if you asked how or why people had no good answer for it and so it was just this this mechanistic thing that would automatically get better in one form another and so we were sort of in this quadrant of extreme indeterminate optimism I would say for about a quarter of a century from the period of 82 to +(12:10) 2007 an indeterminate pessimism is probably what characterizes most of the rest of the developed world I would say Japan's been in this zone for for the last the last 20 years people have a sense the future is going to be not that great maybe a little bit worse and nobody has an idea of what to do and I think sort of Europe has weirdly drifted into this indeterminate pessimistic quadrant as well where people think the future is worse but nobody has any idea what to do um you can sort of argue where you put China on this some people +(12:48) would put in the determinate optimistic quadrant I tend to put it on the determinant pessimistic quadrant it's very determinate people in China have a they know what it's going to look like in 20 years they're going to build out the highways in the cities and for the most part it's it I think it is going to be a somewhat poorer version of the developed world people will get old before they get rich there's a very specific way to to track against that but China comes out on this quadrant that's very different from the you +(13:21) 80222 2000-2007 if you if you think of these quadrants in terms of in terms of the sort of a financial way you could describe this which is that if you're optimistic you don't need to save a lot of money if you're pessimistic you save a lot of money so if you're super optimistic you know the future is going to be better there's no need to save any money and so you end up with a savings rate that's that's very low if you include a government borrowing in the u.s. today the u.s. savings rate today +(14:00) stands at minus 6% so we're still incredibly optimistic about the future we don't need to save any money because the future will automatically be better and so we're still maybe in an Indian summer of indeterminate optimism on the other hand for example I would describe China's quite pessimistic because it has a savings rate of something close to on 30 30 percent of GDP gets saved and so people even though there's some ways in which things are getting better from year to year people still think they'll +(14:31) be old before they get rich and therefore they need to save a lot of money so you have this low savings to high savings access from optimism to pessimism and then if you sort of that's like investing in cash or bonds or things like that saving in cash or bonds or things like that if you invest in specific things there's specific company's buildings ideas you invest in if you have a definite determinate view you'll have a high investment rate if it's indeterminate you will not know what specific things to invest in and so you +(15:08) end up with a low low investment on low investment rate and so so one of the strange things about indeterminate optimism is that it's the quadrant that has both low savings and low investment and the question I will come back to towards the end is whether that's a state that's whether that's a stable quadrant at all is it possible for the future to be better when no one saves and no one invests because no one's thinking and everyone's outsourcing all the thinking to other people one of the other ways you can +(15:43) sort of describe this difference on I want to give a few different axes for describing this shift from determinate to indeterminate ways of the future the mathematical version is that the dominant form of mathematics used to be calculus and it's shifted to probability and statistics the structural way is that in a determinant world you're focused on substance there are some specific substantive things you focus on in an indeterminant world all that you end up focusing on are processes and so what people talk about our world is +(16:17) what's the process for doing thing something much more than what is the specific thing you're trying to do to talk about specific things is to definite and that seems to two to weirdly to weirdly wrong in practice you can sort of think of these very different types of quadrants that dominate the indeterminate optimistic world is dominated by finance and law because these are the kinds of process oriented disciplines that that you pursue if you have no idea about the future if it's fundamentally about +(16:50) making sure that the piping of the system it works but you have the sense the system just sort of works automatically on in a determinate optimistic world that's probably a world in which there's much more room for engineering art you know very specific things it's people who have ideas about the future that are radically different from the present it's people who have dreams about the future that nobody else shares and that they're you know willing to work towards and where the dreams of +(17:23) a substantive Lee different and radically better world are not subsumed towards some sort of random process you know determinate pessimistic would be wartime rationing you know indeterminate pessimistic all you end up doing is buying insurance I've spoken in some context about to being a bubble in education and I think you can think of the education bubble a form of indeterminate pessimism where people are they don't really know what the education is for but it's fundamentally acts as an