The Blackwell Companion to Sociology (40 page)

to household income, especially in the rural areas of the developing world. If

these children are to be sent to the school, household income will go down

substantially at least in the immediate future. For richer families, however, this opportunity cost is negligible relative to their total income. On the other hand, the cost of raising an extra child is pretty low for a poor family, given the bare minimum of food, clothing, and shelter with which they are accustomed to get

by. For richer families, this cost is not so small. Besides, their opportunity costs of raising children will also be high if working mothers have to give up well paid jobs, or if they have to employ paid workers, to look after the children.

These relationships suggest that as a family climbs up the income scale, the

cost of raising children will go up while the cost of educating them will go down.

Parents will then be more inclined to limit fertility and to educate their children, i.e. to trade off quantity for quality. It follows, then, that, for a given level of per capita income, a society with a more equal income distribution will have lower

fertility and higher education than one with a less equal distribution, because a more equal society will have fewer poor families. Both these consequences ±

lower fertility and higher education ± will in turn help a poor economy to grow

faster.

The final category of theories ± the socio-political instability theories ± ba-

sically formalises the age-old idea that gross inequalities are likely to incite 158

Siddiqur Rahman Osmani

violence, rebellion, or attempts to sabotage the established social order by those who feel severely deprived. The resulting breakdown in the rule of law will

create uncertainty in the enforcement of property rights, and in the absence of

certainty about property rights investors will have little incentive to invest. As a result, the economy will grow more slowly. By contrast, a more equal society will enjoy greater certainty of property rights and hence more robust growth.

The proponents of all these theories have tried to test them statistically by

using data drawn from a large number of countries. The test essentially involves an attempt to discern from cross-country data whether countries with lower

initial inequality enjoyed a higher rate of growth, other things remaining the

same. The initial tests almost invariably lent support to the hypothesis that

equality promotes growth, but they were subject to a couple of problems.

First, the inequality variable was measured from poor-quality distributional

data, in the sense that the data were hardly comparable across countries and

sometimes even in the same country at different points in time because of wide

differences in the nature of surveys used to collect income distribution data.

Second, the tests were done in a manner that made it impossible to discriminate

between the alternative theories.

Both these problems have recently been addressed in an important work by

Perotti (1996). He has used carefully screened distributional data, and tried to discriminate among alternative theories by deriving testable implications that are different for different theories. Using a sample comprising both developed and

developing countries, he has confirmed the earlier finding that equality does

appear to have a positive effect on growth. He also finds that among the four

mechanisms discussed above, there is strong empirical support for the instability mechanism and somewhat weaker support for the endogenous schooling and

fertility mechanism, but not much support for the other two.

An interesting result emerged when Perotti tried to test the hypothesis separ-

ately for rich and poor countries. He found that equality had a strong positive

effect on growth in the rich countries, but no significant effect, either positive or negative, in the poor countries. This finding would seem to strike a severe blow to the cause of egalitarianism in poor countries. It used to be said in the olden days, when most people believed in a trade-off between equity and growth, that

equity is a luxury only the rich countries can afford, because only they can afford to make the necessary sacrifice in terms of growth. Perotti's findings would now seem to suggest that the rich countries are even luckier than that ± they don't

need to make the sacrifice after all. They can eat the cake and still see it grow, while the poor countries are still left out of the party.

Further reflection shows, however, that such a negative conclusion may not

really be warranted. In the first place, as Perotti himself notes, the negative effect of inequality on growth could be true as much for the poor countries as for the

rich, and it is only some statistical problems that hide the fact for the poor

countries. Even if this is dismissed as a triumph of hope over facts and the finding is taken at its face value, one can still argue that there is no support for the traditional view that the quest for equality will actually retard growth. This

argument provides at least a weak case for adopting egalitarian policies, even in On Inequality

159

poor countries. It is arguable, however, that the case for egalitarian policies is in fact much stronger, Perotti's finding notwithstanding. There are two fundamental reasons why his finding cannot be taken as decisive evidence for the inability of equality to promote growth in poor countries, even if all the statistical

problems are left aside.

The first reason lies in the very nature of the statistical exercise ± regression analysis ± that forms the basis of this finding. All one can infer from this kind of exercise is that incremental changes in inequality have no effect on growth. That leaves open the question of what happens when radical reforms are undertaken

to bring about large changes in the distribution of income. There are in fact some plausible models in which small reductions in inequality will not promote

growth, but large reductions will. These models are characterized by the so-

called threshold effect. For example, in the models of capital formation with

credit constraint discussed above, it is possible that in order to make any

worthwhile investment it will be necessary to spend a large sum of money to

pay for large fixed costs. In that case, a small-scale redistribution resulting in a small increase in the income of a credit-constrained poor household may not be

enough for it to overcome the minimum threshold of fixed costs. Greater equal-

ity will then do nothing to promote growth, but a large-scale redistribution will solve this particular problem.

