Learning from Errors in Entrepreneurship

I use administrative and survey-based micro data to study the relationship between expectation errors, belief updates, and subsequent corporate decisions of a representative sample of French entrepreneurs. After overestimating their development and hiring prospects, optimistic entrepreneurs update downward, entrepreneurs who underestimate their prospects update upward. Although optimistic and pessimistic types are persistent over time, I show that expectation errors decline over time within individuals, suggesting that entrepreneurs learn from their past errors over time. Making errors and learning from them have real effects. The ability to correctly forecast growth and update after an error lead to better start-ups' performance.