insurance +(17:56) policy to avoid falling through the larger and larger cracks in our society and so anyway you can sort of think about these different quadrants and depending on which one dominates there's sort of very different kinds of industries that that end up dominating you know to have a picture of what definite optimism or determinate optimism looked like you know we can we can go back to these classic examples from from US history so there are things like like the construction of the Transcontinental Railroad in the in the +(18:32) 19th century uh where it was a radically different future where the world you know would be connected it's a gigantic undertaking by today's standards nobody would do it people would say why are you building this railroad to nowhere it doesn't make any sense it's costing too much money but somehow as it was built the future sort of took care of itself you know the classic mid 20th century example I always like to cite is Robert Moses who was this some somewhat forgotten figure he was probably the +(19:10) most important person in in New York more powerful than the mayor or the governor of New York City New York State respectively for about a quarter century he started by becoming the parks commissioner he ended up having seventeen different positions in government and he sort of show up with his bulldozer and build parks and levels and neighborhoods and build some highways people could have access to the parks he built all the sort of roadways on Long Island the FDR expressway and there was sort of you know one thing +(19:40) after another that got rebuilt it sort of stopped in the mid 1960s when he planned to build a highway connecting Brooklyn in New Jersey it was going to go through the southern end of Manhattan they're going to sort of raise Greenwich Village to the ground the neighbors sort of objected they started saying we're not quite sure that the future actually is be better than the present there's gonna be this highway with sort of skyscrapers right on top of the highway and and it's sort of it sort of stopped and at that +(20:10) point people basically you know what and once there was no longer a definite view of the future you also ended up with a place where people stopped building things in New York and so it's possible that Moses was very wrong that most of the things were misdirected but it did have this sort of powerful coordinating function and and once the idea of the future disappeared and people no longer believed that there was a future that was that looked very different from the present and that was radically better than the present people +(20:43) stopped being able to build anything and nothing new an infrastructure sense has been built in New York State for close to half a century on any meaningful scale again to sort of illustrate how different past ideas of the future were for those of those those of you familiar with the San Francisco Bay Area there was this thing called the rubber plan up in sort of the late 1940s it was a plan to basically build two large earth and Rockville dams one was between Marin County and Richmond the other between San Francisco and Oakland it was going +(21:22) to turn on the North Bay in the South Bay into two giant freshwater lakes there'd be 20,000 acres of new landfill there's going to be a 32 lane highway built around the entire Bay Area and you have sort of 30 story skyscrapers built around the entire San Francisco Bay you know rubber was sort of a schoolteacher amateur theater producer but in the world of the 1950s this idea was taken seriously enough that you had congressional hearings there was you know a giant mock-up of the plan it was concluded there would be too much of +(21:58) operation and wouldn't quite work and so people sort of gave up on it after a while but again you cannot even imagine someone someone who's a schoolteacher and amateur theater producer being able to have a plan to just re-engineer a huge geographical area like this and and change and change the world in a in a radical way this is again sort of where we're in this world that's extremely different from the world of just 50 or 60 years ago and of course you know all the classic all the classic versions of +(22:38) science fiction cities underwater cities on the moon cities on Mars cities in outer space sort of radically different and very definite ideas of the future that would sort of become self-fulfilling prophecies of one sort or another and when you sort of look at these pictures today these things look they don't look futuristic they look dated they look like they look like they really are sort of from the past which is sort of a against for this very strange way in which things have changed um you know the classic by contrast the +(23:17) classic version of indefinite optimism is a portfolio investing theory it is that you should invest in a in a stock market index that's the way you get the highest risk adjusted returns because the motion of stocks is like the movement of atoms in the universe it's fundamentally random and we can't know anything about it we can just actually we can maybe describe the laws the statistical laws that describe the randomness but what specific stocks or specific companies or specific projects you should invest in you can never know +(23:54) and but you know the stock market generally moves in a in a northeasterly direction and therefore the most important thing is to find the way to get maximum diversification at minimum cost