The End of Poverty

Over the fold, my draft review of Jeffrey Sachs The End of Poverty. Comments appreciated.

There are really two fundamental issues in development. The first, and biggest, is to determine what policies lead poor countries on to a path of self-sustaining growth, and how other countries can assist in this process. More than two hundred years after Adam Smith wrote The Wealth of Nations and fifty years after development economics became a recognised area of specialisation within economics, this question has not been answered.

There has been some progress. One analysis of the problem, arguably the most popular at the outset has been fairly conclusively refuted. This analysis drew on the seemingly successful experience of the Soviet Union and on concerns about the terms of trade for primary products to conclude that the optimal policy was a combination of autarky (no trade with other countries, or at least no trade with capitalist countries) and central planning.

The autarkist analysis has been turned on its head: the standard path to development, modelled on that of Japan, is now seen as beginning with labour-intensive export industries, such as light manufactured goods and tourism, followed by a climb up the skill ladder to more complex goods and services, based on imported technology, and then to domestically generated technological innovation.

A corollary of this has been general agreement of the importance of education, which was rechristened as investment in human capital in the early 1960s. By contrast, concerns about accumulation of physical capital have been downplayed. In part this is because the early stages of rapid growth are commonly characterised by high savings rates (contrary to what economic theory would predict) and in part because the standard path involves importing capital equipment in its early stages, by contrast with the Soviet Path in which the capital goods industry was the main initial focus.

A final point, obvious enough in retrospect, but not initially recognised is the benefits arising from the rule of law, and, conversely, the damaging effects of civil conflict. This point can be overstated, particularly in the work of writers like Hernando de Soto who focus almost exclusively on property rights and the ownership of capital. Nevertheless, in the absence of reasonable levels of security for people and property, there is less benefit in undertaking economically productive activity, than in seeking to appropriate wealth through the seizure of political power. Under such conditions, sustained growth is difficult to achieve.

All of this is significant progress, but it leaves fundamental questions unanswered. From an initial position of a poor economy, primarily reliant on subsistence agriculture, how does the process of growth get started? Should export-oriented industries be assisted, or is it better to adhere to a purist free trade policy? Should capital markets be more or less regulated than in developed countries? What role should governments play in health and education? Can wealthy countries help the process and if so, how?

There are no generally agreed answers to these questions. In particular, there is no generally accepted recipe according to which modest amounts of aid from rich countries can reliably generate self-sustaining economic growth. As a result, discussions of aid are commonly tinged with an air of hopelessness.

It seems more useful to focus on a second question, less intellectually ambitious, but more directly pressing. What can be done to alleviate, and if possible end, the suffering associated with extreme poverty, from famine to epidemic disease to chronic hunger and malnutrition?

In The End of Poverty, Jeffrey Sachs argues that extreme poverty can be ended in this generation, and that a feasible strategy to achieve this goal already exists, and needs only funding and political will to be implemented.

Before considering the arguments in detail, it is worth attempting to look at the problem in quantitative terms. The standard statistical measure of extreme poverty is an income per person of less than $1 US per day, as estimated using the Purchasing Power Parity method of the World Bank to compare incomes in different currencies and countries. An income between $1 and $2 per day is defined as moderate poverty.

People in extreme poverty are chronically hungry and lack access to basic health services and to elementary education. Under these conditions, the achievement of self-sustaining growth is almost impossible.

Of the world’s population of 6 billion, about 1.1 billion are estimated to live in extreme poverty, and about 1.6 billion in moderate poverty. While these numbers are huge, it is a striking fact that, for the first time in human history, the majority of people in the world are not, by these standards, poor, and only about one in six is extremely poor. This has profound implications for the feasibility of the task of ending extreme poverty.

Consider the total amount of additional income that must either be generated by poor people, or transferred to them by others, to lift them out of extreme poverty. To get an upper bound estimate, let’s ignore their existing income altogether. One dollar per day for 1.1 billion people is about $400 billion per year, a bit less than the US defence budget, and about 1 per cent of the income of the world as a whole.

Even this modest number overstates the difficulty of the task facing the wealthy countries of the world. Although poverty in East and South Asia has declined rapidly, a majority of the world’s extreme poor are still found there, with hundreds of millions in both India and China.

