Is African poverty falling?


This page in:

Maxim Pinkovskiy and Xavier Sala-i-Martin (PSiM herafter) have confidently claimed that “The conventional wisdom that Africa is not reducing poverty is wrong” and that “African poverty is falling and is falling rapidly.” This sounds like good news. But is it right?

We must first be clear about what we mean when we say “poverty is falling”. What many people mean is falling numbers of poor. However, PSiM refer solely to the poverty rate—the percentage of people who are poor. (There is no mention of this important distinction in their paper.) And it is not falling over their whole period of their analysis, which goes back to 1970. Rather they find that the poverty rate has been falling since the mid-1990s.

Here we agree: aggregate poverty rates have fallen in Sub-Saharan Africa (SSA) since the mid-1990s.  Shahoua Chen and I came to exactly the same conclusion in our research, for the World Bank’s global poverty monitoring effort, although our methods differ considerably and (no surprise) I prefer our methods.

However, Chen and I also point out that the decline in the aggregate poverty rate has not been sufficient to reduce the number of poor, given population growth. That is important.

PSiM obtain lower poverty rates than us and they appear to have a somewhat steeper decline. These differences reflect the different methods.

Two points to note here: (i) Chen and I show that the poverty decline in SSA tends to be larger for lower poverty lines (in the region $1-$2.50 a day) and (ii) PSiM’s method attributes the entire difference between GDP and household consumption to the current consumption of households, and they assume that its distribution is the same as in the surveys. These assumptions are very unlikely to hold, and they give an overly optimistic picture.

In effect, PSiM are using a lower poverty line than us.

I did a check using PovcalNet. If one uses a line of $0.90 a day then one gets a decline in SSA's headcount index from 42% in 1996 to 34% in 2005, which looks very similar to their results using "$1 a day".

The number of poor in SSA living under $0.90 a day rose over this period, though only slightly (252 million in 1996 and 256 million in 2005). This, too, is consistent with the results of Chen and Ravallion, though higher poverty lines tend to show larger increases in the aggregate poverty count.  

Another important difference is that Chen and I are more cautious about the data limitations. There are not enough good household surveys available yet to be confident that this is a robust new trend of a falling poverty rate for SSA. PSiM are not so restrained, as is plain from their title!

PSiM do not tell readers just how few survey data points they have actually used after 1995. Indeed, readers of their paper may be surprised to hear that there is any uncertainty about the trend decline since the mid-1990s; their main graph has 30 annual data points since 1995. But these are not real data points in any obvious sense; rather they are synthetic (model-based) extrapolations based on national accounts and growth forecasts.

We have national household surveys for all but 10 of the 48 countries in SSA since 1995. However, for only 18 countries do we have more than one survey since 1995; for 30 countries, there are is at most one survey since 1995.

As we warn explicitly in our paper, this is not yet sufficient survey data to be confident about the (promising) downward trend for Africa’s aggregate poverty rate that PSiM have announced with such confidence.

Hopefully we will see a confirmation of the emerging downward trend for Africa in the years ahead, as more (genuine) data emerge. 



Martin Ravallion

Martin Ravallion, Edmond D. Villani Professor of Economics, Georgetown University

Omololu Fagbadebo
March 08, 2010

Poverty in Africa is not because of the derth of resources but because of leadership curse. As long as the corruption-invested leadership and their western capitalist collaborators hold sways in the corridor of power there can never be a falling in the poverty level in the continent. African leaders are corrupt and morally bankcrupt. They care less about the welfare of the people. They are self centred and that explains why policy failures pervades governance.
In practical terms, there is no poverty reduction rather there is poverty agravation

thandi tshabalala
March 09, 2010

slowly but surley i think that africa will come out of the state of distress but, with the current wars going on, can we really say the 'poor peoples' rate is declining? i think not because there are countries who don't even have a functioning government thus we cannot already conclude africa's state of poverty already. i think this researcg focused alot on the countries that are rapidly developing and 'coming out' of thier state of turmoil, disaster and poverty...what about the countries that have been in war since forever or have never really been able to get out of their disasterous situations? yake a country like somalia for example. how can a counrty like that be helped externally and internally?

sanguv simon peter fomonyuy
March 09, 2010

it is true that, poverty reduction trategies have been put in place to reduce african poverty. However, the decline in the aggregate poverty rate has not been sufficient to reduce the number of poor, given population growth, and mismanagement of development funds. thus african poverty is not falling

Dr. Florence Kithinji
March 16, 2010

Hello all,

What is our going definition of povery here? Is it merely in economic terms? Is poverty not also in relation to political, social and technological aspects of growth?

What are the non- quantifiable aspects that define poverty/ I think we need to look at those really marginalised as the baseline from which to study the probable reduction in poverty in Africa. There are no absolutes.

Dr. Kithinji

Nicholas Ngepah
March 16, 2010

I think the debate over African poverty is good at this time in order to wake some institutions from sleep and perhaps redirect attention towards African development after a while of concern with economic melt-down. To say that African (monetary)poverty is decreasing or would even decrease in the nearest horizon in real terms is difficult to accept. In nominal terms this may be possible. There are various reasons
First, one can agree with Martin Ravallion that the use of lower poverty line underlies the apparently decreasing African poverty. This is even quite serious given that the rate at which prices of basics (energy, food, transport, housing etc) increase is hardly taken into account in evaluating monetary poverty. The poverty line cannot continue to stay at one or two dollars per person per day. Secondly, the dollar value has been falling, and with US foreign debts and uncertainty of China's possible moves, the dollar may even depreciate further in the future. As such, a dollar-based poverty line may not be an appropriate yardstick for poverty evaluation. In fact, someone who is poor by one dollar today is actually poorer than another who was poor by one dollar in 1990.
Even though Africa’s real growth per capita has increased to 2.1 per cent from2000 to 20005 (African Development indicators, 20007), one has to question the sources of growth. There are suggestions that growth in (most) African economies is sustained more by high commodity prices than good management practices. In corrupt countries with centralized governments, such growth is even unlikely to trickle down to the poor. However, there are a few countries in Africa that have shown evidence of good management, e.g. Ghana, but it may still be questionable whether with good governance, Africa has a level ground with others to sustain growth. Can Africa goods compete with Chinese goods? Can Africa have equal opportunity in the acquisition of market shares especially in developed countries? We seem to be entering an era where (export led growth) may be a zero sum game. Think of the Doha rounds!!!! On these bases, given that poverty reduction is the combined fruits of sustained economic growth and inequality reduction, the future as regards African poverty is still somewhat bleak.
One may have to only believe (that an invisible hand may do something) in order to accept any significant future reduction in African poverty in real terms.

Hudson Lucky Masheti
March 10, 2010

Compare this case with my small village where I hail from;

We understand that the information on African poverty is falling and is falling rapidly is a statistical analysis not house-to-house (door-to-door) feasibility.

In my primary education, we were about 40 students. Only 15 made it to secondary school, 9 made it to university and or colleges, less than 6 secured good jobs. This shows that 34 are casual workers. How then is poverty falling with such indicators?.

A farmer in my village planted a half an acre tea plantation while single in early 1940s. He got married to two wives and has 15 children that are not well educated though only one boy is educated and one call married to a working class man. Boys are now married with children and girls have children but not married, still depends on their parents with their children for survival.

For sure can this two case studies indicate that African poverty is falling? I can say that by statistics- it is falling but door-to-door feasibility it is on the rise.

It is true that it sounds like good news. But it is not right. For much needs to be done by the Africans themselves eg; by transforming their AID depedency mentality, leadership culture, political interests, lure of gain, policy structures to name but a few and set up developmental structures and partnerships by learning to conserve for the future generation.

March 08, 2010

Presently all country leaders have selfish, so they can't reduce the poverty. This condition have in African country, when will come to unselfish leaders / people to service, that time will be reduce the poverty otherwise not possible. World bank also giving lot of money, but no use to their. So first, we will create the service mind leaders after that we can change everybody. Thanks.

Daniel Altman
March 08, 2010

It is certainly worth noting that the population of the poor may not be shrinking in sub-Saharan Africa even if poverty rates are going down. But what if we saw the same thing in the United States - would we react the same way?

If poverty rates were falling in the United States, but the absolute number of the poor were not, it probably wouldn't be big news. Take an example from a statistic we hear about much more often: unemployment. There were about 7 million unemployed people in the United States at the beginning of 1978. There were also about 7 million unemployed at the end of 2007. At the beginning of 1978, the unemployment rate was 7.1 percent. At the end of 2007, it was 4.8 percent. So, was the United States better off in 2007, even though the absolute number of unemployed people was the same?

