The Great Escape: Health, Wealth, and the Origins of Inequality by Angus Deaton (Princeton University Press)

"Life is better now than at almost any time in history. More people are richer and fewer people live in dire poverty.” So begins the economist Angus Deaton’s study of changing levels of well-being. But he immediately continues: “Yet millions still experience the horrors of destitution and of premature death. The world is hugely unequal.” The aim of The Great Escape is to tell a twofold story: the “escape” of many from what we now see as premature deaths, and the persistent inequalities, both within and between nations, that prevent this happy “escape” from being a reality for many of the world’s people.

The book begins in a touching but somewhat misleading way. Deaton’s father, he tells us, “escaped” from a dismal life in a Yorkshire mining village, where he never got more than a few years of schooling. Eventually, through opportunities opened up by military service in World War II, he studied at night and became qualified as a civil engineer. Through effort and determination he then ensured that his son would get a first-class education. By now, sending his two children to Princeton, where he also teaches, Deaton himself has made sure that his children received an education that he judges to be “vastly superior in its depth, range of opportunities, and quality of teaching” to his own narrow education at Cambridge. Onward and upward!

Deaton tells us that his father’s story is “an example of what this book is about.” But it isn’t: it has nothing to say about changing health and sanitation conditions, which is what most of the book is really about. Even more important, it is a classic story of success through personal effort (with a little help from the army), and it says nothing about the role played by government services, which is where Deaton will argue that much of the responsibility for the “escape” rests. Indeed, those who achieve success through personal effort often conclude from their success that government services are a useless crutch, and that poverty is caused by laziness and stupidity. Deaton knows much better, and the whole point of his book is to insist that institutions and government action matter; so I wish he had resisted the temptation to begin with what is a charming digression. I wish he had also scrapped the all-too-catchy title, alluding to the 1963 film about the escape of Allied prisoners of war from a German POW camp during World War II. Comparing malnutrition and infectious diseases to the Third Reich suggests a grotesque exoneration of the worst in human behavior. And even though Deaton will ultimately impute causal importance to many human actions (medical discoveries, government action and inaction), it is surely not helpful to compare this complicated range of mainly well-intentioned activities to Hitler’s hideous projects.

Once we get past these initial missteps, Deaton proves an engaging and sure-footed guide to the “endless dance between progress and inequality” that is humanity’s history, particularly in recent years. Income and health have improved almost everywhere since World War II; there is not a single country where infant and child mortality is not lower than it was in 1950. And yet inequality is famously skyrocketing even within the richer countries. Even there, many people are denied “escape” from premature death, bad education, and lack of political voice. And inequality between nations is also rising. Although some very populous developing countries, particularly China and India, are making great strides, thus tilting the global average in a rosy direction, other countries appear to be falling even further behind.


To understand these problems and what can be done about them, we must begin by understanding what has already happened. But before we can do that, we must understand what we are talking about. Unfortunately, The Great Escape contains no detailed analysis of the concept of human welfare and no extensive confrontation with rival normative conceptions of welfare and development. But the reader who patiently searches for evidence of Deaton’s views will find a lot of it, together with some compelling arguments against opposing conceptions.

Basically, Deaton’s view is that welfare, or the good human life, has many component parts, each one valuable in its own right. The good life is much more than money: it includes health, education, freedom from discrimination and oppression (including “not to be the victim of others’ search for enrichment”!), and the ability to participate in democratic society on a basis of equality. Even that list of components is too simple, given that each component is itself plural. Health, for example, has many dimensions, including physical robustness, cognitive well-being, and life expectancy. Deaton is at his best when he insists that measures of well-being rest on ethical judgments, even in an area as apparently simple as health. Accentuating life expectancy, for example, has the effect of prioritizing mortality decline among the very young. Meanwhile a focus on height, used judiciously, can supply a part of the picture of how people are really faring in the prime of their lives, since height, affected by maternal and child nutrition, is strongly correlated with both bodily robustness and cognitive capacity.

