Wednesday, December 30, 2009

How far is too far: The sunk cost fallacy

"We've come too far and spent too much to turn back now."

That's the RDC's argument-of-last-resort for finishing Beale Street Landing. In the previous post, I tried to put a percentage on the descriptor "too far." One City Council person suggested that over 50% is "too far".

I believe he is not alone. Most people think of the halfway mark as a threshold -- since it is the point at which you are getting closer to the end than to the beginning.

In my calculation we are actually just reaching the 30% point, provided the steelwork contract hasn't been signed. But is 30% still "too far to turn back?" For many people, yes it is -- but not because they are thinking rationally.

It's a classic problem that is discussed in every business school and it has a name: The "sunk cost dilemma." "Sunk costs" are those that have already been incurred and are not recoverable.


Large pile of sunk costs at the foot of Beale Street, May 23, 2009. Photo by Michael Cromer.


In traditional micro-economic theory, you should not consider sunk costs when you make a decision about a future course of action. They are losses already incurred and have no relevance to the future. To make a strictly rational decision, you should only consider the prospective costs of your various options.

The question should not be "How much have we spent? ($11 million.)"

The question should only be whether spending $26 million more makes sense, given what we know today.

But people have emotional biases. One common bias is people's reluctance to accept the reality of their losses. People would rather proceed (even with a low probability of success) than to quit and have to acknowledge a 100% loss.

Therefore, people do consider sunk costs and allow them to influence their decision-making -- even though that kind of thinking is fallacious. The sunk cost fallacy results in too many "zombie" projects (and wars) that never seem to end.

To be continued...

No comments: