Development as Freedom

by Sen, Amartya

  The importance of this connection was a crucial point of focus of my recent work on India, done jointly with Jean Drèze, dealing with economic reforms.12 In many ways, the economic reforms have opened up for the Indian people economic opportunities that were suppressed by overuse of control and by the limitations of what had been called the “license Raj.”13 And yet the opportunity to make use of the new possibilities is not independent of the social preparation that different sections of the Indian community have. While the reforms were overdue, they could be much more productive if the social facilities were there to support the economic opportunities for all sections of the community. Indeed, many Asian economies—first Japan, and then South Korea, Taiwan, Hong Kong, and Singapore, and later post-reform China and Thailand and other countries in East Asia and Southeast Asia—have done remarkably well in spreading the economic opportunities through an adequately supportive social background, including high levels of literacy, numeracy, and basic education; good general health care; completed land reforms; and so on. The lesson of opening of the economy and the importance of trade has been more easily learned in India than the rest of the message from the same direction of the rising sun.14

  India is, of course, highly diverse in terms of human development, with some regions (most notably, Kerala) having much higher levels of education, health care and land reform than others (most notably, Bihar, Uttar Pradesh, Rajasthan and Madhya Pradesh). The limitations have taken different forms in the different states. It can be argued that Kerala has suffered from what were until recently fairly anti-market policies, with deep suspicion of market-based economic expansion without control. So its human resources have not been as well used in spreading economic growth as they could have been with a more complementary economic strategy, which is now being attempted. On the other hand, some of the northern states have suffered from low levels of social development, with varying degrees of control and market-based opportunities. The need for seizing the relevance of complementarity is very strong in remedying the diverse drawbacks.

  It is, however, interesting that despite the rather moderate record in economic growth, Kerala seems to have had a faster rate of reduction in income poverty than any other state in India.15 While some states have reduced income poverty through high economic growth (Punjab is the most notable example of that), Kerala has relied a great deal on expansion of basic education, health care and equitable land distribution for its success in reducing penury.

  While these connections between income poverty and capability poverty are worth emphasizing, it is also important not to lose sight of the basic fact that the reduction of income poverty alone cannot possibly be the ultimate motivation of antipoverty policy. There is a danger in seeing poverty in the narrow terms of income deprivation, and then justifying investment in education, health care and so forth on the ground that they are good means to the end of reducing income poverty. That would be a confounding of ends and means. The basic foundational issues force us, for reasons already discussed, toward understanding poverty and deprivation in terms of lives people can actually lead and the freedoms they do actually have. The expansion of human capabilities fits directly into these basic considerations. It so happens that the enhancement of human capabilities also tends to go with an expansion of productivities and earning power. That connection establishes an important indirect linkage through which capability improvement helps both directly and indirectly in enriching human lives and in making human deprivations more rare and less acute. The instrumental connections, important as they are, cannot replace the need for a basic understanding of the nature and characteristics of poverty.

  INEQUALITY OF WHAT?

  The treatment of inequality in economic and social evaluation involves many dilemmas. Substantial inequalities are often hard to defend in terms of models of “fairness.” Adam Smith’s concern with the interests of the poor (and his outrage at the tendency for those interests to be neglected) related naturally to his use of the imaginative device of what it would look like to an “impartial spectator”—an inquiry that offers far-reaching insights on the requirements of fairness in social judgment.16 Similarly, John Rawls’s idea of “justice as fairness” in terms of what can be expected to be chosen in a hypothetical “original position” in which people do not yet know who they are going to be provides a rich understanding of the demands of equity, and yields the anti-inequality features that are characteristic of his “principles of justice.”17 Patent inequalities in social arrangements can also be difficult to justify in terms of reasonableness to actual members of the society (for example, the case for these inequalities being one that others “cannot reasonably reject”: a criterion that Thomas Scanlon has proposed—and powerfully used—for ethical evaluation).18 Certainly, severe inequalities are not socially attractive, and momentous inequalities can be, some would argue, downright barbaric. Furthermore, the sense of inequality may also erode social cohesion, and some types of inequalities can make it difficult to achieve even efficiency.

