How to spend $1m reducing climate change

Suppose you had $1 million to spend on tackling climate change.  How would you spend it to get the best bang for your million bucks?

Would you spend it on stopping the slash-and-burn of forests?  Perhaps on switching to nuclear energy?   More energy-efficient buildings?  Building cleaner power stations?

According to a recent paper by David Wheeler and Dan Hammer, climate change experts at the Center for Global Development, the answer is (drum roll): you would do much, much better to spend your money on a combination of family planning and girls’ education in developing countries.

This table, based on data in their paper, shows how many tonnes of CO2 would be abated for your $1m:

Intervention Tonnes of CO2
Family planning & girls’ education combined 250,000
Family planning alone 222,222
Girls education alone 100,000
Reduce slash and burn of forests 66,667
Pasture management 50,000
Geothermal energy 50,000
Energy efficient buildings 50,000
Pastureland afforestation 40,000
Nuclear energy 40,000
Reforestation of degraded forests 40,000
Plug-in hybrid cars 33,333
Solar 33,333
Power plant biomass co-firing 28,571
Carbon Capture and Storage (new) 28,571
Carbon Capture and Storage (retrofit) 26,316

The logic, of course, is that if there are fewer people on the planet, then we will generate fewer greenhouse gas emissions.  Population policies are important because there are many people in developing countries who want smaller families, but don’t have access to the family planning services they need to achieve this.  Education is important because educated girls want (and are more able to insist on) smaller families.  That’s why these interventions are important and cost effective, both individually and especially when done together.

Win – win

This approach is particularly attractive because, in addition to helping to slow global warming, there are other, very significant benefits for the citizens of developing countries of access to family planning and to education for girls.

The other day I reported here that if donors invested about $180 million a year to provide modern contraception to every Ethiopian woman who wants it, this could set off a virtuous circle of rising income per capita, lower desired family size, greater use of contraception, lower numbers of children, and so rising income per capita.  My back of an envelope calculation found that a decade of access to modern family planning would have roughly the same effect on incomes in Ethiopia as the entire international aid programme in Ethiopia does today.

As well as environmental and economic benefits, there are important social and health benefits for women and their families, which strengthen the case for these investments over and above the cost-effectiveness figures shown above.

Making choices

Of course in an ideal world we would do all of these things.  But although it is inconvenient to acknowledge it when you are busy trying to save the world, resources for averting climate change are limited. We should make informed choices to reduce carbon emissions in the most cost-effective and sustainable way we can with the resources available, to secure the biggest and broadest benefits.   These figures from the Center for Global Development imply that investment in family planning and girls’ education would be a far better investment than the UN Reducing Emissions from Deforestation and Forest Degradation (REDD), which aims to spend $30 billion a year on incentives for developing countries to reduce deforestation and forest degradation.

We would get three or four times as much bang for our buck – in terms of climate change benefits – from population policies and girls’ education as we would from even the most cost-effective investments in forestry (stopping slash-and-burn), and in addition we’d get the broader economic and social benefits for the people of developing countries.

So why isn’t this, in fact, where we are spending the climate change money?  Something to do with the power of industry in the environmental lobby? (Update: See Eliot’s comment below)

(The figures in the table above are calculated from Table 2 and and Table 5 of The Economics of Population Policy for Carbon Emissions Reduction in Developing Countries, David Wheeler and Dan Hammer, Center for Global Development Working Paper 229)

39 thoughts on “How to spend $1m reducing climate change”

  1. Interesting post. I think lateral thinking like this will be very important in the fight against climate change. However, I think the post (and possibly the paper – no time to read it) omits one very important element: when will those carbon savings be delivered? Mostly they will be a long time in the future (are they factoring in exponential savings?). Switching from fossil fuels to, say, wind power, will generate carbon savings into the future, but they will also generate immediate savings. And if we don’t start making some pretty large cuts in carbon emmissions pretty darn quickly, a lot of the debate may become academic. So yes, invest in family planning by all means – although I criticised the $180m estimate, I do not disagree with the principle – but we also need to make some serious efforts with rather more short term impacts.

