Rational Relationships: The Sunk Cost Fallacy
"We must do whatever it takes to justify what we've already done" - Stephen Colbert
The sunk cost fallacy is one of my favorite concepts. I first encountered it in business school. In the business word, "sunk costs" are any past costs that cannot be regained. The sunk costs fallacy describes an occurrence whereby managers will overvalue a project based on the sunk costs invested in it, rather than the prospective future gains.
A simple example is buying a stock. If you pay $1000 for ten shares of stock, that $1000 is a sunk cost. The current value of the stock is independent of the price that you paid for it. The only way to accurately value the stock is to determine its price at the current moment or to attempt to estimate its future value. However, if the market will currently pay only $900 for your ten shares, then there is psychological pressure on you to continue to think of your stock as worth $1000, even if its actual value is less than that.
The sunk cost fallacy comes from people's natural tendency toward loss aversion. For almost everyone, the pain of receiving something and then losing it is greater than never having it in the first place. In other words, we tend to feel losses more strongly than we feel corresponding gains. Because of this, we tend to try to avoid losses more than we try to pursue gains. The sunk cost fallacy is a result of loss aversion, because we tend not to see sunk costs as "losses" until we dispose of the object paid for. When you buy those shares of stock, you have lost $1000, but it doesn't feel like you've lost $1000 because you've simultaneously gained a stock that's valued at $1000. If the price of the stock drops to $900, you've lost the equivalent of $100, but it doesn't feel that way unless you sell the stock. If you sell the stock for $900, you've gained $900, but it feels like a loss of $100. This is the feeling that enables the sunk cost fallacy.
This is a problem for business people. Let's say my company decides to research a new widget, incurs substantial R&D costs, and tasks me with the decision whether to introduce the widget into the market. There is a temptation to see the large amount of R&D put into the widget and conclude that the widget will be profitable and we should launch it. However, the amount of R&D spent has no bearing on whether there is sufficient demand for the widget. If I actually relied on the sunk costs to make my decision, it would be a disaster. Profitability depends on demand, production costs, overhead, advertising, and a whole host of other factors, none of which are sunk costs. Sunk costs are completely irrelevant to the value of the widget.
The sunk cost fallacy is that little voice that encourages us to finish the book once we've read half of it and decided that we don't like it. It's what keeps us driving the wrong direction rather than turn around (literally and figuratively). It's what keep us using the fancy $150 universal remote long after it's apparent that the cheap $10 remote is more user-friendly and useful.
It's also what keeps us in bad relationships. People change. Often, those changes will result in formerly good partners no longer being good matches for each other. In those circumstances, it's best that couples break up or transition to some other form of relationship. It is often the case, though, that couples will look at their history and conclude that too much time, effort, and energy has been invested in the relationship to end it.
This is a mistake. There are certainly plenty of reasons why long-standing partners might not want to break up. Their experience with each other may show them that they are only in a temporary rough patch. Their lives may be so entangled that leaving the relationship would be incredibly painful. Their issues may just not be as bad as they seem.
But it's a mistake to think that the amount of investment in a relationship automatically adds value to that relationship. It doesn't. The value of the relationship consists of what is happening in the present and in the future. The past is done. The past is useful in predicting the future, but the past by itself doesn't actually add any value. The length of a relationship or the amount of effort put into a relationship doesn't actually add value. If it's clear that a relationship won't serve you in the future, your previous investment in the relationship won't change that.
Rationally evaluating our relationships requires acknowledging that their value is derived only from our reasonable future expectations. Sunk costs are unrecoverable.
Copyright 2014 by Wesley Fenza