I think most Americans would say yes. The question is, when will we begin to evaluate sub-Saharan African countries the way we evaluate the United States? How far does the absolute number of poor people have to fall in sub-Saharan Africa before we switch to looking at poverty rates? Or should we start looking at them now?

At the very least, we should justify using one measure or another. Perhaps we can say that 1) a high absolute number of poor continues to represent a significant human tragedy, but also that 2) the fact that this number has not risen even as the region's overall population has grown is a success. A little more explanation, on the part of all the authors involved here, would be welcome.

George Amadi
March 07, 2010

I find this commendable on the part of Ravallion. A pat on the back here, a good word there (they need not be unassailable) can bolster African Leaders' self-esteem -- an ingredient currently lacking in their approach towards economic development and transparency, on thanks to Western media organisations,

Otabor Isaac
March 09, 2010

The various national governments in Africa can do more in reducing poverty in their various countries,by strategically implementing policies that fight against poverty.

Meanwhile,if poverty in Africa is reducing,then the number of poor people should reduce in Africa.But,what we experience in Africa is that the rich are getting richer, whereas,the poor are getting poorer.

Finally,there should be more serious commitments by African Governments in particular and Africans indeed,in fighting poverty.

Jean de Dieu Bakundukize
March 16, 2010

Yes, Africa can end the poverty but is going to take many years due to the development policies in Africa are not constructive and reproductive as in other continents. We need to improve the policies formulations which are poverty pro poverty alleviation and also to make sure that there is no more corruption in bureaucracy systems, even the developed countries they should to pay the leadership of developing countries trough aids. It is kind of congruating of wrong doing of poverty creation and generation in Africa.

John Burton
March 12, 2010

We need much better data on African poverty and global poverty. Many debates about the links between growth and poverty could be resolved with evidence rather than complicated data manipulation. The fact is that we don't have good enough information. Martin Revallion tells us that only 18 out of 48 countries have more than one household survey since 1995. 10 have no survey and 20 have one survey. How can we operationalise the MDGs on this basis? - it is like driving the car at night with no lights on a dark road. Surely it would be in everyone's interest to have a new survey at least every 5 years, using a consistent methodology so that we can compare results accross time and accross country. What are the World Bank and other donors doing to fill the data gaps?

Rodney Atuhaire
March 16, 2010

To suggest that poverty is falling in Africa, something must be wrong with those indicators. When I think about what my home country men (Uganda) in the rural areas have to go through to survive this research must be seriously flawed. All indicators point to the fact that our leaders are ruining our economy to the extremes that conventional means of measuring poverty become irrelevant. There is so much corruption that so called business can afford to put up public structures with no one renting the facilities and still remain economically sound. yet we have a wanainchi in the village who cant afford a 5-litre jerrican of water. Rather resorting to unconventional means of supporting his family aka roadside robberies.

Give us a break and comeup with another meaningful research topic if this has not yielded common sense.

John Chiwuzulum Odozi
March 16, 2010

the phrase "poverty is falling and falling rapidly in Africa" defines the progression of poverty over time in terms of the number of poor people and the percentage of those that are poor. Does the phrase sound right?

Martin Ravallion and Maxim Pinkovisky seem to agree that the percentage of poor people(poverty rate) has been declining since the mid 1990s and also agree that the decline has not been sufficient enough to reduce the number of poor people in Sub-Saharan africa because of population growth. In essence given population growth, poverty in terms of number since the mid 1990s can be said to have remained stagnant.

However, both seem to disaggree in terms of methods used in reaching their conclusion. this a major headache in Economics. And the solution in my opinion is to be honest enough to declose limitations and to conclude with caution. second it is important to back up analysis with specific country case studies.

March 16, 2010

How do you measure poverty? Do you consider both birth rate and death rate? if the birth rate is increasing more than the death rate, then poverty is increasing rather than falling. The population figure used in calculating per capita are sometimes estimated population figures.

March 18, 2010

Academically, the Chen Ravallion analysis appears to give a fair picture, because it takes into account population growth (assumably population decline too). Not only that, the PSiM uses lower poverty line. However, we at ACRO have a different line of thinking which is based on common sense and natural justice, or perhaps better said, based on present realities as far as Africa, SSA, ic concerned.