  And yet attempts to eradicate inequality can, in many circumstances, lead to loss for most—sometimes even for all. This kind of conflict can arise in mild or severe form depending on the exact circumstances. Models of justice—involving the “impartial spectator,” or the “original position,” or not-reasonable-rejection—have to take note of these diverse considerations.

  Not surprisingly, the conflict between aggregative and distributive considerations has received a remarkable amount of professional attention among economists. This is appropriate since it is an important issue.19 Many compromise formulas have been suggested for evaluating social achievements by taking note simultaneously of aggregative and distributive considerations. A good example is A. B. Atkinson’s “equally distributed equivalent income,” a concept that adjusts the aggregate income by reducing its accounted value according to the extent of inequality in income distribution, with the tradeoff between aggregative and distributive concerns being given by the choice of a parameter that reflects our ethical judgment.20

  There is, however, a different class of conflicts that relates to the choice of “space”—or of the focal variable in terms of which inequality is to be assessed and scrutinized—and this relates to the subject matter of the previous chapter. Inequality of incomes can differ substantially from inequality in several other “spaces” (that is, in terms of other relevant variables), such as well-being, freedom and different aspects of the quality of life (including health and longevity). And even aggregative achievements would take different forms depending on the space in which the composition—or the “totaling”—is done (for example, ranking societies in terms of average income may differ from ranking them according to average health conditions).

  The contrast between the different perspectives of income and capability has a direct bearing on the space in which inequality and efficiency are to be examined. For example, a person with high income but no opportunity of political participation is not “poor” in the usual sense, but is clearly poor in terms of an important freedom. Someone who is richer than most others but suffers from an ailment that is very expensive to treat is obviously deprived in an important way, even though she would not be classified as poor in the usual statistics of income distribution. A person who is denied the opportunity of employment but given a handout from the state as an “unemployment benefit” may look a lot less deprived in the space of incomes than in terms of the valuable—and valued—opportunity of having a fulfilling occupation. Since the issue of unemployment is particularly important in some parts of the world (including contemporary Europe), this is another area where there is a strong need to seize the contrast between income and capability perspectives in the context of inequality assessment.

  UNEMPLOYMENT AND CAPABILITY DEPRIVATION

  That the judgments of inequality in the space of incomes can be quite different from those related to important capabilities can easily be illustrated with examples of some practical importa
nce. In the European context, this contrast is particularly significant because of the wide prevalence of unemployment in contemporary Europe.21 The loss of income caused by unemployment can, to a considerable extent, be compensated by income support (including unemployment benefits), as it typically is in Western Europe. If income loss were all that were involved in unemployment, then that loss could be to a great extent erased—for the individuals involved—by income support (there is, of course, the further issue of social costs of fiscal burden and incentive effects involved in this compensation). If, however, unemployment has other serious effects on the lives of the individuals, causing deprivation of other kinds, then the amelioration through income support would be to that extent limited. There is plenty of evidence that unemployment has many far-reaching effects other than loss of income, including psychological harm, loss of work motivation, skill and self-confidence, increase in ailments and morbidity (and even mortality rates), disruption of family relations and social life, hardening of social exclusion and accentuation of racial tensions and gender asymmetries.22

  Given the massive scale of unemployment in contemporary European economies, the concentration on income inequality only can be particularly deceptive. Indeed, it can be argued that at this time the massive level of European unemployment constitutes at least as important an issue of inequality, in its own right, as income distribution itself. An exclusive focus on income inequality tends to give the impression that Western Europe has done very much better than the United States in keeping inequality down and in avoiding the kind of increase in income inequality that the United States has experienced. In the space of incomes, Europe does indeed have a clearly better record both in terms of levels and trends of inequality, as is brought out by the careful investigation reported in the OECD (Organization for Economic Cooperation and Development) study prepared by A. B. Atkinson, Lee Rainwater and Timothy Smeeding.23 Not only are the usual measures of income inequality higher in the United States than is the case, by and large, on the European side of the Atlantic, but also the U.S. income inequality has gone up in a way that has not happened in most countries in Western Europe.