and and do something like a portfolio investing you know in this ship one of the strange things that happens in the shift from definite to indefinite views of the future is that there is this shift from engineering to finance and and and one of the things that happens is that money somehow becomes much more important and so the soundbite +(24:28) version of this is in a definite world money is a means to an end because there are specific things you want to do with money in an indefinite world you have no idea what to do with money and some money simply becomes an end in itself which seems always a little bit perverse you just accumulate money and you have no idea what to do with it that seems for like a bit of a crazy thing to do but I think that's actually what what happens a great deal and so to illustrate one way that this flow might happen if you you know start a +(25:00) successful business you know you sell the company or you sell shares to investors in an IPO you make some money question what do you do with the money you have no idea because nobody knows what to do with anything and so you give the money to a large bank to help you do something what does the bank do it has no idea so it gives the money to a portfolio of institutional investors what do the instance each institutional investor do they have no idea and so they all just invest in a portfolio of stocks not too much in any single stock +(25:36) ever because that suggests you have opinions or you have ideas and that's very dangerous because it suggests that you're somehow not with it and then what do the companies do that get the money they've been told that all they should do is generate free cash flows because if they were to actually invest the money in specific things that would suggest the companies had ideas about the future and that would be very dangerous and so one of the worst things you can ever have is a company that's uh +(26:02) that's not profitable in in in this indefinite world and sort of the the contrarian idea that I always like saying is that we always like investing in companies that are losing money we don't like investing in companies that are making money because the companies that are not profitable are actually the companies that have a lot of ideas about what to do with their money whereas a company that's massively profitable on some level is a company that's out of ideas and it's especially crazy in a world +(26:31) where on where the interest rates are zero and you actually get paid less and less on the money and then of course the companies are profitable to generate cash flows the cash flows eventually go back to people and you sort of cycle and repeat and this is sort of this is the rough flow that happens in the world of indefinite optimism on the problem is you know it's somewhat of a you know this is a bit of an extreme picture but in effect it's a hermetically closed loop and at the end of the day no one's +(27:03) doing anything real with the money it's completely abstracted and what ends up happening is there are fewer and fewer things you can do and and one of the one of the sort of financial ways to illustrate this is if you look at the real interest rates on a 10-year bonds in the US which is this is a measure of how much interest you earn on bonds - what the expected inflation rate is and it's basically been trending steadily downwards today it's at minus 0. +(27:37) 6 percent so 10-year bonds are yielding about 2% the expected inflation for the next decade is 2.6 percent so when you invest in bonds in real terms you're expecting to lose minus 0.6 percent a year for a decade and it shouldn't even be surprising because there's no one in the system who has any idea what to do with the money this has been sort of a consistent critique of the huge deficits the US is running people constantly are saying you know there's a point where the bond market is going to blow up and +(28:09) the interest rates will go higher and one of the really big mysteries is that this has not happened for year after year and I think the the fundamental the fundamental reason this has not happened is that people actually have no idea what to do with the money the last big idea people had on what to do with money that was not sort of circular was to buy houses and to invest in housing and that was sort of the decade the idea of the last decade that idea has gone out of fashion now that people no longer want to buy houses they +(28:41) have absolutely no ideas what to do with money the interest rates the real interest rates are going steadily more negative and so there's some sense in which this system of indefinite optimism is on is gradually sort of running out there's a way in which um you know what the the natural drift is for something from finance to insurance I attend to all I'm not going to give my whole anti Warren Buffett lecture here but I think I think there's a way in which Buffett was prescient and ahead of the curve and +(29:19) basically we oriented most of his businesses towards insurance companies in disguise which is sort of the world of indefinite pessimism and that's what what dominates in that sort of world and we can sort of see how this indeterminacy affects us in very very many different fields so if we look at politics in an indeterminate world the most important position in politics is the pollster and what do you do in politics nobody has a clue but what you do is you take a poll and and the polls tell you what to do they don't really +(29:56) tell you anything on a long-term basis but they sort of tell you incrementally what you do at any any given time and and as we've tracked towards us more and more indeterminate world there's a way in which you know Paul taking has become more and more dominant and so the way we talk about political campaigns and elections is sort of how are people doing in the polls much more than what ideas they're talking it's sort of you know if Martin Luther King were here and said I have a dream about a future +(30:28) that's really different you know the question with how does that poll um it's a it would never and and that's sort of the way we avoid this there obviously are cases where this goes very badly wrong my sort of Exhibit A and my apology to all the Palin fans and the audience here but but but you know people always say it's this incredible mystery why McCain picked Palin in 2008 I think it's not a mystery at all on the basic heuristic was that you you looked at all the Republican senators and governors in 2008 +(31:01) Republicans weren't very popular they're mostly running around 40% in the polls Palin was at 85 in Alaska because when oil was $140 a barrel Alaska was like Kuwait everybody was getting huge checks from the government at state of Alaska she was pulling extremely well and so there was no question that you would go with the person who pulled the highest and so you went with you went with Palin even though perhaps it couldn't scale to the US as a whole because the US was not producing way more oil than it was +(31:33) consuming and so you know and then there's sort of our ways in which you know the 2012 re-election gets recast as a contract contest between say nate silver and dick morris silver was a better poll taker and so he won rather than that obama had better ideas or the ideas resonated better or that's what people wanted to hear it's not the substantive way that we talk about politics um to talk about government more generally you know even though government spending is still about the same as it has been for the last 50 or +(32:06) 60 years as a percent of GDP more and more of it is shifted towards transfer payments which are of course a way of saying that the government has no ideas and what to do with money it simply moves it to other people and people it's assumed that people have ideas but you don't have specific ideas of what to do there is a way in which you can see indeterminacy in literature sort of if you sort of take a some you know classic science fiction the 1960s Space Odyssey classic you know the text was updated +(32:38) automatically every hour one could spend entire lifetime doing nothing but X orbing the ever-changing flow of information from a new satellite so a specific definite view of a radically different future which you know maps reasonably closely to today's internet Neuromancer 1984 the sky above the port was the color of television tuned to a dead Channel future is fundamentally static if you think of it in terms of philosophy there's sort of our different ways to to map these quadrants but I would say the optimistic indeterminate +(33:17) quadrant is fundamentally the quadrant of someone like Rawls or Nozick people only think of them as opposites but they're they're really both in the same category Nozick it's sort of the libertarian version you have no idea what to do and therefore the individual is paramount Rawls you're an avail of ignorance you have no idea what society you will be born in and therefore social democracy is paramount you get two very different ideas but they both take their starting point from sort of complete +(33:48) indeterminacy about the state of the world the classic determinate one you could be on the left you could be on the right but it was there was some sort of sort of Marx Hegel all these people where there was you know even going back to bacon or Locke there was some sort of definite way that things would would get better and you could sort of work towards a better future in one way or another and then of course you have the determinate pessimistic indeterminate pessimistic you think tend to be more classical ones the determinant +(34:22) pessimistic would be like Plato and Aristotle there's a definite way things happen even though there's sort of limits and you can't be too hopeful about technology you're really improving the state of the world in any fundamental way and then probably the classic version of indeterminate pessimism and I think this is sort of the philosophical category that we're back in is sort of the epicurean Lucretia's view of the universe where there is nothing but atoms and the void the atoms randomly move throughout the +(34:50) void they run sometimes bump into each other stuff happens eventually they break apart things fall apart chance roads everything there's nothing specifically you can do that makes any sense everything event you know things sort of all go to pot marijuana farming might be a good thing to go into in this world but basically this is actually this is actually in some ways become the dominant the dominant view and it ends up being sort of strangely stoic because there's really nothing you can do about these larger forces that that will +(35:28) ultimately buffett you in the the most you can hope to achieve the certain amount of equanimity and indifference about fundamental randomness and meaninglessness of the universe you know indeterminacy and death we think about the process of aging and death very