April 2008 newsletter – Letter from America – The passions, the politics… and the data

In his latest letter, Angus Deaton comments on the findings of the Gallup/World Poll of people in 150 countries. These show a remarkably consistent link between per capita GDP and various measures of well-being. However, US citizens it seems do worse on happiness and sadness measures than their incomes would predict and they are astonishingly dissatisfied with their healthcare system — even more so than the British.

Politics is constantly being shaped by new events and new data. And politics and politicians spin the data, impugn its accuracy, or highlight it, as circumstances demand. At a deeper level, there are links that run from politics and events back to the construction and the quality of data. When Newt Gingrich was Speaker of the House of Representatives, he threatened to ‘zero out’ the Bureau of Labor Statistics if it did not ‘improve’ the consumer price index to better reflect supposedly unmeasured quality improvements, and to stop over-compensating the elderly. Lawsuits buzz around the US decennial census like flies around road-kill. The US Constitution requires that political apportionment be adjusted to the results of the census, and because non-response is non-random with respect to party affiliation, even the most arcane statistical adjustment procedures become bones of political contention. There are also links that are more unusual and less predictable.

Jim Clifton, who is the CEO of the Gallup Organization, recalls a conversation with then Secretary of Defense Donald Rumsfeld, shortly after 9/11, in which they concluded that they really didn’t know why the attacks had taken place. They concluded that a major reason for their lack of understanding was that there was no Gallup Poll for the Middle East, or other Muslim areas of the world. This story reveals much about the dependence of decision making in Washington on polls but that is not my point here. Clifton, regularly hailed as one of America’s more visionary CEO’s, committed Gallup to fill the gap, and instituted a unique data gathering exercise, all the more so by being undertaken by a profit-making firm, not by the World Bank or the United Nations. This is the World Poll, whose aim, literally, is to construct a continuing random sample of all the people of the world. The core questionnaire, which takes about twenty-five minutes to administer, contains a wide range of questions, not only about attitudes, but also about income, employment, health, and well-being. The first wave was in 2006, and covered random national samples of 150,000 people from almost 150 countries. Although not all countries in the world were covered, the World Poll reached many places for which such data are not often collected, including not only China, Myanmar, Afghanistan, Iraq, Iran, and Cuba, but a number of the poorest countries in the world, including Mali, Togo, Chad, Zimbabwe, and Sierra Leone. The Poll missed out on Sudan in the first wave, but covered it in the second round in 2007. Gallup has always prided itself on its links with academia, especially with psychologists, and my Princeton colleague Danny Kahneman has been heavily involved in the World Poll from the beginning, and Clifton has also brought in a number of economists including (and in the interests of disclosure) the writer of this letter.

The World Poll reveals much about the world, about the US, and about the relationship between them. In line with Kahneman’s interest in measures of well-being, the World Poll asks questions on life-satisfaction-measured on an eleven point ladder — from the worst (0) to the best (10) possible life for you — as well as questions on affect that allow a more direct assessment of happiness and sadness. Across the world as a whole, there is a remarkably close relationship between individual and average national life satisfaction and per capita GDP; each doubling of income brings with it four-fifths of a rung on the ladder, a relationship that holds among rich countries as well as poor. Yet, as is the case for the relationship between life-expectancy and GDP, Americans are less satisfied with their lives than would be predicted by their incomes. They are much less satisfied than the Danes or the Finns, and almost as dissatisfied as the British. The US comes out even worse on the happiness and sadness measures. People in richer countries enjoy themselves more, they smile and laugh more often, and they are less prone to depression and sadness, but Americans do worse than is warranted by their incomes. (The British are not easily characterized: they smile and enjoy themselves more than all but the inhabitants of six other countries in the world — Laos, New Zealand, Ireland, Holland, and Costa Rica — but they are sad and depressed about as much as are Americans.)

The most stunning findings, however, are related neither to life satisfaction nor to affect, but to anything to do with health. Americans are less likely to be satisfied with their personal health than the citizens of almost any other rich country; Japan, in spite of its extraordinarily high life-expectancy, is the exception. Americans’ life-expectancy is also low by the standards of the rich world, but not low enough to justify their health perception. But if Americans are dissatisfied with their health, they are astonishingly dissatisfied with their medical and healthcare system, which is the most expensive in the world, absorbing sixteen percent of GDP, more than twice the corresponding figure for the UK. Yet only a half of Americans have confidence in their health and medical system; even in Britain, which recorded one of the lowest figures in Europe, two-thirds of people trust their system. The US not only lags behind Europe, but it ranks 88th out of the 120 countries for which these data were collected in the first wave. The US is not only behind such traditional good performers as Cuba, Vietnam, or Malaysia, but Americans trust their healthcare system less than do Indians, Iranians, Afghans, or Angolans. These findings almost certainly do not reflect a distrust of the personal medical care that Americans actually receive once they are in the system. There is no immediate prospect of health tourism from the US to Afghanistan, Iran, or Angola. Rather it is that Americans lack confidence in their healthcare system, and they worry about the increasing fraction of the population that is uninsured, about whether those of us who are insured will be allowed the treatment that our doctors recommend, about the extent to which insurers will actually pay, and about insurance arrangements in a future where costly medical innovation makes today's arrangements difficult to sustain. Which brings us back to the politics, where health and health insurance are at the forefront, not only of the debate between Republicans (cost-control and market incentives, and an apparent denial of adverse selection) and Democrats (universal insurance), but also the current debate between Senators Clinton and Obama (whether and how insurance should be ‘mandated’, given that a universal single payer system is seen as politically impossible in the US.) 

What about people in the Muslim countries? It turns out they are a lot like Americans; they want the same freedoms and the same democracy that Americans want, and the vast majority reject violence and extremism. They also believe that democracy should be built around their religion, a position that is far from unknown in the US. Yet nearly all of the inhabitants of Muslim countries have more confidence in their healthcare systems than do Americans.

Related resources