The Planning Fallacy: Why Everything Takes Longer Than You Think

The Sydney Opera House was supposed to take six years to build and cost seven million dollars. It took fourteen years and cost 102 million dollars. The Scottish Parliament building was budgeted at forty million pounds and came in at over four hundred million. The Denver International Airport, the Chunnel, Boston's Big Dig — the same story, over and over, at every scale.

This is not just a problem with large public projects. It is a fundamental feature of human cognition, documented at every scale from individual tasks to national infrastructure: we consistently underestimate how long things will take and how much they will cost.

Kahneman and Tversky named it the planning fallacy in 1979. The mechanism is related to optimism bias — we imagine things going roughly to plan, discounting the inevitable obstacles, delays, and unexpected complications that experience should have taught us to expect.

When asked to plan a task, people focus on the specific details of that task — what needs to happen, in what order — rather than on the base rate of how similar tasks have gone historically. This is what Kahneman calls the inside view. The outside view asks: how long have comparable projects actually taken? The outside view is almost always more accurate. But it requires stepping back from the particular task and thinking statistically about classes of similar tasks, which does not come naturally.

There is also a social dimension. People presenting plans are often incentivised to make optimistic projections — to win contracts, secure funding, or avoid the discomfort of saying this will be harder and longer than you want to hear. This produces systematically misleading forecasts at the organizational level.

Reference class forecasting is the most effective correction. Before estimating how long your project will take, look at how long similar projects have actually taken historically. Take that distribution seriously. Adjust from the outside view rather than the inside view. Add a buffer that your inside view tells you is too pessimistic, because your inside view is not calibrated.

The uncomfortable truth is that this correction feels wrong even when you know about the planning fallacy. Optimism is a feature. It gets projects started that would never begin under realistic projections. The question is whether the optimism is worth its costs.