    Owen replies: That seems to be the second thing you’ve expressed doubt about without having any actual basis for your scepticism. Perhaps you’d better read the paper? Do you have evidence about the relative speed of changes through population policy as opposed to constructing new power stations? If it does turn out that the benefits for the environment of population policy come through slower (which I doubt, but I don’t know) then can you show that the benefit of reducing carbon emissions a little earlier (eg through forestry policy) is more valuable than a reduction which is four times bigger a little later?

  2. Owen – nice post. I think some people are convinced by the population argument, but I think you’re assumptions about the effect of population policies on economic growth are pretty suspect. There is no convincing evidence out there that that the chain of causality runs from —> lower birth rates —> higher economic growth.

    I think most demographer and population economists would argue that the actual relationship is more nuanced, and that it’s more likely be that higher incomes create a demand for lower family sizes, for plenty of reasons. Many also believe that the surge in population that comes from the demographic transition (the period where birth rates are high but death rates have fallen) can and has had benefits in places like India

    Back on the envelope calculations in your previous post like “Hence if Ethiopian women could achieve the reduction in family size they currently want, from 5.4 to 4.0, this would increase growth of GDP per capita by approximately 0.35% a year.” are, I think, pretty misleading.

    Coming back to the CO2 argument, the Wheeler and Hammer paper uses assumptions on the effectiveness of family planning intervention and education. These assumptions are not necessarily based on casual studies – in the case of female education, they are based on cross-country regressions (the same studies we quickly dismiss in the aid-growth literature).

    I don’t think anyone believes that family planning and education don’t have a negative effect on population rates, but the estimates used in papers like these should be treated with extreme caution, especiually when the value for money is of the same order of magnitude as other interventions.

    1. Matt

      Thanks. I agree that it is very difficult to tell statistically whether higher incomes cause lower desired birth rates, or lower birth rates cause higher incomes, or both.

      Common sense suggests that both are likely, and indeed that there could be a virtuous circle of lower birth rates, higher incomes, leading to lower desired birth rates.

      My calculations were influenced by this paper by Carl-Johan Dalgaard & Henrik Hansen which says:

      For each percentage point of fertility reduction per capita GDP growth will likely increase by 0.25%. As explained above, the size of this effect is likely to decline, however, as agriculture becomes less important to economic activity. A difficulty, however, lies in how to reduce desired fertility.

      That was the way I calibrated my little back of an envelope model. I’m not saying it’s right; but it passes the laugh test. I think it is fair to work on the basis that the size of this effect is not zero, as you say, though I agree with you that it is hard to work out with statistical precision what it is.

      I find it easy to believe that we don’t know the best way to reduce desired birth rates. What seems quite clear is that if desired birth rates are falling, but women cannot achieve smaller families because they don’t have access to the relevant family planning services, then this is bad for all sorts of reasons (both because reducing actual birth rates may be good for the economy and because welfare is reduced if people are unable to achieve their preferences for smaller families).

      Finally, the key figure from the Wheeler and Hammer paper is that the cost per tonne of CO2 abated through family planning is $4.50 / tonne. That figure is based on some pretty detailed costings of family planning interventions, read alongside the data for carbon emissions per person in different countries. So that figure – which is the key calculation – is not based on cross country regressions.

      Kind regards

  3. I suspect a lot of those figures – or at least those relating to development interventions – have a fairly tenuous relationship to actually achievable C reductions, especially if those interventions are attempted on a large scale. The Stern report highlighted reducing deforestation (REDD) as being one of the most cost effective measures that could be taken, based primarily, I seem to recall, on an opportunity cost analysis. In conservation circles we were very pleased to get this kind of backing. However, putting into effect his blueprint is proving very much harder. A poor farmer may not be making much money from agriculture, but providing them with solid alternatives to clearing more forest is a lot more difficult than just ponying up a little bit of cash. (As, no doubt, any tropical agriculural expert would have told us.) When you take those kind of caveats into consideration geothermal energy starts to look like a much better bet! (based on the table above)

  4. Oops. Missed your previous comment. It appears that if you edit an existing comment to put a reply in line people subscribed to the comments do not get an email.