These countries are now wealthy enough, and growing rapidly enough to address the problem of extreme poverty primarily with their own resources (India will need modest assistance), though so far these growing incomes have been very unequally distributed. The defeat of the BJP government in the Indian elections of 2004 reflected the demands of India’s rural poor for a fairer share of the benefits of economic progress. China’s leaders will eventually have to respond to similar pressures. The main obligation of the wealth countries here is to resist pressure for protection against the exports of goods and services that are driving growth in East and South Asia

The more intractable core of the problem of poverty is in Sub-Saharan Africa and affects between 300 million and 500 million people. To raise the incomes of all these people out of extreme poverty requires additional income totalling no more than $200 billion per year. Sachs quotes the estimates of the Global Millennium project to suggest that the amount required from wealthy country donors is closer $100 billion per year, and could be achieved by gradually increasing aid between now and 2015, until the total contribution was about 0.7 per cent of GDP, an amount long-promised, but never delivered, by the rich countries.

To mobilise public opinion in favor of such a policy, it is necessary to overcome the well-known counter-arguments. These are, first, that large-scale aid programs have failed in the past, and second that aid is necessarily ineffectual because corrupt and dictatorial governments, particularly in Africa, will capture all the benefits.

As anyone who has followed the debate knows, the first argument is based on a false premise, namely that large amounts of aid have been spent with the aim of alleviating extreme poverty). Americans in particular are prone to the belief (which was in fact true during the Marshall Plan years immediately following World War II) that their government spends large amounts of money on foreign aid.

Sachs quotes a 2001 survey indicating that Americans, on average, believe that 20 per cent of the US government budget is spent on foreign aid (the actual figure is less than 1 per cent). The total amount of aid going to Africa from all sources is around $12 per person per year, and much of this is tied in ways that reduce its effectiveness. It’s scarcely surprising that there’s not much evidence of past aid efforts: how much evidence could we expect to see?

On the second point, Sachs offers a number of counterarguments. Corruption and ineffective government are almost inevitable in a poor country, but Sachs points to empirical evidence suggesting that, taking account of poverty levels, African countries are no more corrupt than others. In this context, it’s important to distinguish between the routine corruption and inefficiency that affects all governments to a greater or lesser extent and the extreme cases (for example, Sudan and Zimbabwe) where governments are, in effect, waging war against their own citizens. In the worst cases of civil war or brutal dictatorship, it may be impossible to implement aid programs at all, or it may be necessary to provide aid in ways that bypass governments altogether. But such extreme cases are the exception rather than the rule.

More generally, although governments may suffer from corruption and inefficiency, this does not constitute a reason for doing nothing. Even significant losses to corruption, and other problems like high transport costs, are not sufficient to offset the startling differences in the costs of effective medical interventions in poor and rich countries. A standard estimate for the cost of a marginal health intervention that would yield one extra year of life for one person in a rich country is $US50 000. The same amount spent in a poor country would yield health benefits hundreds of times greater.

The total expenditure required of the rich countries to finance the Millennium project is minimal in relation to other public priorities. An obvious comparison is the Iraq war, which has already cost $US300 billion. Even more striking is Sachs’ observation that the United States could finance its own contribution simply by repealing tax cuts recently granted to taxpayers with incomes in excess of $US 500 000 per year.

Ours is the first generation in which most people in the world are not poor. We can be, if we choose, the first in which no-one lives in extreme poverty. Sachs makes a compelling case for action. It remains to be seen whether governments, and the voters who elect them, will listen.

37 thoughts on “The End of Poverty

  1. The reference is

    William Easterly, The Elusive Quest for Growth: Economists’ adventures and misadventures in the Tropics, MIT, 2001

    It comes up first when you search for “Easterly” on under “Books”.

  2. I haven’t read the UN report yet. My initial question is, if as the authors says inequality is rising is this due to an increase in absolute poverty or because the living standards of the poor are improving more slowly than those of the rich?

    I agree that relative poverty is a problem but absolute poverty is a much greater one.

  3. I just glanced at the executive summary. This sentence stands out:

    >The proportion of the world’s population living in extreme poverty declined from 40 to 21 per cent between 1981 and 2001.

    Undoubtedly there’s much more to be done and there are many aspects of current development strategies that could be imporved but the incredible good news contained in that sentence seems to be ignored in the rest of the report and in much of the media coverage of it.