It is basically wrong for Africa (SSA, to use your designation) to be poor in the first place, because nature had endowed it with abundance both in terms of natural-material and human resources. But those with might and power had pillaged it openly at first and continue to do so systematically at present.(Africa's human resources was pillaged by slavery and its natural resources by colonialism, neocolonialism and unfair trade etc. Africa, SSA, exchanges usual natural resources and food items for weapons and ammunitions and in most intances its ruling elite transfer the country's resources to be kept in Western and Eastern banks).

It is ACRO's view that justice, democracy and the rule of law are very cardinal to fight poverty and improve equality among the peoples of any African nation, any nation of the world for that matter. Once these conditions prevail in Africa,ie. SSA, with the complete exclusion of any heavy-handed external actor of any form or nature that plays a negative role in the process, there could be tremendous transformation in Africa towards betterment. A greater role could also be created for the WB and the UN to contribute positively in this respect.

Finally, there is what we call in ACRO, the tolerable income gap, that should be maintained in SSA in particular and the developing countries in particular. For instance, using your study as $1 - $1.50 a day for the most minimum, the income gap between the poor and the wealthy SSA businessman/woman, politician, minister etc. could be calculated by deducting the former figure from that of the latter. How much is the gap? 10, 20 or 1000 times? And how much should be tolerated? Here, the idea is not to discourage business but institute fair means of taxation, and a matter of finding a means of retrieving stolen resources from the corrupt politicians, and the like as well as practising equity and justice in the remuneration system both at the private and public sector. In this line of thinking and reasoning, many an anomality could be tackled in each and every African country, SSA.

March 10, 2010

Absolute poverty is decreasing but the gap between the poor and the rich also is increasing. I am also of the opinion that the definition of poverty in SSA should be properly and better defined.

March 19, 2010

It seems that both groups of authors agree that the poverty in SSA is on the decline, at least relatively speaking. When we take a broader measure of poverty ($2.00 a day) we see according to Ravallion and Chen that the decline went from 76.2% (1990), to 77.9% !(1996) to 73% (2005). When we assume a margin of al least 5% the decline does not look very impressive or stable or certain. The high level of poverty posses the question whether we really can speak about a possitive development.

When we look at the absolute numbers we see an increase (as Ravallion correctly menstions). For SSA from 393.6 million in 1990, to 471.1 million (1996) to 556.7 milion (2005). That is an increase of 41% in absolute numbers since 1990.

When we look at the mean consumption of the poorest of SSA (below $1.25) we see that their consumption has increased from 2002 to 2005 with about 1%. (What margin?) The average increase of the GDP per capita was 2.5%. (Margin?)

All this taken together the conclusion based on theese numbers should be that absolute poverty is on the rise, and the relative poverty of the poorest compared with the average income is also on the rise.

The other team of Columbia comes with the good news. However, we see that they used time series of PPPs, that they ignore the fundamentals of inequality and make use of GDP data that certainly cannot be trusted. Considering this it is not likely that their analysis can be called sound. The data limitations in Africa are such that each figure needs to be treated very carfully. Many figuers taken together even more carefully.

It seems reasonable to assume that income is underestimated all over Africa. The more is earned the more will be underreported. Therefore the levels of poverty can be lower as estimated by Ravallion and Chen, the relative share of the poor may also be lower, the consumption may be higher but there is no reason to speak about a promising development. Fact is that we do not have data of sufficient quality to make strong statements. Assuming you have is fooling yourself.

Moreover, the total fertility rate in Africa is not coming done rapidly enough (TFR is about 5). This means that behavior in Africa is not changing. That is bad news for the future. Since we cannot trust the economic data it is possible that we will face an increase of poverty in the coming years. Certainly in absolute numbers, possibly also relatively. Even if we do not have the tools to measure it.

Olofin, O.P.
March 17, 2010

Yes, I agree that Africa can end poverty but this will take place in a very long period in as much as the attitude of self interest of Africans in general as well as their leaders persist. From the look of things we can say poverty is reducing in African Countries even by using common sense of inferences. For example, income in general is increasing and there is no way a man will embezzle countries money and no part of it will be of any benefit one way or the other to anybody in the society. However, there will be increase in the gap between the embezzlers (the rich) and the subjects (i.e. those who dont have the opportunity to embezzle).

PSiM have done great job, but the problem is that of accurate data to determine the reliability of their findings. I,m happy they also noted this in their analysis. Did they consider at least very large number of multidimensional aspect of poverty in African countries? I think if poverty is reducing, the number of the poor should also decrease since those who are getting off the hook of poverty will not want their children (i.e.growing population) to fall victim by making necessary provisions for them and for others who are still trapped. Poverty eradication is what Africans can stop on their own if they decide to shun self interest at least everyone knows what is good for him and if he can try as much as possible to let others enjoy some of the things he enjoys, poverty will be minimised.

April 05, 2010

I challenge the readers that poverty may not be falling in Africa. I don't have data on other African countries other than Mozambique, so I restrict my comments to Mozambique. The country is steadily growing at about 8 percent/year, yet is ranked 172/179 in the Human Development Index. The country has reported a dramatic decrease in poverty from 69 percent in 1996 to 54 percent in 2002/03. But between 2002/03 and 2007/08 things have gotten worse: Caloric availability has decreased, access to extension services has also decreased amid the high fuel prices, adoption of improved technologies has also deteriorated, fewer farmers are receiving price information, and poverty has not decreased. This raises additional questions. Was the remarkable reduction in rural poverty observed between 1996 and 2002 a permanent change in poverty, or just a transitory change? Will Mozambique achieve the MDG of reducing poverty if the cropping season of 2014/2015 is not favorable? If vagaries of the weather are driving most of the changes in rural poverty, then what are the farmers’ prospects with the threats of global warming?


April 05, 2010





March 29, 2010

Thanks for the points raised. I think the point is well made that the PSiM's work cannot be trused.

How much can be believed of the paper of Rvallion and Chen? When I use the data from their own tables I can create the following picture.

Table 1.
Poverty groupings as a share of the total of the poor based on USD 2.50 for SSA (1990-2005)
$1.00 $1.25 $2 $2.50 of
total SSA
1990 57 70 92 100 is 82.5
1996 56 70 92 100 is 84.2
2002 53 67 92 100 is 82.5
2005 50 64 91 100 is 80.5

It is save to say that the relative poverty of the group of USD 2.50 and below has not really changed. It is above 80% of the SSA population. The only changes are at the lower end. But even these changes are not very impressive.

Look at the absolute numbers.
Table 2.
Changes in the absolute numbers: each year compared with 1990
$1.00 $1.25 $2 $2.50
1990 245 299.1 393.6 426.4
1996 17 19 20 19
2002 27 30 36 37
2005 24 31 41 44

Here we see that the group of the poor has grown with 44% since 1990. The group below USD 1.00 with 24%. At some point the absolute numbers are far more relevant than relative numbers. Poverty is absolute.

Of course the truth is that when we do not capture the informal sector and informal labor activities we do not understand Africa. A
coordinated and harmonised approach is needed to measure these issues. This takes careful investments. To be able to measure with statistcal methods the total income in SSA however does not seem to be realistic. Alternative methods need to be considered.

When we look into the future we know that Africa will grow in the next 40 years explosively. Probably with 100% or more. It is easy to predict what is needed to deal with this. It will be extremely difficult to realise these efforts. These are some of the reasons to be extremely careful with painting rosy pictures about the SSA.

Juniours Marire
March 26, 2010

Thumbs up to PSiM for a compelling work! One striking and coincidental feature of their analysis is that Structural Adjustment Programmes (SAPs) and the Washington Consensus (WC) have unleashed unparrelled growth and poverty reduction in Africa. Why? Because the 1990's constitute the greater part of the SAPs implementation period especially for Sub Saharan Africa. Is it really true that SAPs have not only miraculously unleashed Africa's growth potential but also that the growth was (is) pro-poor? Then we might need more of them. But the WC acknowledges (and indeed it's acknowledgeable) that growth was disappointing in the 90's and this is diametrically contrary to the flowery picture PSiM paint here.

The assumption of uniformity of functional parameters for income distribution intra-country and intertemporally is naive!

Perhaps PSiM could have brought aid in to the picture and reflect on whether it explains the miraculous de-pauperisation they paint.

