Causes of low growth in Africa

A fascinating new paper from the National Bureau for Economic research finds that adult mortality alone can account for all of Africa’s growth shortfall over the 1960-2000 period.

This comprehensive analytical and statistical study finds that high adult mortality induces people to invest less, accumulate less human capital, have a large number of children rather than fewer, children who are likely to survive and have be economically productive. This, in turns, lowers economic growth. The effect is economically very substantial. The linkage with fertility is particularly strong: as countries develop, the reduction in mortality precedes, and appears to cause, the fall in fertility (and not the other way round).

In my view, this has important policy implications. We should be careful about blaming Africa’s poor performance on its governments and institutions: this paper finds that the whole of Africa’s poor economic performance can be explained statistically by higher levels of mortality.  Some of this may be an indirect result of poor governance; but the burden of tropical disease is also substantially higher in Africa; and the paper points out that poverty can be self-perpetuating, because countries are too poor to be able to devote resources to fighting diseases and reducing mortality. This suggests that we should focus on fighting disease as a significant investment in reducing poverty and helping to break out of this vicious circle. The proposal  to create market incentives for the development of new vaccines for malaria, AIDS and TB, by promising to buy them if they are developed, would be an excellent start (full disclosure: this is what I work on).

It is also worth noting that this analysis tends to back Jeff Sachs’s claim that there is a poverty trap, as opposed to Bill Easterly’s scepticism which I reported last month.

Source: Death and Development, Peter Lorentzen, John McMillan & Romain Wacziarg NBER Working Paper 11620 (registration required; send me an email if you want me to send you a copy)

Hat tip: Stationery Bandit

Published by Owen Barder

Owen is Senior Fellow and Director for Europe at the Center for Global Development and a Visiting Professor in Practice at the London School of Economics. Owen was a civil servant for a quarter of a century, working in Number 10, the Treasury and the Department for International Development. Owen hosts the Development Drums podcast, and is the author Running for Fitness, the book and website. Owen is on Twitter and

Join the conversation

13 Comments

  1. Need this be mutually exclusive with the idea that poor governance carries the can for low growth? If only I could remember where I read it, but I recently came across a very persuasive piece about how many African governments effectively act to suppress enterprise, discourage investment and, on occasion, seize assets.

    I seem to be perpetually making whiny hey why can’t we all get along style comments on your blog, which must irritate, but surely there’s no need to argue over what is the cause of poverty in Africa, when we can perfectly well say these are the causes. It just strikes me that your observation that perhaps we ought not put so much blame on shitty governments, does not follow from everything else you say.

    Besides, might not high mortality rates also be a consequence of kleptocratic governments? So the statistical survey fingering high mortality might quite valid, but missing the underlying cause? just a thought. Of course the paper you cite may have controlled for this, in which case a lot hangs on how well they controlled for it. I wonder how easily qualitative factors like that are captured.

    Presumably, lots of countries did at some time move from high mortality poverty to low mortality wealth of their own accord, which raises the question of why this hasn’t happened – yet – in Africa. This too may well be dealt with in that paper you cite, which I confess I haven’t time to read.

    funny how I can find the time to air my uninformed opinions, and not read something that might inform them.

    Still, I am a fast typist. And reading econometrics makes my head hurt after about 30 seconds.

    Owen replies: And where did I say that high mortality is the cause of low growth? What I said – and I am simply reporting a very good piece of statistical analysis – is that that “poor economic performance can be explained statistically by higher levels of mortality. Some of this may be an indirect result of poor governance…”. There is a substantial body of evidence that tropical countries have an inherently higher burden of disease – and Africa in particular suffers from a type of mosquito that mainly bites humans – and hence higher mortality, than temperate countries. To the extent that climate causes increased disease, that part of the higher mortality, and (on the basis of this evidence) the consequent lower economic growth, is not the fault of poor governance. I do think it would be a good idea to read the paper before you disagree with it.

  2. I have to agree with Paddy on this one. The temptation is to apportion the large chunk of “blame” to a few factors, but it seems that there are more than enough causes already. This reports point could be refuted (simplistically I admit) by the observation that economic development in the west during and after the industrial revolution was not hindered by low mortality and high birth rates.