Thus Deaton is strongly, and repeatedly, critical of the idea that we can measure human well-being by GDP per capita, a standard shortcut in the development literature, and one with large political implications. (Narendra Modi, India’s recently elected prime minister, campaigned on his alleged development achievement in Gujarat: but closer inspection shows that, while Gujarat did very well on average GDP, it did much less well than Kerala and Tamil Nadu on health and education, in part because of the excellent quality of government services in those states, while Modi appears opposed to a large role for government.) Even if average GDP were the best single number to use as an index of welfare—and Deaton disputes this, making the familiar point that the profits of foreign investment are often repatriated by the investing country, so average household income would tell us more about how people are really doing—no single number is much good, given the complexity of human lives and what is worthwhile about them. Average GDP, moreover, does not include work done in the home (a point often stressed by Nancy Folbre and other feminist economists that has finally made it into the mainstream), and it does not include the value of leisure. And although there is a general correlation between GDP and some of the other good things Deaton mentions, the correlation can be disrupted. The high average GDP in the United States, for example, does not tell us about the inequalities that make for ill-fare (bad health, bad education, lack of political voice) in a distressingly large number of the nation’s inhabitants. These points are not new; they have pervaded the development literature for some time; but it is good to see them ringingly endorsed.

Nor does Deaton succumb to the lure of the once-again fashionable idea that we can measure welfare by “happiness,” defined as moment-to-moment feeling. (“Once again” because the similar view of Jeremy Bentham in the late eighteenth century, soon aptly criticized by his student John Stuart Mill, has now been revived with great éclat by the psychologist Daniel Kahneman, though without attention to Mill’s critique.) Feelings are important, says Deaton, but they are not reliable indicators of how people are really faring, because people adapt to hard conditions and to some extent tailor their satisfactions to what they think they can achieve—the phenomenon known in the economic literature as “adaptive preferences.” Moreover, Deaton adds, some valuable pursuits, such as love and the struggle for justice, require risk and effort, and may be accompanied at times by pain. Happiness, he concludes, is “a poor measure of overall wellbeing.” If we are to pay attention to survey data, he wisely suggests, we ought to prefer “life evaluation” surveys, which at least allow people to ponder many parts of their lives. But the important conclusion to draw, he says, is that there is no single measure of this complex notion, and “no magic question that provides a touchstone for judging well-being.”


This is a very important part of the book, except that it is not really a part: I have distilled these ideas from remarks scattered in many places. For the most part, the book’s analysis returns repeatedly to an emphasis on health and income, and the other dimensions of well-being, such as education and political participation, though often mentioned, are never analyzed. Deaton’s main concern is to study the complex relationship between income and health, and the rest is left as a set of pencil strokes to be filled in by someone else.

Even with regard to income and health, he is not terribly precise: he writes that these are “two of the most important components of well-being, and the two with which this book is mainly concerned.” But can income really be a component of well-being, rather than a useful all-purpose means? It is just some stuff, until it is put to work in the context of life as a set of opportunities. If we imagine a person who has a lot of money but cannot turn it into opportunities for functioning, we should not say that this person has one large component of well-being but lacks others. If he or she cannot use it, it does no good at all. One can make this issue vivid by thinking of people with disabilities who might have money but still cannot access the streets and transportation systems of their country, because they have been constructed to suit the needs of the “normal” person. In the Internet era, such a person can still do a good deal of shopping, converting income into opportunities for activity—but still the difference is huge between this person and the person whose use of income suffers from no unjust impediment. Deaton himself emphasizes the fact that income may or may not cause health: it all depends on what institutional structures surround it.


Deaton begins the historical part of his analysis with a brief account of what we know about well-being in human prehistory. About our hunter-gatherer ancestors, we learn two things of current significance. First, we find that infectious diseases are not endemic to the species, but the product of environmental conditions, especially overcrowding and poor sanitation. Second, we are confronted with the fact that inequality of welfare does not exist “by nature”: our ancestors were committed to sharing, and were able to survive only because they did share. Inequality, Deaton is convinced, is of very recent origin. This means that it is not one of those sad facts of nature about which we should wring our hands and moan; it is an artifact of human society that we can and should change.

From prehistory, Deaton leaps rather abruptly to sixteenth-century Europe (with a few stray remarks about the Greeks and Romans). Here we begin to study the manifold causes of premature death that are basically still with us: smoking, overcrowding, poor sanitation, a variety of infectious diseases, poor nutrition. Things begin to improve with the gradual increase in scientific knowledge: variolation (inoculation with the smallpox virus, the precursor of modern vaccination) leads to a decline in mortality, and then the germ theory of disease makes even further inroads. Deaton ascribes the huge increase in life expectancy from 1800 to 1945 (from forty to eighty years) largely to scientific knowledge combined with understanding of sanitation.

