Tuesday, July 1, 2008

What does happiness research tell us about the trade-off between inflation and unemployment?

The best place to begin is with the misery index. What is the misery index? No, it isn’t a measure of the average happiness level of economists. The misery index is the sum of the inflation rate and the unemployment rate. The concept was apparently created by Arthur Okun in the 1960s (see here) but it wasn’t the best idea he ever had. (Arguably, Okun’s best idea was the ‘leaky bucket’ metaphor. Redistribution of income from the rich to the poor is like carrying water in a leaky bucket. Some of the contents of the bucket is lost in transit. More information can be found here.)

The problem with the misery index is that it assumes that a percentage point increase in inflation creates just as much misery as a percentage point increase in unemployment. That assumption might make some sense if we had no information about the relative amounts of misery caused by inflation and unemployment, but that is not the case.

The results of happiness research suggest that unemployment has large negative effects on satisfaction with life. Research suggests that an increase in annual income of over $40,000 (over and above unemployment benefits) would be required to give Australian males who are unemployed the same probability of experiencing high life satisfaction as someone who is employed. The corresponding income increase for women was estimated to be about twice as large (for reasons unknown). Larger required income differences have also been estimated for other countries (see Nick Carroll, ‘Unemployment and psychological well-being’, Economic Record, Sept. 2007).

Happiness research suggests that although inflation has significant negative effects on satisfaction with life these effects are smaller than the effects on unemployment. Using a large European data base, Justin Wolfers has estimated that a percentage point of unemployment causes about 5 times more unhappiness than a percentage point of inflation (‘Is business cycle volatility costly’, International Finance, 2003, here).

If we can assume that a similar ratio applies in Australia, would this mean that the Reserve Bank would have to be crazy to contemplate policy action that would reduce inflation by a percentage point if this was likely to result in an increase in unemployment of 2 percentage points? No. In order to do this calculation properly it is necessary to take into account the duration of these expected effects of policy actions. If the reduction in inflation is expected to be permanent and the increase in unemployment is expected to last for only one year, the policy action would make sense. A clever politician might even be able to sell such an outcome to the public as “the recession we had to have”.

The obvious point is that if the Reserve Bank is able to demonstrate its resolve to prevent inflationary expectations from taking hold it will be able to avoid resorting to the unpalatable option of inducing a recession in order to reduce inflation. Unfortunately, the Bank is under pressure from some economists to view recent increases in inflation as having been “imported” and hence to take no action against them. I don’t know how anyone can argue that inflation can be imported in a country that has a floating exchange rate - but that may not mean much because I still count myself among those who think that “inflation is always and everywhere a monetary phenomenon”.

No comments: