What’s wrong with poverty

John Broome

  • On Ethics and Economics by Amartya Sen
    Blackwell, 131 pp, £14.95, May 1987, ISBN 0 631 15494 9
  • The Standard of Living by Amartya Sen, edited by Geoffrey Hawthorn
    Cambridge, 125 pp, £15.00, September 1987, ISBN 0 521 32101 8

Welfare economics is concerned with what economic arrangements we should have, and what governments should do in economic matters. It is about right and good in economics. So it is a branch of ethics, the study of right and good in general. There is no dividing line between theoretical welfare economics and practical ethics.

Economists, however, have generally tried to avoid the kind of philosophical argument that ethics requires. Nevertheless, some of the wildest philosophical opinions circulate among them. In 1871, one of the pioneers of modern economics, W.S. Jevons, wrote in The Theory of Political Economy: ‘The susceptibility [to pleasure] of one mind may, for what we know, be a thousand times greater than that of another. But, provided that the susceptibility was different in a like ratio in all directions, we should never be able to discover the difference. Every mind is thus inscrutable to every other mind.’ For several decades starting in the 1930s, particularly after the publication in 1932 of Lionel Robbins’s An Essay on the Nature and Significance of Economic Science, it was dogma among economists that one cannot in principle compare one person’s good with another’s. At any rate – this was Robbins’s own view – statements that depend on comparisons like this must be excluded from ‘economic science’. Robbins thought that interpersonal comparisons of good were ethical and therefore unscientific.

This idea gave rise to a style of welfare economics that still lingers. In the United States, indeed, it still predominates. Virtually everything that happens in an economy is good for some people and bad for others. The dogma says economics cannot compare the benefit to some with the loss to others. So it cannot judge whether a change is good or bad on balance. Economics, then, is unable to judge the merits of virtually any economic event. It is even deprived of one of its favourite doctrines. ‘It was not possible to say that economic science showed that free trade was justifiable,’ Robbins said in 1938. Welfare economics was left feeling the cold.

It recovered by taking up the study of efficiency. A state of affairs is technically known as ‘efficient’ if there is no other state of affairs that is better for someone and no worse for anyone. You can say whether or not a state is efficient without making any comparison between the good of different people. So this is something that welfare economics was allowed to do. Concentrating on efficiency was seen by welfare economics as a way of purging their subject of unscientific ethical commitments.

Under certain (rather improbable) conditions, free trade is efficient. Because of this, there is no shortage of economists who think economic science shows that free trade is justifiable. They even think it shows that restrictions on trade are unjustifiable. They have slipped from identifying efficient states, which can be done without interpersonal comparisons, to claiming that efficient states are better than others, which generally cannot. Robbins was right about this. Without interpersonal comparisons they are not entitled to say, for instance, that free trade is better than a system of tariffs discriminating in favour of poor countries. Free trade is better than discrimination for the inhabitants of rich countries, but worse for the inhabitants of poor.

Welfare economics concentrated on identifying efficiency because, being denied interpersonal comparisons, it had nothing else to do. From that it slipped into pursuing efficiency regardless of other goals. The result of trying to avoid ethics, combined with some muddled thinking, was that economists found themselves making claims with strong and often indefensible ethical implications.

Today welfare economics is once again doing better. It is beginning again to take account of other values besides efficiency. It is interested in distributive justice, for instance. This, clearly, is its responsibility. Economic forces determine how the benefits and harms of economic progress are distributed within each society and around the world. And they do it desperately unfairly. It is shameful that the ethical branch of economics refused for so long to say anything about this injustice.

A leader in the return to responsibility and good sense is Amartya Sen. Sen’s ultimate motivation, I am sure, is a practical one. He is moved by a deep concern for India, for the poor of India, the women of India, and all the world’s poor. A great deal of his work is in practical economics. Poverty and Famines (1981), one of his most important books, is about the processes that lead to famine. Choice of Technique (1960) is about the degree of capital intensity appropriate for economic growth. He has worked in Bengal villages in order to compare the degrees of malnutrition of girls and boys. But Sen is also prepared to pursue the theoretical bases of his arguments as far as they need to be pursued. To attack poverty effectively you need to decide exactly what poverty is, what is wrong with it and what needs to be put right. And eventually you need to dig up and destroy the roots of a welfare economics that declines to address these questions. This will carry you to philosophy.

Sen has never acknowledged a boundary between economics and ethics. He brings philosophical arguments to bear where they are needed in economics, and combines them skilfully with formal analysis. He has also used the methods of economics to illuminate questions in philosophy. Economics embodies a lot of experience in manipulating systems of preferences, and Sen has applied this experience, for instance, to questions in political philosophy that have to do with mediating people’s conflicting interests. In the area between economics and moral philosophy, and in the foundations of welfare economics, he is the author of a long series of ideas and arguments each of which has stimulated a vast amount of discussion. On Economic Inequality (1973), to take one example, is still the most clear-sighted work on the theory of inequality. He has achieved such a dominant position that those of us who work between economics and philosophy find it hard to remember that this is only one part of his immensely influential body of writing.

In On Ethics and Economics, Sen describes the harm economics did to itself by running away from ethics. He mentions, as I have, the refusal to make comparisons between the good of different people, and the unreasoning obsession with efficiency that results. Sen argues that it was in any case a mistake to think that making these comparisons was an ethical activity. I am not so sure of that. When a change benefits some people and harms others, the gain to some has to be weighed against the loss to others. And ‘weighing’ here does not mean simply determining which quantity is the greater. It means determining which quantity should count more in judging the overall goodness or badness of the change. In fact, I doubt whether there is any useful distinction to be made in this context between a person’s gain or loss and how much that gain or loss should count in the overall judgment. In my view, comparing the good of different people is intrinsically evaluative – which is not to deny that it is factual. Lionel Robbins thought it was evaluative, too. As I see it, economics is more tightly bound up with ethics than even Sen suggests.

The view you take about this will depend on what you think a person’s good consists in. This is the central topic of The Standard of Living, a book consisting of two Tanner Lectures given by Amartya Sen in Cambridge, together with four commentaries by others and a reply by Sen. Tanner Lectures are normally published in volumes containing lectures on many disparate subjects. This little book makes Sen’s lectures much more accessible, and the commentaries are a valuable addition. The lectures are devoted to clarifying and analysing the notion of the standard of living. At first sight, this would seem to be an economic notion, referring to a person’s material conditions. Sen uses it more broadly, however. He includes within a person’s standard of living all the aspects of her well-being that have to do with her own life. He excludes only the good or bad feelings she might have that are caused, through sympathy or antipathy, by what happens to other people. His account of the standard of living is an ambitious one: not much less than a full account of what a person’s good consists in.

The account begins with what Sen calls ‘functioning’. These are the aspects of a person’s life, more precisely the ‘doings’ and ‘beings’ she achieves, that are valuable in their own right. Sen then defines a person’s ‘capabilities’ as the functionings she could achieve given the circumstances of her life, even if she chooses not to. If she has a bicycle, she has the capability of going out to meet friends, even if she actually stays at home. A person’s standard of living, Sen thinks, is an aggregate of her functionings and capabilities: both should count, but he is not dogmatic about the weight that should be attached to each.

If you are to define the standard of living this way, nearly all the work will be done by your initial selection of appropriate functionings. Which aspects of a person’s life are valuable? In setting up the apparatus of functionings and capabilities, Sen is offering, not so much an account of what a person’s good consists in, as a framework for an account, a framework consistent with any theory of good. If you think, for instance, that a person’s good consists in having good feelings, you can call having good feelings a ‘functioning’, and say it is the only functioning of value.

If you did that, Sen would not be pleased, however. He disagrees with the theory that good consists in good feelings, and argues against it in the book. In talking about functionings as ‘doings’ and ‘beings’, he is deliberately pointing in a different direction. Here are some of his examples of functionings: being literate, having a long life-expectancy, being well-nourished, being free from the risk of famine. Functionings are much more overt features of a person’s life than having feelings.

