Goodhart's Law

Back in 1975 professor Charles Goodhart stated:

Goodhart's Law

Any observed statistical regularity will tend to collapse once pressure is placed upon it for control purposes

More in general Goodhart's Law is Goodhart's law is a generalized social science expression of the so called 'Lucas Critique'. Named after Robert Lucas who argues that it is naive to try to predict the effects of an (economic) change entirely on the basis of relationships observed in historical data.

Actuarial examples...
Some remarkable actuarial examples of Goodhart's Law are:

• Using coverage ratios to control pension funds
• Using z-scores to control banks or pension funds
• (Under)estimating longevity risk for decades
• Measuring (long term) rates (stocks, bonds, inflation ) for valuation purposes

Historians...
Early 2009 Goodhart proclaimed : “One of the lessons of the recent crisis, a lesson for bankers and for regulators, is, hire fewer mathematicians (actuaries) and physicists who build models on the basis of data that they can observe over relatively short period, and hire a few more historians who know what can go wrong even if they don’t necessarily have a good data basis to put into particular models”.

Although Goodhart is probably right, actuaries should keep an eye on the process....


Future developments are a function of Data, Probability, Uncertainty, Experience and above all Common Sense.

Just like the Roman God Janus, actuaries should look at the future, with the past in mind. And to do just that, we need all the help in the world, especially from historians....


Related links:
- Performance Persistence of Dutch Pension Funds (2010)