    It seems to me that the low mortality is the symptom not the disease. And the disease is the combination of concentration of economic and political power in the hands of a very few incompetent and/or brutal “Big Men” over the last several decades. Economic development in every single African country north of Botswana and south of the Sahara has invariably been choked off by kleptocrats. Oil, mineral, and crop revenues have been bled from the continent year after year, businesses and land was (and still is) appropriated/nationalized and redistributed to the clans of the presiding leader,to the detriment of the health and education of millions of Africans. Almost invariably, the idealistic socialist policies of the 1960s resulted in economic ruin by 1975.

    Large, uneducated, idle populations were left to starve (women and children) or join gangs and militias from Angola to Zimbabwe.


    Sachs, a brilliant economist and no doubt a good man, misses the point (it seems to me). Browbeating the west to get to .07% of GDP seems presupmtuous, when the hundreds of billions of dollars in aid over the last few decades seems to have resulted in a poorer, more desparate population. Sachs needs to explain how and why it will be different this time.

    Nevertheless, the “solution” may lie in partnerships like the one you point out. Western corporations need to embrace “benevolent” capitalism, developing programs that allow everyone to benefit, including the shareholders. (This is not greed, just common sense)

    Instead of traveling the world with Angelina Jolie, I’d like to see Jeffrey Sachs traveling with Jack Welch or Peter Drucker (still alive?). Instead of a G8 conference, perhaps a conference of Fortune 20 companies and the top 10 aid organizations. This may already be happening at some level, but it seems that the world is still just focused on the “investment” not the strategy.

    (By the way, I like your blog – and my 4 year old is named Owen)

    Owen replies: Thanks for your kind words about my blog. I assume you mean ‘high’ mortality where you have written ‘low’ mortality.

    I find it strange that you accuse me of simplistically apportioning the blame on a few factors, and then proceed to do exactly that yourself (“the disease is the combination of concentration of economic and political power in the hands of a very few incompetent and/or brutal Big Men”). This is a gross caricature of a much more varied and nuanced problem. Some countries have enjoyed considerable prosperity and growth in the hands of undemocratic governments (Singapore, South Korea); and some democratic countries have not fared particularly well (South Africa, Mali). The colonial legacy of unaccountable governments and western support for kleptocrats (eg Mobuto, Mengistu, Houphouët-Boigny, Eyadéma, Traoré, and Banda) has undoubtedly played a role in Africa’s poor governance; as has the artificial divisions between ethnic groups resulting from the Berlin Conference. The willingness of rich countries to channel huge funds into bank accounts for natural resources (timber, oil, diamonds etc) may also contribute to the problems of political economy in poor countries. But as I noted above in my response to Paddy, some causes of high mortality in developing countries are not related to poor governance.

    Sachs does not miss the point. His argument for 0.7% (not 0.07% as you have written) is based on a serious analysis of the needs of developing countries, and a substantial body of evidence about what works. In a separate post I explained – at least to my satisfaction – why the argument that “hundreds of billions of dollars in aid over the last few decades seems to have resulted in a poorer, more desparate population” is not correct – I hope you’ll take a look. However, I do not believe that aid alone is sufficient (I do not know if Sachs believes this, but he often seems to imply it.) There are many things that need to change, in poor countries and rich countries, to increase the prospects for the reduction of poverty. But there is all the difference in the world between saying more aid is not the only answer (which I agree with) and saying that more aid won’t help (which I don’t agree with).

    Who is your son named after? I was named after Robert Owen and Wilfred Owen – of which I am very proud.

  3. I think you’re being a bit hard on Paddy – after all, it’s not hard to mistake the claim that a particular factor “can account for all of Africa’s growth shortfall over the 1960-2000 period” for the claim that this is the sole cause of the shortfall. And did you read the whole paper before you agreed with it? If so, well done, because I don’t always manage it 😉

    There is obviously some causation from low income to higher mortality, but you’re quite right that there is an independent cause of higher mortality in Africa too – geography. Of course, I don’t know if the paper controls for this, so like Paddy I’m sitting here writing about a paper I haven’t read. But then again if I spent my whole time reading them I’d never get to comment on them …

    Owen replies: I have the advantage that I am paid to read this stuff.

    This research shows that, statistically, higher mortality can explain Africa’s growth shortfall.

    It would be hard to claim that higher mortality is solely caused by poor governance – there are other factors, such as geography.

