What the economic theory of sunk costs tells us about plan continuation bias
Plan continuation bias is a recognised and subtle cognitive bias that tends to force the continuation of an existing plan or course of action even in the face of changing conditions. In the field of aerospace it has been recognised as a significant causal factor in accidents, with a 2004 NASA study finding that in 9 out of the 19 accidents studied aircrew exhibited this behavioural bias. One explanation of this behaviour may be a version of the well known ‘sunk cost‘ economic heuristic.
Economists argue that sunk costs (what we have spent in the past) should never be used when making rational decisions about future events. So for example if I’ve bought a ticket to the ballet, I should not decide to go on the basis that I have spent money on the ticket. Instead the decision should be based on what my current circumstances are, e.g. ‘I’ve just learned that my favourite ballerina is not dancing, therefore I won’t go’, rather than ‘I have spent all this money and I don’t want to waste it…’.
Of course as the example above illustrates, and behavioural economists point out, this is not exactly how human beings really behave. In practice people tend to apply the so called sunk cost heuristic to a greater or lesser degree. In summary the heuristic reflects that human beings are averse to loss so we tend to bias our decision in a way that demonstrates our reluctance to write of any sunk cost as a loss.
There are also two predominant factors that characterise the heuristic. The first is an over optimistic estimate of probability of success, possibly to reduce cognitive dissonance having made a decision. The second is that of personal responsibility, when you are personally accountable it is difficult to admit you were wrong.
So it’s not hard to see that the first factor could well be at work in skewing a pilot’s risk perception and biasing the subsequent decision to ‘press on’ in a worsening situation. Having made that first decision, the likelihood of subsequent success would be viewed more optimistically. Which may explain why the NASA Ames study of the 19 aviation accidents concluded that most could have been avoided by executing a missed approach, and also why the report also concluded that aircrew seemed to find it difficult to do just that.
And of course pilot’s are responsible to their company for minimising passenger delays and operating costs which again makes it more difficult having committed, to admit that the course of action was wrong based on an uncertain future. Given that the consequences of a divert would have been certain, immediate and embarrassingly apparent to colleagues, passengers and company.
For me the sunk cost heuristic provides an elegant and powerful explanation for how rational and well trained air crew could have their risk perception of a worsening situation progressively skewed by the series of incremental decisions that led them to that point in the first place.