differently from the way we used to sort of the classic early science version was that death was a specific problem to be solved there were specific diseases to solve specific ways to conquer death on the contemporary way to think about it is fundamentally through the prism of +(36:06) insurance which is again sort of indeterminate more on the pessimistic side and the main thing we can do is the actuarial math you know what is the probability that you are going to die in a given year if you are that old so if you're 30 years old you have about a 1,000 chance of dying in the next year if you're 80 years old you have a 1 in 10 chance of dying in the next year and and all we can do is sort of describe these these probabilities and I'm not saying that you know by the way that this probabilistic view has nothing to +(36:38) it I just want to sort of illustrate how it sort of dominates our thinking and so instead of trying to find a cure for death or a solution to the problem of aging the best we can hope to do is figure out better ways to calculate the probabilities better ways to give people cell people various life insurance policies and and thereby suggest the sort of pseudo mastery of a process that's fundamentally random and indeterminate of course you know the basic problem is that eventually the luck runs out oops you see how this works +(39:26) so anyway um you know at some point your luck runs out and so I think I think this question of whether you know indeterminate optimism is possible in the long run is this is this core question and is something like this this frame that we've had where it's just one coin toss after another and you know you're probably not going to die in the next hour or the next six months you'll probably be lucky it will probably land heads for the time being but you never think very far about the future is that +(40:06) actually an ultimately stable quadrant or not and so so you know if you want to sort of frame this as a general question it's you could say you know could an iterative process lead if not the best of all possible worlds at least to a world where there's a path of monotonic and potentially never-ending improvement that's sort of a the core idea of this of this world of indeterminate or indefinite optimism on you know the paradigmatic Pro an indeterminate optimist example is Darwin in Darwin's +(40:39) theory of evolution where you know you basically over billions of years develop this proliferation of different life forms and that's sort of the that's the paradigmatic example that we apply to all these different fields and we think that things like that work now I do think there are some paradigmatic counter examples the one that I want to sort of highlight is the paradigmatic counter example of failed in determinism are is failed cities and and you can sort of you can certainly give various examples there's you know Los Angeles +(41:16) which probably should have been the greatest city in the United States with you know fantastic weather and somehow you know everything gradually went wrong with the sprawl LA is still one of the better places in the world you know there's the example of South Paulo I was there for a day about a year ago it's about a 12 mile drive from the main downtown area to the airport it takes about 35 minutes if there's no traffic for hours if there's traffic on the nine lane highways sort of take the helicopter at dusk which is sort of +(41:48) endless lines of red lit cars backed up bumper-to-bumper on it's sort of 30 million people living in enclaves of a quarter to half a million each and of course how Paulo is still you know vastly better than places like Mumbai or Lagos Nigeria or places like that and if you had to sort of give a single reason why the convergence theory of globalization is probably going to fail is that most cities in the developing world looks something like this and and they will not actually be improved in any incremental step-by-step +(42:27) way sorry there's there's no actual incremental way to to improve something something like this and you know if you look at the populations most of the people the world at this point are are living in these sorts of places the the sort of policy debates we have at this point are things like economics versus environmentalism which we we describe as these radically different perspectives on the world but they're really just differing views of different indefinite futures as a sort of maybe economics can +(43:03) be a little bit more optimistic environmentalism is a little bit more pessimistic I personally think that as long as that's the frame environmentalism will always win because indefinite you know optimism is unstable and the ecological critique of economic thinking is that is that the economist says we don't need to think about the environment because people will solve problems every step of the way and then the counter-argument is this this probably does not work but of course the the issue with both is that somehow +(43:36) you're subject to these these much much larger forces it's the market its nature they're fundamentally a sort of random unknowable statistical and and you can't think about them coherently one way or another and of course this kind of approach is also sort of very endemic to the way people think about when they start businesses just to sort of segue back to that and what one sees most commonly are sort of this this methodology where you have no idea what you're doing but it should