    To my mind this is important because it suggests that if we can find a way to break in to the vicious cycle (poverty -> disease -> mortality -> poverty) – for example through vaccinations which have been shown to be very cheap and effective – then we might be able to make a substantial difference, whether or not Africa’s current troubles are caused in part by poor governance.

    I do worry that the comments above from Paddy and Paul suggest a mindset of blaming the victims: put bluntly, some people seem to want to say that Africans are poor because they can’t govern themselves properly. My aim in this post was, and is, to highlight evidence of opportunity, not to cast blame.

  4. Owen

    Yes – high mortality. Sorry.

    First, while I may have a certain grasp on the issues here, I’ll concede that you are better informed and in a better position to offer up an analysis of the problems. With that in mind, re-read my first line, “The temptation is…”. Its not a dig at you, just at what I see as a general tendency to simplify this issue – on all sides of the debate. My goal is to avoid this trap, and while it may look like I am holding up one “cause” I am really at the beginning of the puzzle personally, and don’t expect to have any hard answers any time soon.

    However, that does not mean that I can’t have my theories, and while I respect Sachs (thanks for pointing out my decimal error) even you admit that he may “believe” that “aid alone is sufficient”. He probably does not, but as a spokesman, I think he understands that his job is to get a soundbite out. This is the same criticism that I and many have about politicians – they think that their audience needs simplicity. Oh well.

    There is no question that Western governments and corporations have been enablers. But while there is no one cause, it is easy for many to raise the ghosts of colonialism or the cold war to remind the west of their sins. This can have the effect of turning some off of the subject, or of allowing others to have a sense of self satisfaction without really adding anything to the discussion.

    My view, however, is that African rejection of their colonialist past, along with the generally positive attitude in the 1950s and 1960s towards command economies, led to the embrace of disastrous economic policies. Even where the best intentions may have been at work (such as in Ghana under Nkrumah) the attempts of the government to control every aspect of the economy led to substandard growth and dependence on aid.

    Again, without being critical, I am curious about the analogy of mortality in the west and 18th/19th century economic growth. If 20 years were added to the average lifespan in sub-Saharan Africa, would we then see an alleviation of the problems in question?

    Finally, I do think that aid, and probably more aid, is necessary. But the current model seems not to be working.

    And, I have admittedly not read the paper – yet. I will do that.

    My hope is that the next/current generation of African leaders are able to become the driving forces behind governmental and economic reforms, and that massive aid becomes unnecessary within the next few decades. Without their involvement, and without significant private enterprise economic development by Africans for Africans, we will not see an end to this cycle in our lives.

    Just my rambling opinion though.

    BTW, my wife’s favorite book is A Prayer for Owen Meany. Also, Saint Owen was captured by pirates in the 6th century and was known for his love of reading and learning.

    Owen replies: “If 20 years were added to the average lifespan in sub-Saharan Africa, would we then see an alleviation of the problems in question?” – The NBER study quoted above suggests that we would, yes. This is supported by other studies: in 2003, Nordhaus found that half of US growth in the first half of the twentieth century was directly attributable to improvements in public health. IIRC, life expectancy in SSA is now at about the levels it was in the UK in the 1840s.

    Yep: here’s the graph:
    life_expectancy.png

  5. Any chance of a gander at the full paper? One thing that puzzles me:

    “The linkage with fertility is particularly strong: as countries develop, the reduction in mortality precedes, and appears to cause, the fall in fertility (and not the other way round).”

    This is a surprise? I thought it was simply common knowledge. From here (yes, I know you love this site):
    http://www.techcentralstation.com/100104E.html

    To an economist, the falling birth rate is easily explainable. We’ve got rich, fat and happy. This gives all of us, both male and female, more choices and more choices mean that the opportunity costs of any specific one are higher. Offered alternatives to bearing 12 children, women find that bearing one or two is quite sufficient thank you, they’d also like to have some of that self-esteem found through a career, that leisure of a sabbatical, well, simply more of the delights that modern civilization has to offer. Those who read a little Darwin might also point to modern medicine as lowering the fertility rate. Not just the availability of contraception but a change in the number of children desired instead of the number thrust upon those who enjoy the carnal delights. For if the meaning of life is to have descendants who then go on to have more (a very rough and ready outline of Darwinist theory) then the fact that almost all children survive to themselves be able to breed means you need to have fewer of them. Or, in econspeak, the opportunity cost of having few children has fallen.

