African poverty is falling…much faster than you think

Xavier Sala-i-Martin, Maxim Pinkovskiy

06 December 2010



The picture of Africa as a place of collapse, hunger, disease and death is slowly fading. Both official statistics and the popular press acknowledge a nascent “African Renaissance”, as the continent is enjoying its longest and strongest growth spurt since independence.

Nevertheless, it is still widely believed that this growth is primarily driven by oil and natural resource prices, and that it is confined to well-connected elites in geographically advantaged countries. The popular image is that the poor majority in all African nations and many African nations as a whole are stuck in “poverty traps” created by unfortunate geography and calamitous history. For example, the prospects of meeting first Millennium Development Goal of “halving, between 1990 and 2015, the proportion of people earning an income less than $1 a day” seem to appear bleak for Africa; the UN writes in its latest Millennium Development Report that “little progress was made in reducing extreme poverty in sub-Saharan Africa” (UNDP 2008).

We disagree. The sustained African growth of the last 15 years has engendered a steady decline in poverty that puts Africa on track to meet the Goals by 2017. If peace is established in the Democratic Republic of Congo, and it returns to the African trend (which is what happened to other African nations that were formerly at war), Africa will halve its $1/day income poverty rate by 2013, two years ahead of the 2015 target.

Moreover, African poverty reduction has been extremely general. Poverty fell for both landlocked and coastal countries, for mineral-rich and mineral-poor countries, for countries with favourable and unfavourable agriculture, for countries with different colonisers, and for countries with varying degrees of exposure to the African slave trade. The benefits of growth were so widely distributed that African inequality actually fell substantially.

Measuring African poverty

In recent research (Pinkovskiy and Sala-i-Martin 2010), we use the methodology of our previous paper (Pinkovskiy and Sala-i-Martin 2009), to combine the standard Penn World Tables GDP series with a comprehensive inequality database to estimate African income distribution for the period 1970-2006. For countries and years with inequality data, we compute the distribution of income by fitting a lognormal distribution to the inequality data, whereas for countries and years without inequality data, we interpolate inequality on the basis of neighbouring years. If a country has no inequality data for the sample period, we interpolate on the basis of the average inequality of countries with inequality data. 

Figure 1 presents our main result:

  • Using the $1/day definition of poverty adopted by the Millennium Development Goals, African poverty declined strikingly, from 41.6% in 1990 to 31.8% in 20061.
  • Poverty seems to co-move with GDP almost perfectly.
  • African inequality has also fallen over this period, almost entirely reversing its rise since 1970, but still remaining at a high absolute level (Figure 2).

Figure 1. African poverty and growth

Figure 2. African inequality

Thus, during the period of positive and sustained African growth (1995 to 2006), not only did inequality not fail to explode as would have been the case if all the growth went to a narrow elite, but it actually declined substantially.

Our estimates of African inequality allow us to measure African welfare, e.g. by Amartya Sen’s (1976) index of GDP-per-capita x1 minus the Gini coefficient. African welfare declined substantially between 1970 and 1995, but the trend was reversed dramatically between 1995 and 2006 (Figure 3). During this decade however, the two components of the index moved in the same direction. Mean income increased and overall inequality declined. Hence, African welfare improved.

Figure 3. African welfare (using the Sen measure)

Implications for the Millennium Development Goals

Tables in our working paper show the expected decline in poverty by 2015 if present trends continue, and the expected date of attainment of the goals for the baseline estimation method and multiple variations to the estimation procedure. We consider alternative methods of extrapolation, distributional assumptions other than the lognormal, an “adjusted” method of fitting distributions to the data that makes use only of data in the middle of the income distribution that should be less affected by survey misreporting, recovering the distribution by inverting the Gini coefficient, and using other sources of GDP besides the Penn World Tables. The data compiled by Angus Maddison and the World Bank’s calculations of GDP after its revision of purchasing power parity estimates in 2007.

We also consider what happens if any of the eight largest African countries is dropped from the analysis. We see that Africa will probably halve poverty relative to 1990 sometime between 2015 and 2020, with the baseline estimate being 2017; a few years late relative to the Millennium Development Goal target. However, being a few years late to achieve the goal is much better than not making progress towards it at all. The main point is that Africa has been moving in the right direction and, while progress has not been as substantial and spectacular as in Asia, poverty has been falling and it has been falling substantially. We should not let the literal interpretation of the Millennium goals turn good news (Africa is rapidly moving in the right direction) into bad news (Africa will not achieve the goals on time) (Easterly 2009).

One reason why the Millennium goals are projected to be achieved several years late is the poor performance of the Democratic Republic of Congo (DRC) over the last decade. Naturally, this poor economic performance has to do with the war that took place in that country during that decade, and which is now drawing to a close. If we exclude the DRC from our baseline computation, Africa halves poverty by 2012, three years ahead of time. (For more extensive robustness checks for the scenario without the DRC, see our paper; these show that for most of nations, the MDG is actually achieved ahead of the 2015 target date.)