The reality that PSiM are running away from is that institutions matter. There are quite a number of African Countries that have abolished property rights and are growth disasters. Governance crisis and corruption have increased dramatically in Africa and these explain more than PSiM imply. In fact there is substantial income missing in Africa on account of corrupt political leadership that, in some mineral rich countries, owns smuggling syndicates that trade with Islamic (and Lebanese) fundamentalist regimes. So that resources are in the hands of elites and that they are not accounted for in national income data is not an exaggeration. Indeed this makes inequality rampant. The political office has created a lot of distortions to increase lootability of resources and impoverish the non-elitist.

Whether a country is mineral rich or not matters. PSiM's work (Fig 17) shows mineral rich countries have not only a higher GDP per capita but a steeper ascent than mineral poor nations but in terms of poverty rate fall, mineral poor countries have a steeper plunge. Why? That reflects the resource curse is frictionary in the economic machine in such countries and for that reason (we have more political disorders in some mineral rich countries and voicelessness - which amounts to poor poverty reduction effort). Maybe an interactive component of corruption, wars, smuggling and ethnic violent insurgences have been underplayed in this analysis. On the other hand they down play the mineral price boom which has seen many mineral rich countries experience huge leaps in per capita income growth.

But perhaps data deficiency could be a major stumbling block. Note that the Shadow Economy has tremendously grown in most of SSA following SAPs in the 90's. The activities are literally in light industry, foreign currency trading, mineral smuggling and gold panning, small-cross border trading, mini-retail work among many. These are generating tremendous incomes and indeed poverty rates might be falling in line with PSiM's notion although we can't substantiate that without adequate data. Excessive informalisation really goes with uncaptured or miscaptured growth and poverty reduction

Nonetheless, this work is compelling

Efon Epanty
May 03, 2010

We fail because Africans who know their countries and territories better than you are not given the chance and opportunities to work on projects that deals directly with the plights of their communities. This is not the case with Asian Countries. Until international institutions stop this practice and allow we, africans, to develop sustainable and workable initiatives for our continent and countries, western imperalists ideas and development theories would not work.

May 04, 2010

I think that you are touching one part of the problem. How in the world would an international organization gave the similar recipe for all countries whether in African or Asia to devalue their currencies and let them compete for the same northern countries's market. When in fact its fosters more inflationary trends and eat up all possibility of competitive edge that it was supposed to provide.
The main reason companies fail is the system of red tape where anyone in the chain of the system can hampers your initiative from the low level of the administration who is supposed to convey your authorization for a simple signature to political opponent who wants you not to succeed. We have a long way to go as long as we don't take in account that the success of one is the success of long in global. The recent news that are coming from Africa seems to give hope to lot of people who start to think that we are may be "doomed".

Anonymousugbebor olabisi
June 22, 2010

let the world deliver nigeria from legislators whose pay and allowances equal more than what the rest of the nation earn then and only then will our poverty be reduced HELP!777

Armstrong Ongera, Jnr.
December 08, 2010

Public-private partnerships required to overcome poverty. An enabling investment environment, less corruption, and proper planning, and implementation will lift most people from poverty.

Omololu Fagbadebo
December 09, 2010

Recently, I had the opportunities of visiting some areas in Nigeria dotted with MDGS' water supply projects. All the taps are dried. Investigations revealed that the taps worked onl on the day the projects were commissioned. Inspite of the MDG water taps, the people have to go to the streams to fetch water for their domestic needs. The projects were decoys for the leaders to share the money allocated them.

December 26, 2010

When you are in the corner and have no cash to go out from that, you would require to take the loan. Because that will help you unquestionably. I get bank loan every single year and feel OK because of this.

Expert Reviews
April 01, 2011

Last time I visited South Africa the economy really seemed to be doing much better. I think Africa will be on the forefront of the next economic boom.

Omololu Fagbadebo
April 05, 2011

Weneed a more wholistic information on other Afriacan countries.

September 12, 2011

I really hope this is true, we all deserve equal possibilities in life

June 03, 2013

Hello there, are you having difficulties with the web hosting? I had to refresh the page about million times to be able to get the page to load

Finest Briar
August 11, 2014

Many small businesses are struggling through this recession like those that make handmade pipes but will emerge stronger through the other end. Africa and its people have a lot to offer the world through similar small and cottage businesses.