    Similarly, short life spans leading to underinvestment. Who’s surprised?

    The solution might be disease but I would proffer an alternative…sanitation. That’s what reduced the English death rate. And done by local authorities largely, with minimal intervention by the central state. Or in France, by the private sector.

  6. Tim – thanks. I’ve sent you a copy of the paper by email. If anyone else wants a copy, please let me know by email or here in the comments.

    You are right: the conclusion itself is no surprise, as the authors readily admit. But it surprised me how strong and clear the effect is. Statistically speaking, once you control for high mortality, there is nothing left to explain about low growth in Africa. And as you say, high mortality is something that we can pretty easily do something about. Sanitation, clean water, vaccinations and access to medicines would make a huge difference. All easily within our grasp.

  7. sheesh, misunderstanding city. I don’t know where you think I said I disagreed with the paper! I am perfectly happy to accept the arguments about the economic effects of high mortality, or at least I see no immediate grounds for rejecting them. You could add in the effects that high infant mortality and poor nurition has on education – the Economist did a good piece ages ago on how hard it is to learn when you’re starving. The only real question I have is on what’s cause and what’s effect, hence me wondering whether/how they controlled for bad governance, but as you (and I) said, I ought to read the damn thing. I’m paid for doing something else.

    The the not these thing was something I inferred by reading your argument as follows:

    high mortality rates can explain low growth
    therefore we ought to be careful about blaming poor governance

    I just didn’t think it had to be a case of either/or, that was all. re-reading your original post, I see I failed to pick up on this bit: “some of this may be an indirect result of poor governance”. Well, that just raises the question of to what extent poor governance is to blame, it doesn’t suggest to me that we should be careful about blaming it.

    You say: To the extent that climate causes increased disease, that part of the higher mortality, and (on the basis of this evidence) the consequent lower economic growth, is not the fault of poor governance

    Well, maybe. I mean Finland has to deal with it being cold enough to easily freeze to death. Are there not other continents/countries with high incidences of tropical disease that are better off that Africa? Again, please don’t react as if I’m disagreeing – I’m just raising questions.

    And please have no fears for my “mindset”. On that subject I find this puzzling:

    some people seem to want to say that Africans are poor because they can’t govern themselves properly

    If that’s how you interpret it, no wonder you are apparently so hostile to the idea.

    I really don’t think the idea that kleptocratic, corrupt and incompetent governments are responsible for much of Africa’s woes, is tantamount to ‘blaming the Africans’ for their problems. Unless we’d also blame the Russians for Stalin or the Cambodians for Pol Pot. I guess there is some sense in which people are responsible for how they are governed, but primarily the responsibility for despotism lies with the despot, wouldn’t you say?

    I also note in your response to Paul that you say “Some countries have enjoyed considerable prosperity and growth in the hands of undemocratic governments (Singapore, South Korea)” Well, quite. Undemocratic governments can be growth friendly. The point about many African governments is, surely, that democratic or not, they have acted to inhibit growth.

    You say: “If 20 years were added to the average lifespan in sub-Saharan Africa, would we then see an alleviation of the problems in question?” And you show that the answer would appear to be yes. Ok, but perhaps for that to happen, we’d need to see better governance. We still, as far as I can see, haven’t untangled cause and effect.

    Seems to me that Paul, Tim & I have tried to comment constructively and in good spirit, we’re not saying “Owen you’re wrong”.

  8. that last sentence was unwarranted – Owen please strike it out. that was just me getting defensive.

    if the main point that you were making is that besides trying to improve governance in Africa, we should also be going after things like sanitation and vaccinations – well hell yes. Go Bill Gates!

    my dad – a civil engineer – likes to say that nothing saves lives like a good sewer

  9. Actually this does seem like potentially good news.

    I think it means that if we can reduce mortality, then people will tend towars smaller families with more rsources per person and that we can hope that as they become wealthier reforms will occur in society with some multiplier effect.

    Since there are a number of possible and understandable methods that may succeed in increasing life expectancy then we can *hope* these will work rather than depending on social interventons we are still pretty clueless about.

Leave a comment

Your email address will not be published. Required fields are marked *