basically be +(44:13) a never-ending form of a be testing on Darwinistic a be testing might work if you have billions of years but in practice you tend to run out of money well before then and the problem is that somehow the search space of all businesses is much bigger than the search space of great businesses so it's somehow the a/b testing I think generally is a somewhat too inefficient process and the sort of iteration process but you end up with this question you just end up on some very low-hanging hill where the iteration is +(44:49) going to do something that incrementally improves things at every step in time but you maybe you just end up you know in a slightly better part of an infinitely large slum or something like that to use the failed city example there's a machine learning and there's sort of all these different ways where you do not think about the future it's most character strongly characterized I think in a way by the very short time horizons and I think one of the things that's true both on the startup side and +(45:17) that may be even more true on the side of people who invest in startups is that anything that takes longer than a year is considered unreal and fake because we can have no opinions about the future and so everything has to be on a super short time horizon and and and we're sort of dominated by by people who do these kinds of things and there's a question you know how well how well does this actually work and this is this is this is the official religion that we have today and I'm not you know I'm I +(45:54) don't believe in the official religion on but I think you know I think I just want I want people just to at least be aware that is this is the religion that that it's all statistical it's all luck driven and I do think the the most striking thing is that the most successful businesses in some ways don't quite seem to fit this pattern even though you know we we end up talking about it and so you know the you know the you know after the 2007 crash we have seen this return to technology but it was characterized by +(46:35) it was most characterized by businesses that had very definite ideas I think the the iconic one for for the last number of years at least till jobs passed away was Apple which was of course very much the opposite of you know an indeterminate business Apple was one where there was a multi-year vision of the future that was sort of being executed against there are legitimate concerns whether it still has that vision now that jobs is is no longer there but that's really what what is going on and of course that's not really +(47:10) the way we typically talk about it we typically talk about it and we look at you know how mean Jobs was to all the employees and and so we you know they're people I know who are running businesses and they sort of hand out they hound out hand out books describing how bad jobs was to his employees to make their employees feel better about themselves and and so and and I think the the real question you need to ask about something like Apple is you know why did the people put up with this bad behavior and I think it was because it was actually +(47:43) this very countervailing narrative of the future that in a determinate world you know one of the most important metrics is the robustness of the plan what I often call the secret plan that you're working against on you know the the kinds of companies that we've looked at it founders spend over the years the ones that have done best and want been ones that have somehow tracked against this this very very long-term vision of the future and I think sort of the one the one closing thought I would have on +(48:14) this is that on the one most characteristic thing of companies with a plan is that they do not sell and it there often are times that you should sell businesses you know I started PayPal there were reasons that we there were reasons that it was the right decision for us to sell the business when we did in 2002 to eBay but but I do think that the most successful businesses somehow have an idea of the future that's uh that's very different from the present it's not fully valued and and therefore there's +(48:52) there's no point at at which you you should sell I've told this anecdote before but the the most important moment in my mind in the history of Facebook occurred in July of 2006 the company had been around for about two years at the time was still just a college site there may be eight or nine million people on the site the revenues were I think tracking to about thirty million no profits and we got we received an acquisition offer from Yahoo for a billion dollars and so we had the board meeting on Monday morning on in July +(49:35) 2006 or three of us on the board Zuckerberg myself Jim Brier and and you know full disclosure I think that both Breyer and myself on balance thought we probably should take the money and run but but you know Zuckerberg started the meeting and first thing he said was well you know it's kind of a formality we have to have a quick board meeting shouldn't take more than 10 minutes you know we're obviously not going to sell here and then we say well you know which actually talked about this a little bit +(50:10) more you know a billion dollars is a lot of money and we in some ways we hashed the entire the entire discussion we had today here these are like you know you know you own like 25 percent the company there's so much you could do with all the money you'd make well I don't really