In fact, when one recalls that the recent theoretical interest in the growth-

promoting effect of equality has its origin in the experience of East Asia, one

should realize that it is large-scale redistributions that are really relevant. What happened in East Asia was not marginal changes in tax-expenditure policies that

can do no more than tinker with the distribution, but radical land reforms that

fundamentally altered the distribution of income. It is arguable that if greater equality did have a growth-promoting role in East Asia, it was only because

distribution was altered so radically.

The second reason why Perotti's finding cannot be regarded as decisive is that

it deals exclusively with the distribution of private income. This focus may be

too narrow, especially in the context of human capital formation. For improving

the education and health of poor children, public provision of good quality

services may be more important than small increments in the private incomes

of their parents ± the celebrated examples of Sri Lanka, Costa Rica, China, and

the Indian state of Kerala prove this point. In poor areas, where private schools and health facilities have not developed because the potential clientele are too poor to make the necessary investment worthwhile, a slight increase in private

incomes of the poor will fail to buy any extra education or health services. Public provision will be much more effective in this situation. This line of argument

suggests that egalitarian policies that take the form of greater public provisioning for the poor may well be growth-promoting, even though redistribution of

private incomes may not be.

It follows from the preceding argument that if one wants to test the effect of

inequality on growth, then it may not be enough to look only at private income

distribution. One would need to test whether growth is related to a measure of

inequality that reflects the distribution of both private income and public

160

Siddiqur Rahman Osmani

provisioining ± for example, inequality in educational achievement. Such a test

has recently been carried out by Birdsall and Londono (1997). Their results show that while income inequality has no significant effect on growth, educational

inequality has a significantly negative effect. This finding indicates that a

broader concept of equality may well be growth-enhancing.

This broader concept is of course nothing other than the equality of basic

capabilities discussed above. It was noted there that on the grounds of ethical

plausibility equality in the space of capabilities can arguably claim superiority over equality in any other space, such as utility, income, or resources. The

arguments of the present section suggest that equality of capability has much

to commend itself also on the grounds of economic interaction between distribu-

tion and growth. Policies aimed at achieving equality of basic capabilities can

not only allay the traditional concern about the growth±equity trade-off, but

may even lead to a virtuous circle in which economic growth and equitable

distribution reinforce each other.

12

The Persistence of Poverty in a

Changing World

Melvin L. Oliver and David M. Grant

Poverty is pain; it feels like a disease. It

attacks a person not only materially but

also morally. It eats away one's dignity

and drives one into total despair.

A poor woman, Moldova ( World Bank

Group, 1999, p. 25)

Globalization is rapidly changing the world we live in and the social context

within which poverty exists. On the one hand, technological innovations are

bringing people closer together in both the real and virtual worlds. Expansion of the global economy has brought an increasing number of people into formal

economies and created a growing interdependence among people, nations, and

international institutions. Democratic institutions are proliferating along with ideologies that endorse citizen participation. The Internet has dramatically

improved access to information and is eliminating boundaries and borders to

revolutionize the distribution of goods and services. Yet one of the world's oldest social problems ± poverty ± remains stubbornly present despite rapidly shifting

global institutional and organizational structures. The prophesy of the New

Testament, that ``ye have the poor with you always'' (Mark 14:7), seems as

true today as ever.

From a sociological perspective, confronting the issue of poverty raises a

number of challenging questions concerning the social forces responsible for

the persistence and intractability of poverty. What is poverty? Who is poor?

How can poverty be measured? Why are people poor and why do people remain

in poverty? And what strategies are most effective in reducing poverty? Provid-

ing complete answers to these questions is far beyond the scope of this chapter.

162

Melvin L. Oliver and David M. Grant

Rather, we highlight recent trends and dimensions in the incidence of poverty in the United States, one of the most advanced technological societies, and in a

number of the world's least developed regions and societies. This strategy, we

believe, reveals much about the multidimensional, complex web of factors

responsible for poverty, as well as some of the common poverty-reducing strat-

egies that seek to provide and build resources among those in need.

To understand the extent of poverty, we first address definitions and measures

of poverty, focusing on the difference between absolute and relative measures of poverty. We then move to a discussion of poverty in the USA and in less-developed societies. Each of these sections addresses broad trends in the incid-

ence of poverty to identify who are poor and the complexity of factors related to why they are poor. The final section of the chapter considers novel strategies to reduce poverty that go beyond traditional income transfers policies. To reduce

the violence of poverty, we argue, requires a focus on developing and building

assets among those in need. We use illustrative examples from the USA and India

to demonstrate the relationship between assets and shifting people from eco-

nomic marginality to economic security.

Defining and Measuring Poverty

The concept of poverty appears to be quite straightforward: poverty simply

means to be poor, to be lacking in the basic material resources for an adequate

human existence. At base, poverty is a function of the impact of macroeconomic

forces as they are mediated by the mechanisms of governmental and institutional

structures. Poverty is associated with expansive forces far beyond the reach of

the poor themselves, such as economic trends, technology, financial institutions, the division of labor, political structures, infrastructure, health care, courts and criminal justice systems, the military, and much more. Yet the experience of

poverty is personal, amorphous, and multidimensional. As a normative social