The Engine of Capitalism

Core idea

Most people believe they are better drivers than average. Most people believe their projects will succeed at higher rates than similar past projects. Most people who start businesses believe they have a 70%+ chance of success, while the actual success rate for new businesses is roughly 35% over five years. The evidence of optimism bias is robust, cross-cultural, and domain-general.

But optimism bias is not simply a cognitive error to be corrected. Kahneman argues that it may be the psychological engine of capitalism: without the overconfidence of entrepreneurs who consistently underestimate the probability of failure and overestimate the probability of success, many ventures that have positive expected social value would never be started. The individual pays the full cost of the failure; society captures some of the gain from the successes. This collective overconfidence is economically functional even when individually harmful.

Why it matters

The ubiquity and scale of optimism bias

A survey of small business owners found that 81% rated their chance of success as 7 out of 10 or better; 33% rated their chance as 10 out of 10 (certain). The actual survival rate at five years is around 35%. The gap between subjective confidence (averaging around 7/10) and actuarial probability (3.5/10) is large, consistent, and robust to information: knowing the base rate does not eliminate the gap.

The same pattern appears in mergers and acquisitions (executives consistently overestimate synergies), government projects (The Outside View), major capital investments, and competitive entry decisions (companies enter markets whose capacity is already sufficient to eliminate profit for all entrants).

Focused thinking and competition neglect

One mechanism for optimism bias is what Kahneman calls competition neglect: planners focus on their own capabilities and plans while ignoring that competitors are making similar plans simultaneously. When asked to estimate market share, people generate a scenario of their own product's capabilities and infer share from there — without asking "how many similar products will compete, and why would ours succeed against them?"

This is WYSIATI applied to planning: the inside view constructs a scenario from what is available (the current project, the team, the market opportunity) while suppressing what is absent (the competition's plans, the base rate of similar entries, the historical distribution of outcomes).

The dual role of optimism

Kahneman is careful to distinguish individual-level harm from social-level function. For any individual entrepreneur, optimism bias leads to investments they would not make with accurate probability estimates. Many of these investments result in loss. But the aggregate effect of all entrepreneurs acting on overconfident probability estimates may include some ventures that produce large positive externalities — products, jobs, technologies — that the individual entrepreneur could not have captured fully anyway. The social value of entrepreneurial risk-taking partly depends on the individual being wrong about probability.

This does not make optimism bias "good" — it makes it complex. For individual decision-making, the prescription remains: use the outside view and correct for optimism. For understanding economic dynamism, optimism bias may be part of the explanation.

Key takeaways

Mental model

Copy sourceZoom

Practical application

Managing optimism bias organizationally:

• Separate the decision from the motivation: use reference class forecasting for probability estimates (removing bias), while allowing optimistic effort and energy to drive execution. These can coexist.
• Require explicit competition analysis: before any market entry or product launch, require an explicit analysis of competitive plans, not just current competitor position.
• Build in de-biasing at the proposal stage, not the post-mortem stage: by the time a project is running, sunk costs and momentum make it very difficult to revise estimates. De-biasing must happen before commitment.

Example

Three companies in the same industry all decide to build competing digital platforms, each estimating they will capture 40% of the market. They are entering a market with one large incumbent. Collectively, they are assuming 120% market share is available to new entrants — ignoring each other. Each company has competent teams and reasonable business cases, constructed from inside views. Competition neglect means none of them factored in that the 40% share must come from the same pool that all three are targeting, plus the incumbent.

One platform succeeds; two fail. From the social perspective, the industry has a new competitive option. From the individual perspective, two companies consumed significant capital and talent on a venture whose competitive probability was far lower than the inside view suggested.

Related lessons