Deaton opens up the public-choice perspective that informs his analysis as a whole, though always somewhat vaguely. (This perspective uses economic tools to think about interest groups, about the division of work between markets and politics, and about the many ways in which politicians do not always act for the benefit of society.) Does economic growth matter? Certainly, grants Deaton. But turning money into better lives takes effective public health programs. Growth by itself does not guarantee public works. Here collective political action is crucial, and there are many impediments to effective collective action.

After World War II, progress speeds up because of penicillin and DDT, but today progress is sliding backward in many places, and we see people dying of the same old diseases. Immunizations and oral rehydration therapy help, but the bottom line remains the same: there is no stable progress without effective public health programs provided by government. Returning to the contention that growth is the main cause of progress, Deaton avers that growth is needed to supply food and the money for sanitation and medical services. Too exclusive a focus on growth, however, is a mistake, as we can see from China: when growth became the main focus, public health and health care programs suffered. Here Deaton usefully contrasts the state of Kerala in India, where effective government health services have made progress possible without Chinese-style coercion. Much of Deaton’s prior work has been devoted to India, and he is particularly fascinating here, and full of valuable data, showing the many reasons why institutions are crucial. In Rajasthan, which has a weak state government, “doctors” exploit the poor without regulation or licensing. In Kerala and Tamil Nadu, by contrast, government takes the responsibility for maintaining decent standards. We should not forget the extent of government’s role in keeping people safe. Not just government clinics, but the very basic fact that one can have confidence in the qualifications of doctors and nurses—all this requires a regulatory state, and that in turn requires the public will to support one.

Today the richer countries have by and large moved beyond a first wave of diseases—that is, infectious diseases and diseases of poor sanitation. (Deaton mentions but does not dwell on the continuing devastation of HIV / aids in both rich and poor nations, which of course bears out his thesis that effective interventions require, and do not always receive, effective government action.) Today those same countries are also moving beyond a next set of diseases, cardiovascular disease and cancer, through reduction in smoking and a range of medical discoveries. The next frontier is confronting diseases of aging, such as Alzheimer’s disease and other dementias, and providing quality-of-life interventions such as dialysis and joint replacements. These interventions are extremely expensive, but progress is being made.


Deaton’s readers might at this point be thinking that things are pretty good in America today—onward and upward, from infectious diseases to cancer and heart disease, and now to dementia. In an important chapter, however, Deaton corrects this supposition by showing that inequality in both income and health has been steadily on the rise in the United States. All is not well for a large proportion of Americans. After 1970, growth does not make a dent in the poverty rate. (Deaton provides an excellent discussion of debates about the definition of poverty.) He documents shortfalls in education that have contributed to the unequal opportunities of working-class people. But as usual he emphasizes political factors: the declining power of unions, the presence among the poor of many people who are unable to vote, including illegal immigrants but also convicted felons. He also notes that another potentially vulnerable group, the elderly, have done somewhat better through political clout.

In his most extensive use of a public-choice perspective, Deaton argues that lobbying has been a key factor in the increase in incomes at the top. He also argues that growth at the top can for such reasons become self-reinforcing, as the extremely rich assume more and more political clout, creating a plutocracy that disenfranchises others. Institutions hijacked by the elite may also, he suggests, be inimical to growth. And government bailouts, caused by such institutions, have encouraged excessive risk-taking, itself inimical to both growth and well-being.

This chapter is the book’s most frustrating, because its insights are extremely important and their development is extremely thin. Deaton makes assertions about the importance of education and its role in guaranteeing opportunity, but he provides no data and no analysis—not even discussing the well-known work of the economist James Heckman, whose work on early intervention in childhood education would have supported Deaton’s claims. Surprisingly, even well-known health data are missing. Amartya Sen showed in a famous article in Scientific American that the health status of people in Kerala in India is equal to that of American citizens in Harlem. This neat fact receives no mention, and yet it would have supported so well both Deaton’s earlier claims about what excellent government services have achieved in several Indian states and his observations about political failure in the United States. Nor do we have any analysis of how government failures contribute to the comparative ill health of many Americans, or of how other richer countries have achieved somewhat better health outcomes. All these insights are suggestive, but we must take it from there and do some further work.