    Any way, as to the meat of your retort, I was stung into scanning the CGD paper. Their claims for the C reductions for population reduction stem from a paper by O’Neill et al in PNAS. That shows small savings up to 2020, medium savings 2020-2040, and more substantial ones after that with some exponential effects, as you would expect. There would presumably also be some kind of time lag, as you allude to, between the education / family planning investments, and the reductions to population growth rates, which I did not see mentioned by Wheeler and Hammer. So my initial query (nowt wrong with that, surely!?) turned out to be not far off the mark.

    As I said in my second comment, actually turning this investment into real C savings is a considerable challenge, whether in forestry (where I do know what I’m talking about) or family planning (where I have only the anecdotes from all my public health friends). Of course that means that doesn’t mean we shouldn’t try. The potential payoffs can be very big. But a sensible portfolio approach to managing climate change risk would combine these approaches with more certain, more immediate efforts.

    And while I take your point about reading the papers on which I am voicing skepticism, if I wanted to write learned academic papers that is exactly what I would do. But I have a day job which takes up most of my time, and the blogosphere is for commentary, questioning and vigorous debate. My commentary is informed by the gap I frequently detect between international and national level planning and economic modelling, and the reality as implemented on the ground a long way from the urban centres.

  5. Hi Owen, the obvious point is that if you want to reduce birth rates, the best place to start is those countries with the highest per capita emissions, and the highest emitters within those countries. So let’s start in Australia, rather than Kenya, where per capita emissions are tiny. I had a row with some population controllers about this just before Copenhagen, when they tried to piggybank on the interest in climate change (wow, that seems a long time ago)by offering a ‘population offset’. See for more
    best wishes

    1. Duncan – You are right that you’d get a relatively big reduction in emissions for a given reduction in fertility in Australia; but you probably wouldn’t get much reduction in fertility for your spending. There is probably not much unmet demand for modern contraception in Australia – whereas there is in Ethiopia or Pakistan. We don’t know much about how to reduce desired fertility (which is what you’d need to do in Australia) but we know that reducing actual fertility in countries where there is unmet demand for family in planning is quick and relatively cheap.

      I suspect that means that reducing carbon emissions in Australia is not in fact a cost effective way of reducing global warming but I’d like to see the numbers. If you are right that reducing climate emissions by reducing fertility in Australia is the most cost effective way of reducing global warming, and if that is what we are trying to achieve with this money, then perhaps we should indeed spend the money this way.

  6. While reducing number of people on the planet through family planning and girls’ education in developing countries is important, it doesn’t affect the behavior of people in the developed nations who will continue responsible for the vast majority of greenhouse gas emissions.

    Developing countries’ carbon emissions are expected to surpass emissions of industrialized countries in the next decade. However, much of the increase will be due to China and India fueling their economic development with fossil energy. Reducing birth rates may help us address climate change, but the question then is whose birth rates?

  7. Probably an obvious question –

    While reducing birth rates is important (whether in rich or poor societies), will donor funds also attempt to promote programs and campaigns to reduce consumption in rich societies (residing in developed as well as developing countries)?

  8. Duncan and Owen,

    Although I don’t think anyone has looked at Australia, David Wheeler did run some numbers for the United States that aren’t in the final paper and as Owen suspects, the bang for the buck wasn’t dramatically different from many developing countries. Yes per capita emissions in the US are currently high, but the cost of providing family planning is also much higher. When David did a rough ranking of countries the US fell somewhere in the wide middle in terms of bang for buck of investing in family planning as a way to reduce emissions. There is a brief consideration of the United States in the O’Neill paper referenced above and in another paper by Scott Moreland and colleagues at the Futures Group. See for both papers.

    All the best.

  9. Hi Owen,

    It is interesting to see this effect. Most interventions I can think of have negative tons of CO2 saved. However, I think there must be other interventions out there with a positive eternality on climate. A brainstorming to identify these clear win-wins is indicated.

    However, I have a moral issue with the thinking behind the calculation (this moral issue does not make the result false). Every woman, should be able to have the number of children they wish for. In most places this is way less than the actual number they have. For me this is a moral right. However if the climate crowd jumps on the bandwagon and buys gold standard carbon offsets with anti-conception programmes, the reason we offer these reproductive health services don’t stem from the rights of the women, but from our wish for limiting her offspring. We could push the birthright below the child wish.

