The number crunchers

About ambiguities in the official poverty statistics of the world bank

Actually, it should be relatively easy to determine whether poverty in this world has increased, decreased or remained approximately the same in the last ten and twenty years. If you ask the world bank and the international monetary fund (imf), the answer is astonishing: they do not know.

Yet they go to such lengths. Data is collected, criteria are established, and the results are compiled and processed in elaborate procedures so that they are available to decision-makers in politics and business. The poor of this world may be poor, but their poverty is measured with the most advanced methods available.

Contrare results

The world bank, as professor angus deaton describes in his text is world poverty falling?? For the imf, the world bank is effectively the only organization in the world that collects and manages global poverty statistics. It is embarrassing that in the same text, professor deaton has to examine how two independent studies of the state of affairs come to almost contradictory conclusions at almost the same time. And this on the simple question of whether the number of people living on a dollar a day has grown or fallen over the last twenty years. Deaton describes the central riddle at the very beginning of his text:

The first table in the world bank’s 2000 world development report "attacking poverty" shows that the number of people worldwide living on less than $1 a day grew from 1.18 billion in 1987 to 1.20 billion in 1998, an increase of 20 million.

Barely two years later, another world bank publication ("globalization, growth and poverty: the path to a just world economy") states in a central listing that the number of people living in poverty fell by 200 million between 1980 and 1998 and showed no signs of increasing between 1987 and 1998. The decrease in poverty was later confirmed in a press bulletin, accompanying the publication of a text named "the role and effectiveness of development aid", which was published shortly before the un conference on development aid in monterrey, mexico in march 2002. In this bulletin it states: "in the last twenty years, the number of people living on less than $1 a day has dropped by 200 million, even as the world’s population has increased by 1.6 billion."

Can these statements be reconciled? Has there really been such a dramatic reduction in poverty in the last two years?? Or has the bank just changed its interpretation of history??

[all citation translations mh]

Data problems

Although prof. Deaton tries hard to explain the inconsistencies in the following text, he has to admit in the end that his interpretation of the measured values is subject to gross uncertainties. While he concludes that the very positive, not to say euphoric, values reported could be basically correct, he expresses his concern about the basic quality of the data material by writing:

Much about this process can make one uncomfortable. This is most clearly expressed in the compulsion to rely on data whose margin of error over a period of a mere two years allows for such starkly different views of what is going on with poverty in the world.

Deaton’s discomfort is not surprising, since he is struggling all the time with incomparable data sets (incomparable because of significant changes to the. Significant changes in the questionnaires during the period of comparison), long-term estimates based on uncertain methods, a fundamental limitation of his considerations to india (because of. Its fundamental importance for the world poverty figures) and others. In the end, the dumbfounded reader has the impression that the years of effort of a luminary in the field of economic statistics have gone into the result that, unfortunately, it is not possible to say anything precise. Genuine despair shines through in deaton when he recommends that the world bank be more careful with its data:

The very success of collecting this data and its importance to policy-making processes calls for improvement in the methods of obtaining and maintaining it. If such reports continue to contradict each other, the world bank risks losing the ability to monitor its own success.

Where is globalization headed??

This could now be an isolated case, but is confirmed by another paper by b. Milanov, who also discusses wildly divergent poverty estimates in a short paper for the world bank. In it, milanov argues that the calculations of andrea boltho and gianni toniolo, who in a december 1999 ie of the oxford review of economic policy reported significant improvements not in absolute poverty figures but in inequality ("inequality") between the rich and the poor worldwide.

The claimed fall of the so-called gini coefficient by four percentage points between 1980 and 1998 is an illusion, in fact three recent studies (schultz, firebaugh and his own) show that the gini coefficient is 10-20 percent higher than amed by boltho and tonioli. In addition, milanovic has been able to prove that the international gini index increased by three percentage points between 1988 and 1993.

Milanovic also cites the conclusions of a paper by m. Lundberg and l. Squire (the simultaneous evolution of growth and inequality", pdf), which found the following on behalf of the world bank: globalization is harming the poor of this world, it was explicitly stated there. Milanovic writes:

In a recent world bank publication, lundberg and squire claim that trade deregulation negatively affects the income growth of the 40 poorest percent of the population, while providing a strong positive stimulus to the income growth of the rest of the population (based on sample data from 38 countries between 1965 and 1992). Die anpangskosten von handelsderegulierungen werden ausschlieblich von den armen erbracht, egal wie lange die anpang auch brauchen mag. Die armen sind verwundbarer fur schwankungen im internationalen preisgefuge, und diese verwundbarkeit wird durch eine offnung des binnenmarktes gegenuber dem weltmarkt noch erhoht.

It could also be said that globalization, as understood and practiced today, makes the poor poorer and the rich richer. It goes without saying that lundberg’s and squire’s paper has opened doors for anti-globalization activists. It is often quoted in anti-globalization publications and is accordingly popular. Who is right? Are the optimists closer to the truth or the pessimists?? B. Milanovic calls for "greater conceptual clarity" and "more fundamental empirical investigation". For him, one thing is certain:

First, global income inequality is much more massive than reported by boltho and toniolo, and recent trends tend to point to a widening of inequality. Second, the positive effects of globalization do not benefit everyone equally. It might be possible to convince the anti-globalization movement of the positive effects of an integrated world market, but this would require arguments based more on facts and less on ideology.

How this is supposed to succeed when the world bank doesn’t even know what the numbers it collects mean remains his secret. For those who in any case do not have access to the data of the rough institutions and the methods of their processing, there is probably nothing left but to stick to the indications.

The increase in worldwide migration movements, the efforts of the rich countries of the north to militantly fend off all impoverished refugees who cannot be utilized, and an ever widening income gap in the rich countries themselves do not exactly indicate that poverty in the world is decreasing. Under these circumstances, it is unlikely that fact-based debates will be able to convince the opponents of globalization of the beneficial effects of globalization.