  And yet if we shift our gaze from income to unemployment, the picture is very different. Unemployment has risen dramatically in much of Western Europe, whereas there has been no such trend in the United States. For example, in the period 1965–1973, the unemployment rate was 4.5 percent in the United States, while Italy had 5.8 percent, France 2.3 percent, and West Germany below 1 percent. By now all three—Italy, France, and Germany—have unemployment rates that hover around 10 to 12 percent, whereas the U.S. unemployment rate is still between 4 and 5 percent. If unemployment batters lives, then that must somehow be taken into account in the analysis of economic inequality. The comparative trends in income inequality give Europe an excuse to be smug, but that complacency can be deeply misleading if a broader view is taken of inequality.24

  The contrast between Western Europe and the United States raises another interesting—and in some ways a more general—question. American social ethics seems to find it possible to be very non-supportive of the indigent and the impoverished, in a way that a typical Western European, reared in a welfare state, finds hard to accept. But the same American social ethics would find the double-digit levels of unemployment, common in Europe, to be quite intolerable. Europe has continued to accept worklessness—and its increase—with remarkable equanimity. Underlying this contrast is a difference in attitudes toward social and individual responsibilities, to which I shall return.

  HEALTH CARE AND MORTALITY: AMERICAN AND EUROPEAN SOCIAL ATTITUDES

  The inequality between different racial groups in the United States has received considerable attention recently. For example, in the space of incomes African Americans are decidedly poorer than American whites. This is very often seen as an example of relative deprivation of African Americans within the nation, but not compared with poorer people in the rest of the world. Indeed, in comparison with the population of third world countries, African Americans may well be a great many times richer in terms of incomes, even after taking note of price differences. Seen this way, the deprivation of the American blacks seems to pale to insignificance in the international perspective.

  But is income the right space in which to make such comparisons? What about the basic capability to live to a mature age, without succumbing to premature mortality? As was discussed in chapter 1, in terms of that criterion the African American men fall well behind the immensely poorer men of China, or the Indian state of Kerala (see figure 1.1, this page)—and also of Sri Lanka, Costa Rica, Jamaica and many other poor economies. It is sometimes presumed that the remarkably high death rates of African Americans apply only to men, and again only to younger men, because of the prevalence of violence. Death from violence is indeed high among young black men, but this is by no means the whole story. Indeed, as figure 1.2 (this page) shows, black women too fall not only behind white women in the United States but also behind Indian women in Kerala, and come very close to falling behind Chinese women as well. It may also be noticed in figure 1.1 that American black men continue to lose ground vis-à-vis the Chinese and the Indians over the years—well past the younger ages when death from violence is common. More explanation is needed than violent deaths can provide.

  Indeed, even if we take higher age groups (say, that between thirty-five and sixty-four years), there is evidence of enormously greater mortality for black men vis-à-vis white men, and black women vis-à-vis white women. And these differentials are not wiped out by adjustment for income differences. In fact, one of the more careful medical studies related to the 1980s shows that the black-white mortality differential remains remarkably large for women even after adjustment for income differentials. Figure 4.1 presents the ratios of the mortality rates of blacks and whites for the country as a whole (based on a sample survey).25 While U.S. black men have 1.8 times the mortality rate of white men, black women have nearly three times the mortality of white women in this survey. And adjusted for differences in family income, while the mortality rate is 1.2 times higher for black men, it is as much as 2.2 times higher for black women. It, thus, appears that even after full note is taken of income levels, black women die young in very much larger proportions than white women in the contemporary United States.