This is an important advance in economics. Economists have traditionally thought of a person’s good as something private: the feelings of satisfaction she gets from her consumption, or something of that sort. This is what opened the door to doubts about interpersonal comparisons of good. Imagine that good consists in feelings, and at the same time adopt Jevons’s wild scepticism about other people’s feelings, and you let in these doubts. An excellent antidote is to recognise that a great deal of a person’s good is not private at all. If someone has a house, a car and a job, she is, as a general rule (but not, for instance, if she is blind), better-off than someone who does not. If she is too mean to appreciate it, that does not make it any less true.

The material things a person owns do not themselves constitute her good. Owning a car does not do a blind person much good at all, because it does not help her to get about. The point of functionings is to take account of such circumstances. Functionings are what a person manages to achieve with the possessions she has. But they must be understood as public achievements like getting about, rather than as private achievements like having good feelings. A causal line extends from owning possessions, via the overt things you do with them, to feelings. Functionings are not too far down this line. Sen proposes that they, together with her capabilities, are what make up a person’s good. He includes capabilities because, he argues, a person’s capabilities reflect her freedom. They count as good because freedom is good. The standard of living, therefore, is a compound of functionings and capabilities.

Now, I like the idea of capabilities. And I agree that a person’s standard of living should be based on her capabilities, perhaps even to the exclusion of functionings or anything else. But my reason is different. I think Sen is wrong to conceive the standard of living as broadly as he does, as almost the whole of a person’s good. Our ordinary view of it is narrower: and it is this ordinary view which is at work in making capabilities seem the right basis for the standard of living. The ordinary notion of a person’s standard of living picks out what the rest of us – not the person’s family or friends, but us strangers – should be concerned about in her good, or at any rate what we have responsibility for. And what we should be concerned about is whether she is equipped to have a good life. Whether or not she takes the opportunity is her business. If a person has a house, a car and a job, if she is not blind, if she has no emotional handicap or other disability, then if she chooses to stay at home and mope, that is up to her. Certainly, her life will be less good. But I think her standard of living will be no lower. We are concerned with her capabilities: her functionings are her business.

This, I suggest, is why the standard of living seems to be an economic concept. It is not exactly that, however: Sen makes this point and he is surely right. It is nonetheless a concept designed for the economic or political sphere – the sphere of strangers – rather than the sphere of personal relations. This explains another thing. Although it is an important advance to realise that not all goods are private experiences, it is nevertheless true that some experiences are goods. All the same, we may not need to take account of experiences within the standard of living. A few days ago, the fountain outside my office was left flowing during a frost. For two days pinnacles and chandeliers of ice grew one on another, while water gushed from encrusted crevices. The sun turned all this into a tower of sparkling light. The sight made the biting cold worthwhile. I wish Jevons had been there, because I found I could easily tell that the other people who stood gazing at this radiant sight were enjoying it as much as I was. A few people were dull enough of soul to pass it by. Their lives are, I think, less good than they might have been, but their standard of living is no lower. This is because we strangers have no responsibility for their experiences: only for the more public features of their lives. Sen’s overt capabilities do not encompass all of a person’s good, but it is plausible that they are all that count in her standard of living. The standard of living is a practical concept, designed for economic and social action. He is right to connect it with capabilities and functionings. This approach allows it to be deployed easily and effectively to assess poverty. Among the poor it is chiefly a few very basic capabilities that are in question: to be able to eat, to be able to appear without shame in public, to have access to a newspaper. These are what the rest of us should be concerned with. I am less sure that this is the right way to understand a person’s good in general: but that is not Sen’s real purpose.

Sen has been developing his idea of capabilities for some years now. On Ethics and Economics displays the wealth of Sen’s ideas in the area between economics and moral philosophy. But it is a short book, and many of the arguments are too compressed to be clear. The Standard of Living offers a nice presentation of Sen’s viewpoint. John Muellbauer raises several interesting points, particularly about measuring living standards. Ravi Kanbur discusses how uncertainty should be accommodated within the notion of the standard of living. Keith Hart considers the effect on living standards of the spread of the market. Bernard Williams turns a searching eye on the concept of capabilities.

Sen has recently moved from Oxford to Harvard. This is a consequence of the decline that is being forced on British universities. It is also a significant part of the decline.