Deaton now uses the preceding analysis to confront what has been his main topic from the beginning: the inequalities between nations, rapidly growing, that cause many of the world’s people to be utterly unable to avail themselves of the lovely progress he has depicted. Once again he warns the reader of theoretical difficulties: comparison of living standards is very slippery, given that currency conversion does not really correspond to buying power. The poverty line, difficult to draw within a nation, proves all the more elusive across nations. Moreover, we cannot compare easily by thinking about things that people feel they need, since people care about very different things. Deaton here uses an example that American visitors to Britain will find quite convincing. Most British people highly value and feel that they need Marmite, which I would describe as a disgusting yeasty salty paste that spoils any piece of bread onto which it is spread. Many Americans, on the other hand, intensely value bourbon, which British people mainly dislike. (But how could it possibly give rise to the revulsion occasioned by Marmite?) Life is like that across the board, so we need to figure out some middle ground between income, which is too abstract (not capturing the availability of specific things that people care about), and Marmite, which is too specific. Few development studies are up to the task, and so we do not really have a very good account of global inequality.

But we do know that inequality is on the rise: some countries are progressing, others are left behind. Here I wish Deaton had referred his readers to the many pages of tables in the Human Development Reports of the United Nations Development Programme: not the famous Human Development Index, which tells us relatively little, but the tables after the index, which compare nations on dozens of parameters in the intermediate zone, such as life expectancy, infant mortality, maternal mortality, access to safe drinking water, mean years of schooling, and so on. These tables support Deaton’s general claim, and they give a good idea of how development experts are addressing the measurement problem.

But why haven’t the poorer nations caught up? Deaton rejects the facile answer of overpopulation. It was a mistake, he argues, to say that more people means poorer people. Things are much more complex than that, as by now we see, and draconian limits on childbearing are not only objectionable from the point of view of human freedom but also often counterproductive, since they may ultimately diminish the political clout of poorer classes. Above all, Deaton argues, the answer to the “escape” question is to be sought in the nature and the quality of political institutions. The “escape” requires effective governance, the rule of law, an effective tax system, secure property rights—and public confidence in all of these.


In what will surely prove the most controversial part of his book, Deaton argues that the problems of global poverty will not be solved, and, indeed, are often made worse, by extensive foreign aid and international private philanthropy. People in rich countries want to help; they feel a moral imperative to help. But when they give money to poor countries, the money usually does not help to solve the problem. As the economist Peter Bauer argued long ago, if all conditions for development except capital are present, then capital will soon be generated locally or will become available to government or private businesses out of loans from abroad. If the other preconditions of development are not present, inflow of capital will be unproductive and ineffective.

To Bauer’s basic argument, Deaton adds some more specific arguments born of his own long experience. First, foreign aid is rarely guided by the real needs of recipients: far more often it is guided by the priorities and the interests of the donor country. Second, aid is typically given to governments, and all too often it is given to governments that have no interest in helping their own people. It may just entrench the autocrats in power. And even when aid is not actually misallocated, it just takes the place in the national budget that would have been used to address the problem, freeing more money to be used on other projects. Worse still, it often decreases government’s incentive to solve the problem itself. Aid should at the very least be conditional on improved performance, but the absence of real cases in which conditionality is enforced means that we really cannot study the effects of aid constrained in this way. But even well-crafted aid policies tend to weaken the contract between government and the governed, which in Deaton’s view is crucial to long-term economic and social development. If a government can raise funds without needing the consent of its people, institutions gradually become unmoored from popular control and can eventually turn “toxic.” Deaton’s ostensibly hard-hearted advice to stop aiding is directly linked to his central insight about the pivotal role of government, and government’s response to people’s needs. His approach will be contested, but any challenge to it must grapple with the clear argument, through which Deaton has linked his institutional perspective to specific prescriptions.


Is there any sort of aid that escapes Deaton’s strictures? When people are dying in large numbers, Deaton argues, the moral imperative to aid is especially urgent, and there are forms of health aid that are not vulnerable to his criticisms, or at least far less vulnerable. He distinguishes “vertical programs,” focused on a single disease, from “horizontal programs,” focused on broad provision of health services. The latter are mistaken interventions, since long-term success in creating health infrastructure can be achieved only by government, and government is likely to be undermined by foreign aid. The former, by contrast, are often useful—programs dealing with tuberculosis, malaria, HIV / aids, or some other specific disease. Deaton grudgingly allows that “horizontal” programs might possibly be a good idea if governments have a track record of pursuing policies that benefit their people (as in the requirements imposed by the U.S. Millennium Challenge Corporation). But he remains skeptical, in the absence of evidence, that even the best such programs can help.


