Evolutionary strategies


Jason Collins


February 21, 2012

In Tim Harford’s discussion in Adapt of the benefits to experimentation , Harford notes that experimentation by individuals is often at great potential cost. As when a species evolves, what appears to be beneficial experimentation at a societal level involves frequent failure to survive by individuals. While Harford suggests that people should consider experimenting in a manner that avoids failure that threatens survival, the reality is that many people take risks with a large downside.

Harford’s discussion reminded me of a section in Daniel Kahneman’s Thinking, Fast and Slow, where Kahneman notes the economic benefits of optimism. Entrepreneurs often engage in “optimistic” behaviour, in that they vastly overestimate the probability of business success. Kahneman writes:

The chances that a small business will survive for five years in the United States are about 35%. But the individuals who open such businesses do not believe that the statistics apply to them. A survey found that American entrepreneurs tend to believe they are in a promising line of business: their average estimate of the chances of success for “any business like yours” was 60% — almost double the true value. The bias was more glaring when people assessed the odds of their own venture. Fully 81% of the entrepreneurs put their personal odds of success at 7 out of 10 or higher, and 33% said their chance of failing was zero. …

The optimistic risk taking of entrepreneurs surely contributes to the economic dynamism of a capitalistic society, even if most risk takers end up disappointed.

They do not consider their conduct to be risk seeking, as the odds they have calculated suggest they believe they are taking a reasonable bet. Kahneman notes that we need optimistic people to run these experiments, as they are responsible for the economic dynamism in a capitalist society.

The penchant for entrepreneurial activity in people suggests that such optimistic and apparently risk-seeking behaviour may have evolutionarily foundations. Take the historical case where members of a clan left for new territories with significant potential for death. Those who departed were usually on the fringe of the group or faced severe resource constraints. As a result, even if a move to a new frontier was likely to end badly, the low benefits to staying and the potential upside of departing make the departure worthwhile. Or consider the evolutionary bonanza for the early settlers to eastern Canada and the north-eastern United States, who often experienced order of magnitude population increases within five generations of arrival. Many moved from being fringe members of societies they left to comprising significant portions of the new populations.

This brings me to a post by Bryan Caplan on the benefits of meekness. Caplan suggests those at the bottom of the social ladder are “dysfunctionally assertive” and would benefit from being meeker. Caplan states that advice to be assertive often comes from those with power who can afford to be assertive. When someone with low-status stands-up, it may be disastrous.

If we ignore whether there is evidence of the lower classes being more assertive, it raises the question of whether assertive behaviour of low-status people is, on average, costly. There may be a low expected monetary pay-off to an assertive act as they lose jobs or other privileges, but what are the costs and benefits in dimensions that matter? Is the person balancing a meek and certain passage into genetic oblivion, versus a risky shot at reproductive success? We should not derive the optimal strategy by a simple cost-benefit analysis in monetary terms. When facing the end of the genetic line, strategies with an expected negative monetary and social pay-off, but high variance, may be the preferred path.

This issue is similar to a paper I posted about concerning risk-seeking behaviour by those without a mate. A utility function measured in terms of mates leads to significantly different behavioural predictions than one in financial terms.