  4. What good news? That sentence, in isolation, doesn’t say whether the news is good or bad. We need to know if he adjusted for survivor bias. The thing is, there are two ways people can cease to live in extreme poverty – one good but not always very good, and one very bad indeed. Both affect that simplistic statistic. Did their lot improve, or did they die or get forced out of the area into even worse conditions elsewhere or something? That needs to be brought out one way or another.

  5. 19% of the world’s population was approximately 1,000,000,000 in 1980 and 1.2 billion in 2005 – I think we would have noticed death on that scale. (also you have to then assume that the net population growth of 2-2.5 billion over the period was after this mystery die-back.)

    and the “area” in question is the Earth – I think we might also have noticed the massive wave of alien abductions.

  6. You misunderstand me. I was not asserting death etc. on a massive scale, I was pointing out the leaping to a conclusion involved in using that statistic to get to that conclusion. You just can’t do it without bringing in some other information. As it happens, since ethics are not a mere matter of arithmetic, even if the “net” effect were an improvement, any material harm would make it bad news – even if it were wholly incidental, like the human suffering in the Highland Clearances.

    But I was mainly drawing attention to the gap in the chain of logic.

  7. PML, I think we can take as read your objection to the description of any event as “good news”.

  8. The problem with that is, you won’t actually look at the argument once you start to filter my own views out. Besides which, as and when I agree with an optimist, it is not my habit to jump in with an “I agree”.

    Would you like to address the material I put forward, rather than register the fact that I am out of step with the consensu around here? You are drifting close to playing the man if you don’t.

    I even left out the point that the money numbers don’t reflect what happens to people moving out of subsistence lifestyles, I only addressed the meaning of the single statistic in isolation.

    For what it’s worth, I would expect that statistic to conceal a large proportion of what just got called “good news” – but, also, a small proportion of real suffering as people got squeezed even worse, falling out of the statistical base and maybe even dying. So, what I was complaining about wasn’t the “good” or “bad”, but the failure to look more deeply into the statistic. And, of course, the full picture still needs an understanding that a subsistence peasant with $1 pw is better off than a slum dweller on $5 pw – but as I said, that’s outside this statistical measure anyway.

  9. PML, I thought the response from Ian Gould covered the substantive issues pretty well. You can’t fall out of the statistical base of “the world’s population”. And it’s obvious from the numbers that the decline in extreme poverty hasn’t been caused by increased mortality.

    I agree that you need to look carefully at the purchasing power of money, but there’s no strong reason to suppose that there’s a bias to overstatement.

  10. I see that I shall have to provide a full and organised reply to this whole thread.

    But for now, yes you can drop out of the statistical base of the world’s population, in two ways. One, by dying. And two, from the fact that the idealised “world population” is not being counted; when peolple are driven into slum dwelling extremes, for instance, they become much harder to count directly. That means that you have to use estimating techniques to adjust them, if you are using them to bring out the answers to questions like this one.

    For a current example, consider what the statistics will show about the poor people among the victims of Hurricane Katrina. Some will indeed be dead (and make the statics look better). Some will fall off the map and slip between the cracks. Given US approaches, some will wind up in jail after many vicissitudes, and although they will show up under some category or other, they won’t be classed as “extreme poor” any more.

    Now, all this should illustrate the underlying philosophical point I was trying to bring out, that the extreme poverty count is not itself a final result we can use as is to come up with firm conclusions like “good news”. I expect there is a lot of good news within there – but not unalloyed. Which is what I meant, that the statistic itself doesn’t tell the story.

    JQ, you were willing enough to accept that GDP doesn’t tell a whole story; why assume that this statistic does tell the whole story in its area? Particularly when we know that natural disasters do indeed shove poor people out of the system (as they do rich people, but they aren’t in this count anyway).

  11. As PML says, GDP figures don’t mean much by themselves. A nominal increase in GDP may simply reflect monetization of previously un-monetized activity.

    Driving people from the subsistence/gift/barter economy into the money economy may cause a significant jump in GDP, while reducing real standard of living. As I’ve pointed out elsewhere, when the Brits seized the best 20% of land in Kenya for settlers and drove off the native subsistence farmers, and used a poll tax to force other subsistence farmers into the wage labor market, the GDP probably went through the roof. Conversely, if sweatshop laborers homesteaded land currently used to produce cash crops for Cargill, and took to direct exchange through LETS systems and mutual banks, nominal GDP would probably go down considerably–even though real quality of life would soar.

Comments are closed.