Readers of this part of Deaton’s book will very likely be disconcerted. He is telling people who feel guilty and who strongly want to do something that there is nothing they can do. Even worse, when trying to do good, they are actually in most cases doing harm. Such readers should certainly study the opposing arguments of Jeffrey Sachs and other opponents who are fully and fairly discussed here. But I wish Deaton had done more to study other possible types of “vertical” aid. What about aid to education of various types and in various places? Surely one doesn’t get good governance without a widely literate population, and perhaps it is worth getting that by non-ideal means, where local politics is not delivering it, rather than allowing a generation to grow up without the wherewithal to intervene in their nation’s affairs. And what about aid targeted at the environment, or endangered species? The former is on Deaton’s agenda but is rarely mentioned. The latter is completely missing from his framework. So this section of his argument is incomplete, like the others. But it warns usefully about what economist William Easterly recently called “the tyranny of experts.”


Deaton’s insights are important, and this somewhat breezy book, light on analysis and full of gaps, does not entirely do them justice. In many respects, Deaton agrees with the conclusions of Thomas Piketty’s justly lauded Capital in the Twenty-First Century, insisting on the self-reinforcing character of inequality and the damage that it brings with it. But Deaton makes a distinctive contribution to the inequality debate in several ways. His data-rich account of developing countries, added to his insightful analysis of the many aspects of health, supplies a much-needed global perspective and raises important questions about how to address global inequalities. His sketch of a multifarious account of human well-being, together with his criticisms of the dominant GDP paradigm and the newer happiness perspective, supply a sound basis for future policy prescriptions. Most important of all is Deaton’s insistence on the crucial importance of institutions and governmental action, together with his use of a public-choice perspective that sheds light on the difficulties of reaching an adequate political solution when different political groups interact and compete. Piketty’s analysis, crucial though it is, neglects politics and the public-choice angle, as Saul Levmore has pointed out. Deaton opens up this perspective, though he does not seem acquainted with the public-choice literature in any detail. This voice is therefore one we need to hear as we continue to debate these urgent questions. 

Yet the book needs more explicit theoretical argumentation. On the normative side, the scattered insights into the multiple components of human flourishing need to be gathered into one place and treated to a thorough analysis. Since Deaton’s intuitions have much in common with the “capabilities approach” developed by Amartya Sen and others, it would be useful to know his view of the theoretical arguments advanced in support of that perspective and of the numerous debates within it. Since he criticizes the GDP measure but also seems to think it useful in some contexts, we need an explicit discussion of when an admittedly incomplete theoretical measure does useful work. Current debates about how to allocate a value to household labor also should be discussed. The detailed analysis of health needs to be balanced by equally detailed accounts of education, political access, and other aspects of human flourishing that Deaton himself stresses. It is not surprising that an economist whose lifelong preoccupation has been with health did not do this, but it needs to be done if we are to see clearly where we are and where we ought to be heading. 

Deaton’s historical and comparative materials are too impressionistic. The account of inequality in the United States should be buttressed with more comparisons with other rich nations. And the historical account needs to be less scattershot. Writing a comprehensive history of world inequality is not Deaton’s purpose—and yet, instead of leaping over from the hunter-gatherers to sixteenth-century Europe, Deaton could at least have made more, say, of his scattered examples of Roman sanitation and public works. A lot is known about the benefits and shortcomings of Roman governmental action, not only in the water-sanitation realm but also in extending the rule of law. A casual visitor to the ancient Roman military installation in Northumberland can observe that the common Roman soldier had toilet facilities unknown in parts of Britain until close to the present day. (On a summer exchange program in Swansea in high school, I used an outhouse with no flush, whereas those Roman soldiers had running water flowing by in a trench under their seat.) As one can see from that example, the “great escape” is really a more complicated story of advances and retreats. 

Deaton’s important and distinctive ideas make us want the weightier tome that would propel him into the front ranks of the inequality debate, probably the most important debate about human welfare in our time.