Why Africa is held back

It has often been suggested that geography and history matter significantly for the ability of Third World, and especially African, countries to grow and reduce poverty. Collier (2006) argues that coastal countries, or countries that are mineral-rich, will perform better than landlocked and mineral-poor countries in general. Bloom and Sachs (1998) point to adverse geography as a cause of slow development: in particular, countries that have unfavourable agriculture should be poorer than countries with more favourable conditions.

Others have suggested that troubled history may have a persistent effect on growth performance. Nunn (2007 and 2008), for example, argues that the African slave trade had “particularly detrimental consequences, including social and ethnic fragmentation, political instability and a weakening of states, and the corruption of judicial institutions,” which led the parts of Africa most affected by the slave trade to grow much slower than the parts that were not. La Porta et al. (1999) suggest that the identity of the coloniser mattered substantially for development. Since these factors are permanent (and cannot be changed with good policy), they imply that some parts of Africa may be at a persistent growth disadvantage relative to others.

Yet Figures 4-9 show that the African poverty decline has taken place ubiquitously, in countries that were slighted as well as in those that were favoured by geography and history. For every breakdown discussed above, the left panel of the corresponding Figure shows GDP in countries to each side of the breakdown, while the right panel shows poverty rates. While the levels of the poverty series start out matching the hypotheses set out above, the poverty rates for countries on either side of the breakdown tend to converge, with the disadvantaged countries reducing poverty significantly to catch up to the advantaged ones. Neither geographical nor historical disadvantages seem to be insurmountable obstacles to poverty reduction.

Figure 4. Landlocked vs. coastal countries

Figure 5. Mineral-rich vs. mineral-poor countries

Figure 6. Landlocked vs. coastal mineral-poor countries

Figure 7. Favourable vs. unfavourable environment for agriculture

Figure 8. Effect of slavery

Figure 9. Colonial origins


Our main conclusion is that Africa is reducing poverty, and doing it much faster than many thought.

  • The growth from the period 1995-2006, far from benefiting only the elites, has been sufficiently widely spread that both total African inequality and African within-country inequality actually declined over this period.
  • The speed at which Africa has reduced poverty since 1995 puts it on track to achieve the Millennium Development Goal of halving poverty relative to 1990 by 2015 on time or, at worst, a couple of years late.
  • If the Democratic Republic of Congo converges to the African trend once it is stabilised, the MDG will be achieved by 2012, three years before the target date.

We also find that the African poverty reduction is remarkably general.

  • African poverty reduction cannot be explained by a large country, or even by a single set of countries possessing some beneficial geographical or historical characteristic.
  • All classes of countries, including those with disadvantageous geography and history, experience reductions in poverty.

This observation is particularly important because it shows that poor geography and history have not posed insurmountable obstacles to poverty reduction.

The lesson we draw is largely optimistic – even the most blighted parts of the poorest continent can set themselves firmly on the trend of limiting and even eradicating poverty within the space of a decade.


Bloom, David E, and Jeffrey D Sachs (1998), “Geography, Demography, and Economic Growth in Africa”, Brookings Papers on Economic Activity, 2:207-273.

Collier, Paul (2006), “Africa: Geography and Growth”, Journal TEN, Federal Reserve Bank of Kansas City, Fall.

Easterly, William (2009), “How the Millennium Development Goals are Unfair to Africa”, World Development, January.

The Economist (2010), “Uncaging the Lions”, 10 June.

La Porta, Rafael, Florencio Lopez de Silanes, Andrei Shleifer and Robert Vishny (1999), “The Quality of Government”, Journal of Law, Economics and Organization, 15(1):222-79

Nunn, Nathan (2007), “The Historical Origins of Africa’s Underdevelopment”,, 8 December.

Nunn, Nathan (2008), “The Long-Term Effects of Africa’s Slave Trades”, Quarterly Journal of Economics, 123(1):139-176.

Pinkovskiy, Maxim and Xavier Sala-i-Martin (2009), “Parametric Estimations of the World Distribution of Income”, NBER Working Paper 15433

Pinkovskiy, Maxim and Xavier Sala-i-Martin (2010), “African Poverty is Falling...Much Faster than You Think!”, NBER Working Paper 15775.

Ravallion, Martin (2010), “Is African poverty falling?”,, 3 May.

United Nations (2008), The Millennium Development Goals Report.

World Bank (2004), “Millennium Development Goals”.

1 Martin Ravallion (2010) argues that the poverty line should be defined using consumption rather than income, and that greater weight should be placed on poverty counts rather than poverty rates. However, we consider our definition of poverty is consistent with the one used by the Millennium Development Goals. See our response to Ravallion here.



Topics:  Development Poverty and income inequality

Tags:  development, Africa, Poverty

J. and M. Grossman Professor of Economics of Economics, Columbia University

Economist in the Research Department, Federal Reserve Bank of New York