Letter from America – Weathering the storm

Letter from America, Angus Deaton

For the RES Newsletter, April 2012, Online Issue 2

The American economy is limping out of the Great Recession; new jobs are being created, but unemployment is falling slowly, if at all, and it will be many years before the unemployment rate returns to the five per cent level of less than four years ago. The suffering that the recession has brought is little reported in the media. While there is endless speculation about the effects of unemployment and recession on the election prospects of President Obama and his opponent, and while the merits of stimulus and further austerity (firing people to create jobs) are debated, there is little discussion of what the recession is actually doing to people. Apparently, the main effects of interest are those on the chances of politicians in the election.

Happiness and unemployment

One of the achievements of the ‘happiness’ literature has been to show that being unemployed is very bad for people’s well-being. Most people have always known this, though economists have tended to resist, if only because the standard textbook model assumes that leisure is good and that the harm of unemployment resides only in the fact that it reduces income, a harm that is at least partially offset by the leisure that comes with not having to go to work. Yet when people are asked how they are doing, either in terms of an overall assessment of their lives, or by reporting emotional distress, those who are unemployed are much worse off than can be accounted for by their loss of earnings. Jobs mean more than income, and the loss of a job brings a loss of structure and meaning and it is most likely this that brings the loss in well-being. The American data show this, and the size of the effect is similar to those consistently found in Europe.

Given this, we might expect that average well-being would have been badly hit by the recession. But this didn’t happen, or if it did, it happened at the wrong time. Gallup asks a random sample of 1,000 Americans every evening about how their life is going and about their emotional experiences on the previous day; as the unemployment rate rose from five to ten per cent in late 2008 and early 2009, the measures moved only a little. Indeed, overall wellbeing was rising quite rapidly in the late spring of 2009, as the unemployment rate headed to its peak. Should we think that unemployment isn’t so bad after all?

Perhaps the happiness measures are just not very useful, and many economists remain sceptical of them. Indeed, there is some evidence to support scepticism. People are often not sure how to respond to questions about how their lives are going, so that their judgments are seriously affected by the questions that come just before. If people are reminded that they think the US is going in the wrong direction, then they will also report that their own lives are doing badly. When they are not reminded, there is no effect, even when they believe that the country is going wrong. The reminding causes people to reinterpret the wellbeing question. What is happening in the stock market also seems to matter, even for people who have no direct or indirect interest; perhaps it is peculiarly American to see life through the stock-market. Yet, even if the self-reports are taken seriously, it is too much to expect to see large effects in the national averages. After all, even if the effects of unemployment on well-being are large for any individual who is unfortunate enough to become unemployed, ‘only’ one in twenty workers actually lost their jobs during the run up, and about half that fraction in the adult population as a whole. The effect on the average is then less than a twentieth of the effect on the individual, which is small enough to be difficult to see. We might reasonably conclude that happiness measures, however useful in the cross-section, are not so well-suited for tracking over time or for monitoring the state of the economy. Alternatively, they might be doing just fine in elucidating an important truth: sure, unemployment is high, and many are suffering, but only one in twenty workers has lost their job, so why should the rest of us care? It is certainly consistent with my reading of how much the general population seems to care.

The effects on health...

The recession has certainly affected people in many other ways, some of which are a good deal less than obvious. Epidemiologists tend to take it for granted that recessions are bad for health. Richer people have better health than poorer people, and there are many studies on the adverse health effects of losing a job. While it is too soon to assess the effects of the Great Recession, there are certainly reasons to be concerned. The states, who are responsible for much of healthcare for the poor and for many children and elderly, have suffered severe revenue shortfalls, are constrained by balanced budget laws, and are cutting everything that it is possible to cut. Yet, research has repeatedly shown that overall mortality rates in the US are contra-cyclical, that recessions are good for health. Recent work by Ann Huff Stevens, Doug Miller, Marianne Page and Mateusz Filipski argues that the cyclical fluctuations in mortality are largely among the elderly, especially elderly women living in nursing homes or assisted living, whose survival is better when the facilities are better staffed, which happens during recessions.

In the Gallup data people were much more likely to report worry and stress during the period when unemployment was rising rapidly. Worry and stress are unpleasant and may have knock on effects in other aspects of people's lives. My colleague Eldar Shafir, working with Sendhil Mullainathan at Harvard, suggests that poverty and the associated stress may actually compromise cognitive function, because of the overload of having to focus so hard on making ends meet. This effect makes it harder to cope, and can compound economic difficulties by compromising judgment and decision making. There is also some evidence that people turned to religion, perhaps seeking a sanctuary against the difficulties in their lives. During the black days around the Lehman Brothers collapse, there was a noticeable uptick in the fraction who reported that religion was important in their lives. This should not be dismissed as superstition but as an effective defence mechanism.

...and on income

The effects of the recession can be seen even at the very top of the income distribution, among the famous top one percent and beyond. According to tax data assembled and reported by Emanuel Saez, the top one per cent saw a reduction of more than a third in their real incomes from 2007 to 2009, compared with half of that for the population as a whole. But as the recovery began in 2009 to 2010, top incomes resumed their faster growth at five times the rate of the average. Which brings us back to the election. Two years ago, a Supreme Court decision authorized unlimited spending by private individuals on behalf of candidates, and this funding has already reshaped the Republican primaries. It remains to be seen whether the general election will be more heavily swayed by the unemployed or by those at the very top, by the many or by the few.

Page Options