    I don’t think we should refuse the money for reproductive health, but there should some serious checks on the system.

  10. Great post Owen. I would add that well-designed and managed family planning programs in countries with high unmet-need for contraception are often the most effective and cost effective interventions for bringing down high maternal mortality rates (fewer abortions, leading to fewer botched abortions and death; and fewer delivery related deaths). That’s quite a nice positive externality for your climate change dollar invested.

  11. Has anybody mentioned how much value you’d get out of spending that million on education of US citizens? Seems like the ongoing huge omisssions are supported by a conservative lobby, who manages to push their message on climate change being a fraud through to an uneducated American public.

  12. I liked the post and the subsequent debate, but Owen, I felt your final comment was a little off base. Surely the reasons that the environmental lobby fail to tackle this subject might be similar to the reasons that the rest of society has yet to do it justice? After all, there are many many other reasons why girls education and family planning are a good thing? Plus, I would say not unreasonably, environmentalists like trees. There may also be a corporate evil influence, but it’s likely that we’d be where we are whether or not this is true. But more power to your elbow in changing this.

    Also, while I’m here, I think it’s worth pointing out that while the cost-curve approach is useful, indeed vital, it’s unfortunately not definitive as a guide to action. Non-cost barriers to action on climate change (and poverty) are massive in many cases. Other considerations are also important – like politics, timing (as MJ raises), technical capacity and morality (as Sam raises).

    After all CCS is at the heart of many governments climate strategies, not because it is cheap but because they see no alternative to it, if they are to deliver the low-carbon transition in a politically and technically viable way.

    We should be delivering family planning and girls education in Ethopia because they are good things that enrich us all. We should be doing what is possible to preserve forest ecosystems because in many cases when they are gone, they are gone, and their existence enriches the planet. Unfortunately we’re struggling to do either because they are both difficult and made harder when we spend huge amounts on the (relatively easier) CCS project.

  13. I find it completely inane to suggest that by limiting our human presence on earth we are doing any sort of a good thing. The solution lies in finding the best way to make use of the earth; to subdue it without destroying it. Limiting human life so it doesn’t take the place of a tree is sickening.

  14. Pingback: Climate Change and Family Planning » Maternal Health Task Force

  15. Duncan’s point is still right in the sense that there are diminishing returns to the environmental benefits of family planning. Even if you could reduce Ethiopia’s contribution to global warming, their contribution is tiny compared to the OECD countries. And even very significant reductions in birth rates won’t come anywhere close to achieving the needed reductions in CO2 emissions. We need to figure out how to spend 100 billion dollars wisely, not just 1 million.
    In any case, we are currently at the global replacement rate of 2.1 births per woman, though of course some countries and in particular rural areas are much higher. It is how we live that will determine the climate’s future, not just how many live.
    Of course, it is good to invest in girl’s education and access to reproductive health services and contraception because that is intrinsically valuable and a moral right of women and girls, in addition to the many instrumental benefits which flow from it.

  16. Scott, modest reduction in birth rates (1/2 birth per women which is the difference between the UN High and Medium projections) would actually make an enormous impact in reducing CO2 emissions. Of course no single option whether it be family planning, solar, nuclear, efficiency, REDD or CCS will solve the climate problem. Certainly as Socolow and Pacala note we need many wedges. However a recent paper published in PNAS shows that family planning can be one of those wedges delivering in the realm of 10% of the total reductions in emissions many feel is necessary. That is a huge chunk. That means that the modest reduction in fertility would be equivalent to ending all deforestation today, or increasing the worlds reliance on wind power 40 fold. Those are things that could cost hundreds of billions if not trillions, while the estimated cost of providing family planning to all with an unmet need is only $3.7B. See And as you point out well, providing family planning to women who want to avoid pregnancy is the right thing to do anyway for it’s health and rights benefits. The large impact on emissions is just the icing on the cake.

  17. “Easier CCS”? Are we actually considering Carbon Capture and Storage as a viable alternative? A technology thats yet to be proven, raises huge doubts by experts, and has huge technological/physical/economical barriers? I’m amazed nuclear fusion isn’t in the middle as well…

    But back to the actual point of the post. Correct me if i’m wrong but your assuming these numbers based on lower co2 emissions due to smaller families?