know what I do with the money I just start another social networking site but I kind of like the one I already have so why should I sell and it was in some way sort of an encapsulation of this of this entire of this entire +(50:44) discussion and of course some you know you never sort of know that the immediate aftermath was that there was you know an almost you know infinite number of not infinite but there was sort of a large number of you know sort of stories about how you know how in the world could you have a CEO who was so crazy that he wouldn't sell the company who didn't know that you should take a billion dollars this is what you got when you had some who's twenty two years old when you didn't have any adults in +(51:16) the company on you know it probably you know it was just the worst decision anybody had ever made the best best rash you know I was a little bit worried about it we had you know the the sort of Founders Fund ideology we had was that we should always back the CEO we should always back the founder and so we went we we just went with with that as a framework but I think sort of the one you know the one partial rationalization I was able to come up with for not taking the money was that I you know we looked at the history in Yahoo it there +(51:49) had been two other companies where they'd offered a billion dollars that had been turned down it was eBay and Google and and so I concluded that give at least you could actually make a pseudo-scientific argument that that that every case was someone had been offered a billion dollars and had rejected it it had been the correct thing to do but I do think this is a but the you know the argument that Zuckerberg finally you know finally finally came came down on was that you know there were all these new things +(52:27) that we were going to build at Facebook and it was clear that Yahoo wasn't valuing any of the products that had not yet been released and he wanted to have a chance to build those products and since he was pretty confident that Yahoo had and this is not anti Yahoo things I think would be true of almost all these companies that they had no definite idea about the future and therefore did not properly value things that did not yet exist they were therefore undervaluing undervaluing the business and I think this is sort of +(53:02) this is sort of the challenge all of us have is to to work towards the future that's not just static like a dead Channel on television but that's a that's a definite future and that is a radically better future that can that can motivate and coordinate and inspire a number of people to change the world and to go to a world where we where where luck is something for us to overcome and to deal with as we as we go along the way but but not for something that becomes this absolute dominating force that that stops all thought +(53:46) thought before it even starts thank you very much and I'm not going to say best of luck thank you you diff --git a/Pmarca Blog Archives.pdf "b/investor/\360\237\227\243\357\270\217belief/Pmarca Blog Archives.pdf" similarity index 100% rename from Pmarca Blog Archives.pdf rename to "investor/\360\237\227\243\357\270\217belief/Pmarca Blog Archives.pdf" diff --git "a/investor/\360\237\244\234act/MA_deal.xlsx" "b/investor/\360\237\244\234act/MA_deal.xlsx" new file mode 100644 index 0000000..0532972 Binary files /dev/null and "b/investor/\360\237\244\234act/MA_deal.xlsx" differ diff --git a/logistic_model.stan b/logistic_model.stan new file mode 100644 index 0000000..ba5bf8f --- /dev/null +++ b/logistic_model.stan @@ -0,0 +1,26 @@ + +data { + int N; // number of observations + int K; // number of predictors + matrix[N, K] X; // predictor matrix + int y[N]; // binary outcome +} + +parameters { + vector[K] beta; // coefficients +} + +model { + beta ~ normal(0, 5); // prior + y ~ bernoulli_logit(X * beta); // likelihood +} + +generated quantities { + vector[N] log_lik; + for (n in 1:N) + log_lik[n] = bernoulli_logit_lpmf(y[n] | X[n] * beta); + + vector[K] odds_ratio; + for (k in 1:K) + odds_ratio[k] = exp(beta[k]); +} diff --git a/outcome_proportions.png b/outcome_proportions.png new file mode 100644 index 0000000..d16f3cf Binary files /dev/null and b/outcome_proportions.png differ diff --git a/stan/two_step_model_with_latent.stan b/stan/two_step_model_with_latent.stan new file mode 100644 index 0000000..7526293 --- /dev/null +++ b/stan/two_step_model_with_latent.stan @@ -0,0 +1,38 @@ +data { + int N; // number of data rows + vector[N] exec_score; // "founder execution state" + vector[N] idea_score; // "idea quality state" + int invest[N]; // action of investing +} + +parameters { + real alpha0; + real alpha_exec; + real alpha_idea; + + // Latent states + int state[N]; // K latent states + real mu_alpha_exec[K]; + real mu_alpha_idea[K]; + real sigma_alpha_exec; + real sigma_alpha_idea; +} + +model { + // Priors + alpha0 ~ normal(0, 2); + + mu_alpha_exec ~ normal(0, 2); + mu_alpha_idea ~ normal(0, 2); + sigma_alpha_exec ~ cauchy(0, 2); + sigma_alpha_idea ~ cauchy(0, 2); + + for (n in 1:N) { + // Hierarchical priors for latent states + alpha_exec = mu_alpha_exec[state[n]] + sigma_alpha_exec * normal_rng(0, 1); + alpha_idea = mu_alpha_idea[state[n]] + sigma_alpha_idea * normal_rng(0, 1); + + // Logistic regression + invest[n] ~ bernoulli_logit(alpha0 + alpha_exec * exec_score[n] + alpha_idea * idea_score[n]); + } +} diff --git a/vc-regression-test.ipynb b/vc-regression-test.ipynb new file mode 100644 index 0000000..e69de29