PsyFi Search


Friday, 25 July 2014

Z is for Zero Risk Bias

Zero-risk bias is a preference for options that completely eliminate some risk even where alternative, often cheaper, options will reduce the overall risk by more, proportionately.We often prefer the absolute certainty of a smaller benefit to a larger benefit of less certainty.


Let's rewind to 2007/2008 when the world, for investors, had become a very uncertain place. Although, to be frank, it had always been very uncertain, it's just that investors, being myopic creatures, hadn't realized it. Anyway, suddenly the world was full of uncertainty and investors wanted to eliminate it all, at almost any price. The result was a stampede into government bonds, driving bond yields to multi-generational lows, whilst ignoring the astonishingly low prices of some of the world's greatest corporations.

Of course, any corporation comes with some residual risk, but in the final analysis if the world's great firms go bust it's unlikely that the tax revenues that sustain government debt will be far behind. But back then such was the value ascribed to T-bills people were prepared to pay governments (once you take inflation into account) to hold their cash.


We often get muddled about the differences between quantities and proportionality - a rise of 100% in knife crime sounds terrible until you realize that this means it's gone from one attack per year to two attacks. It also seems we prefer large decreases in small risks to small decreases in large ones, even when the overall benefit of the latter is vastly superior to the former: zero risk bias is an extreme form of this behavior, often triggered under conditions of uncertainty.

In fact there are circumstances where zero risk is worth paying for - if you're a householder the elimination of any possibility you might lose your home has a very high value compared to nearly eliminating the risk. But generally, in investment, zero risk equates to very low returns.


Zero risk bias can be reduced by re-framing the problem, which suggests that it's not immutable but is a consequence of the way we think about situations. Focusing on the other side of the equation can help: investing in T-bills back in 2008 would have reduced your chances of losing any money to zero but would also have decreased your chances of making any to about the same amount. By and large it's best to re-frame in terms of absolute quantities rather than relative proportions if possible.

Zero risk bias added to the Big List of Behavioral Biases.

1 comment:

  1. Just like to say that this has been an excellent series of articles. Many thanks.