    But if family planning have a positive effect in gdp, wouldn’t these large poor african families tend to evolve into small rich african families. Rich as in more consuming? As in higher co2 per capita emitting? Like the developed world?

    In other words are you assuming that family planning means lower emissions without being counterbalanced by higher consumption per capita?

    Cause that’s what you have in more developed nations. In fact the trend is the more developed, the more family planning, the lower the family sizes, the higher gdp, higher consumption and higher emissions per capita. (look at Europe and thats generally what you’ll find before cap and trade)

    Again, i might have misunderstood your back of the envelope.

    Secondly, even if it’s the most cost effective, how much of an impact is possible?

    Developing african nations emit aprox 300 million tons (discounting South African Coal mines, thats less than half of germany – How much of this 300 MTon can you reduce with this plan? Might be wrong but i’d bet a few %…

    Seeing that Kyoto originally called for aprox 1100 MTon emission reduction, planning and education should probably be a good quick win. But it’s going to be a drop compared to what other solutions you mentioned can and will have to do. And that’s why everyone looking at the solutions that will make a difference, and not always the most effective.

    An antelope might be fast, but it’s pretty useless when what you need is a horse…

  18. “We don’t know much about how to reduce desired fertility (which is what you’d need to do in Australia) but we know that reducing actual fertility in countries where there is unmet demand for family in planning is quick and relatively cheap.”

    Well, actually, we do know how to reduce desired fertility: economic development. Which is all rather circular yes, but still true.

    And of course, yes, lower population growth will reduce future carbon emissions. But this still leaves us with the question of how wide the gap is between desired fertility and actual fertility. That is, how much less population growth will we get by making contraception available (ie, how can we aid people in moving to their desired fertility)?

    Not a lot it seems:

    “Ninety percent of the differences across countries in total fertility rates are accounted for solely by differences in women’s reported desired fertility. Using desired fertility constructed from both retrospective and prospective questions, together with instrumental variables estimation, it is shown this strong result is not affected by either ex-post rationalization of births nor the dependence of desired fertility on contraceptive access or cost. Moreover, despite the obvious role of contraception as a proximate determinant of fertility, the additional effect of contraceptive availability or family planning on fertility is quantitatively small and explains very little cross country variation. These empirical results are consistent with theories in which fertility is determined by parent’s choices about children within the social, educational, economic, and cultural environment that parents, and especially women, face. They contradict theories that assert a large causal role for expansion of contraception in the reduction of fertility.”

    Sure, providing contraception to those who desire it but do not have it is a great use of aid money. But it isn’t quite as much of a golden bullet as many seem to think. Precisely because the gap between desired and actual fertility caused by the absence of contraception isn’t as large as many seem to think.

    Owen, this is a chapter in my book out at the end of this month. Where do I send the review copy?

  19. I see Tom has made the point I was to make viz. smaller families can become richer increasing their consumption and emissions. I was hoping a comment about this from Owen – will there be one?

    Owen replies: It is an excellent question to which I don’t know the answer, so I’ve sent an email to David Wheeler asking him whether they adjusted for this, and, if so, how. But David is travelling so I may not get an answer right away.

  20. Pingback: The Biggest Bang for the Climate-Change Buck? -

  21. Seems to me that among families with equal resources, the ones with fewer children would have higher standards of living (per capita). Also, fewer children means less demand on resources.

  22. Pingback: links for 2010-11-11 – Kevin Burke

  23. Pingback: Klimaschutz? Familienplanung! « Kantoos Economics

  24. Nice one Owen. It’s about time the environmental NGOs had a good dose of pragmatism. The other side of the population coin is of course that massive increases put pressure on the ecosystem as forests are cut down for firewood and there is increased demand for food (Including bush meat). If we just keep looking for technical solutions to satisfy ever-increasing consumption then we are clearly missing the point.

  25. Pingback: save the environment, wear a condom | Bubblingmurmur's Blog

  26. According to “Adding it up” ( the cost of providing maternal and newborn health services would cost $1.5 billion less if full investment in family planning was made. You can imagine the cost saving to other service areas (like education) too. So there is good economic sense to investing in family planning. However, any discussion about the economic and climate change benefits of fertility decline, must go hand in hand with a recognition of the importance of women’s right to choose, the benefits to gender equality and overall well being. Also the discussion should take place within the context of the International Conference on Population and Development (ICPD 1994), which at least is an international consensus on approach. The dangers of population control and the difficult political positions around family planning and safe abortion are a minefield. Also there needs to be a awareness of the different emissions impact per person in different countries and in different wealth quintiles (the cited report seems to base numbers on different country situations, but does not seem to take into account fertility and access to FP in different wealth groups). Immediate reduction in fertility among groups of people who consume very little, will not have such an impact on emissions. Unmet need for contraception in developing countries is much more prevalent among the poor and marginalised, who may also be the people who are consuming very little – and who may continue to consume very little even in a context of high economic growth or in middle income countries (e.g. Peru in the 1980s and 90s and since then) – unless, of course, there is some sort of massive reduction in inequality. At least it is good to see further arguments for increased investment in family planning and girls education – would just suggest keeping within the context of international policy and nuancing based on the actual emissions that might be saved.

  27. Pingback: make that a double (effect) | Hard Consonant

  28. Pingback: » RT @m4good: Best bang for your… ClimateChange on twitter

  29. Hi Owen and all,
    That’s a really interesting post, and the replies bring up many good point both pro and con. Responding to Jennifer Lentfer and Suvojit’s posts back on Nov 9th, there are several good programs aimed at reducing birthrates in Developed nations. Since availability and information about birth control supplies are already high in these countries, affecting birthrates here often involves helping move *desired* birthrates. Of course, that is often the case in undeveloped countries too. Several groups are effectively working on this, of which one is the Population Institute.

    I think it is everyone’s right and responsibility to take care of their own family size, and also include concern about the environment in their region, and the future of their children too.

  30. Thanks for the heads up about the Wheeler and Hammer article! Your estimations seemed a little high on a few fronts, but I’m sympathetic to the direction you’re taking! I look forward to seeing a more rigorous manifesto put forward in the near future.
    If you’re interested in other rantings and ravings:

  31. Pingback: Cutting international family planning funding: bad news for climate change « Do No Harm

  32. Pingback: links for 2010-11-11 | Kevin Burke

  33. Pingback: 7 billion « quietlyquestioning

  34. Population Control, Evolution, and Individual Rights

    A claim made in earlier comments that women have a right to the number of children they desire is most interesting in light of historic rates of infant mortality, unavailability of birth control, and risks to mothers giving birth. Women were not endowed with the ability to have their desired number of children. Probably the greatest ability to choose the number of births is in modern developed countries with low child mortality rates, readily available birth control, and fertility clinics.  Even so some people have more children than they want and some have fewer.  Even if a substantive argument can be made that it is a woman’s right to have as many children as she wants, individual rights are often limited because expression of one’s right would impinge upon the rights of others.  Is there any doubt that anthropogenic climate change impinges upon the rights of many?

    If a reduction in population growth is accomplished through family planning such that individuals choose to have fewer children or not based on their personal inclination, a problem will arise that will get worse over generations until such a program is totally ineffective.  I’ll explain.  Studies of human preferences show that about half the variation between individuals are the result of differences in their DNA and the other half of the variation is due to differences in their environments. This is true even of how interested one is in religion.  It is very likely to be true of how much one wants to have children and if one wants to have few or many children.  If then, people choose to use birth control or not based on personal inclination, the next generation will disproportionately be the descendants of those with a yearning for more children.  Thus humans will evolve over a few generations to want more children than they currently do and to be less susceptible to encouragement to have fewer children.  To keep such evolution from occurring the number of children people have should be random with respect to their interest in having children.


  35. Dear William, 

    I follow your logic, and I understand what you mean. Probably your analysis is even right, in the long term. This kind of evolution takes time. 

    However, I advise you to meet real world people more often, as your recommendation seems to be difficult to implement in anything but a tyranny. You should be able to do better. 

    in the real world, solutions limiting the rights of people so often end up so wrong, and I don’t know about a lot of cases where they ended up right.  

Leave a Reply

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