How the industry exploits our poor grasp of probabilities
FT columnist Professor John Kay has recently written two columns devoted to the notion that people have a poor grasp of probabilities. The first prompted several letters debating a particular reasoning problem: the Monty Hall Show contest. In his second column on probabilities Prof. Kay points to these letters as proof people really are confused.
I pick from this a much more serious point he makes, which is that understanding probability theory and knowing the odds gives you a money-making advantage over those that do not. This about sums up the problem individuals have in their dealings with the financial services industry. Information advantage is exploited both by professionals at the expense of individuals and by smart professionals at the expense of dumb ones. In both cases the retail customer pays. It is unrealistic for most customers to outsmart professionals so some form of proxy is needed to test when professionals understand the odds and use them to help you, not exploit you.
Information advantage is exploited by product manufacturers when they understand the distribution of probable payoffs better than their distributors or direct customers. Precipice bonds and zeros are examples of the non-trivial probability of catastrophic loss being understood by professionals somewhere in the chain without this being communicated to customers. It can be difficult to know if it is cock-up or conspiracy. For instance, setting endowment insurance premiums at a level providing about a 50% chance of achieving the capital sums required could have been due to industry-wide errors in modeling market and inflation uncertainties or to forcing the investor’s cash outflows to be no higher than a repayment mortgage so as to make them saleable.
Information advantage is exploited by advisers when their commission-based business format makes it profitable to conceal true probabilities. In the same example, commissions on a mortgage endowment were about six times higher than on repayment mortgages. Very few financial advisers have the statistical training or technical means to calculate probabilities but that is hardly surprising if they have no economic incentive to share them with customers.
Prof Kay is right to suggest that probabilistic models may not be perfect but when relied on sensibly are a lot better than nothing. The difference is important enough to serve as a proxy for individuals: what technical resources and language to look for when selecting advisers and what questions to ask about products.
Equally important is that when agents cost you money through their ignorance of probabilities, their arrogance in thinking they knew how the dice would fall or through downright misrepresentation of risk, you treat any of these as a compelling reason to stop dealing with them.
This idea that customers of the retail industry need to be streetwise rather than knowledgable (with practical hints about how to be streetwise) is a theme of my book and is summarised in the Introduction.