Measuring Well-being?

The negative economic impacts of the recent crisis are well known. In the US, for example, unemployment rose from 5% to 10%, a quarter of households lost more than half of their assets between 2007 and 2009, the stock market collapsed, and wages fell.

But what about happiness and well-being? 


A large new data set promises to make it possible to investigate. Since January 2008, Gallup Poll has surveyed 1,000 Americans a day, and the questions cover emotional health and life evaluation. You can have fun here looking at how states vary by how stressed/happy they are!


Using this data, Angus Deaton (Princeton University) is studying how the crisis affected reported measures of well-being, and he presented his findings earlier this week at St Catherine's College, Oxford (unfortunately I don't have the slides). He showed us that, as you would expect, the subjective well-being measures significantly and substantially suffered when Lehman Brothers collapsed and during the US election, and rose at the time of Obama's inauguration. 


The introduction of well-being questions into the Gallup Poll is part of a broader concern that traditional measures of economic health, like Gross Domestic Product, are not relevant to assess well-being, and other factors like the environment too. The recent commission on 'the measurement of economic performance and social progress', set up by Monsieur Sarkozy, brought these ideas to the world stage, and in the UK the Office of National Statistics is introducing well-being questions into household surveys.

There are problems, however, trying to measure happiness and well-being directly, and Deaton faced some of these in his analysis. One issue is that 'hedonic' measures (concerning happiness, stress and worry for example) quickly revert after a negative impact. Our ability to adapt is a great coping mechanism, but it puts into question how informative these measures are of event and policy impacts on welfare. 

The survey design also proved very important. It's not just the way questions are asked, but the order in which they are asked, that affects people's answers. Deaton found one particular inexplicable date on which there was a large jump up in reported well-being. It turned out that from that day questions on well-being were not asked straight after the questions concerning politics (another issue Gallup Poll is interested in). This shows just how sensitive these measures are to irrelevant factors. (And that politics makes us unhappy!)

Deaton ended by showing that the stock market turns out to be one of the best indicators of subjective well-being, which is intriguing and perhaps worrying given that the these measures are trying to measure happiness directly, as distinct from economic measures.