  FIGURE 4.1: Mortality Rate Ratios of Blacks to Whites (Aged 35–54) Actual and Adjusted for Family Income

  Source: M. W. Owen, S. M. Teutsch, D. F. Williamson and J. S. Marks, “The Effects of Known Risk Factors on the Excess Mortality of Black Adults in the United States,” Journal of the American Medical Association 263, no. 6 (February 9, 1990).

  The broadening of the informational base from income to the basic capabilities enriches our understanding of inequality and poverty in quite radical ways. When we focused on the ability to be employed and to have the associated advantages of employment, the European picture looked quite dismal, and as we turn our attention to the ability to survive, the picture of American inequality is remarkably intense. Underlying these differences and the respective policy priorities associated with them, there may be an important contrast in the attitudes to social and individual responsibilities on the two sides of the Atlantic. In American official priorities, there is little commitment to providing basic health care for all, and it appears that many millions of people (in fact more than 40 million) are without any kind of medical coverage or insurance in the United States. While a considerable proportion of these uninsured people may have volitional reasons for not taking such insurance, the bulk of the uninsured do, in fact, lack the ability to have medical insurance because of economic circumstances, and in some cases because of preexisting medical conditions that private insurers shun. A comparable situation in Europe, where medical coverage is seen as a basic right of the citizen irrespective of means and independent of preexisting conditions, would very likely be politically intolerable. The limits on governmental support for the ill and the poor are
too severe in the United States to be at all acceptable in Europe, and so are the social commitments toward public facilities varying from health care to educational arrangements, which the European welfare state takes for granted.

  On the other hand, the double-digit unemployment rates that are currently tolerated in Europe would very likely be (as was argued earlier) political dynamite in America, since unemployment rates of that magnitude would make a mockery of people’s ability to help themselves. I believe no U.S. government could emerge unscathed from the doubling of the present level of unemployment, which incidentally would still keep the U.S. unemployment ratio below what it currently is in Italy or France or Germany. The nature of the respective political commitments—and lack thereof—would seem to differ fundamentally between Europe and America, and the differences relate closely to seeing inequality in terms of particular failures of basic capabilities.

  POVERTY AND DEPRIVATION IN INDIA AND SUB-SAHARAN AFRICA

  Extreme poverty is now heavily concentrated in two particular regions of the world: South Asia and sub-Saharan Africa. They have among the lowest levels of per capita income among all the regions, but that perspective does not give us an adequate idea of the nature and content of their respective deprivations, nor of their comparative poverty. If poverty is seen, instead, as the deprivation of basic capabilities, then a more illuminating picture can be obtained from information on aspects of life in these parts of the world.26 A brief analysis is attempted below, based on a joint study with Jean Drèze, and on two follow-up works of this author.27

  Around 1991 there were fifty-two countries where the expectation of life at birth was below sixty years, and those countries had a combined population of 1.69 billion.28 Forty-six of these countries are in South Asia and sub-Saharan Africa—only six are outside these two regions (viz. Afghanistan, Cambodia, Haiti, Laos, Papua New Guinea and Yemen), and the combined population of these six is only 3.5 percent of the total population (1.69 billion) of the fifty-two low-life-expectancy countries. The whole of South Asia except Sri Lanka (i.e., India, Pakistan, Bangladesh, Nepal and Bhutan) and the whole of sub-Saharan Africa except South Africa, Zimbabwe, Lesotho, Botswana, and a collection of tiny islands (e.g., Mauritius and the Seychelles) belong to the group of the other forty-six low-life-expectancy countries. Of course, there are variations within each country. Well-placed sections of the population of South Asia and sub-Saharan Africa enjoy high longevity, and as was discussed earlier, parts of the population of countries even with very high average life expectancy (such as the United States) may have survival problems that compare with conditions in the third world. (For example, American black men in U.S. cities such as New York, San Francisco, St. Louis, or Washington, D.C., have life expectancies well below our cut-off point of sixty years.29) But in terms of country averages, South Asia and sub-Saharan Africa do indeed stand out as the regions where short and precarious lives are concentrated in the contemporary world.