What should an effective altruist do: save children or prevent unwanted pregnancies? According to the charity evaluator GiveWell, one of the most cost effective health charities is the Against Malaria Foundation. This organization distributes bed nets against malaria. Saving a healthy life year (preventing a disability adjusted life year or DALY) with bed nets costs roughly €10 to €50.
Saving children’s lives with bed nets is good because it increases the total well-being in the world (total well-being is the total number of lives times the average well-being of those lives). But what about concerns of overpopulation and environmental degradation?
There are two considerations, one about the well-being of non-human animals, one about the well-being of humans. First, more human lives means less room for wild nature and hence a decrease in wild animal populations. This loss of wild nature influences the total well-being, but it is far from clear how it affects total well-being. It depends on the average well-being of wild animals: total well-being can increase if wild animals have on average a negative well-being but it can decrease if wild animals have a positive well-being. We do not have enough data to know how a loss of wild nature influences total well-being, so I will neglect this issue for the moment.
Second, an increasing number of children might also result in an increasing competition for scarce resources and hence a decreasing average well-being of humans. There is some evidence that an increasing human population results in a lower average well-being of (future) humans.
But we can also donate to charities that invest in family planning, such as the UNFPA or IPPF, in order to prevent unwanted pregnancies. This intervention has a lot of benefits. It is not directly about saving lives, but rather about preventing lives. Investments in family planning will decrease rapid population growth and avoid the problems of overpopulation and decreasing average well-being. There will be fewer lives, but those who live will have a higher average well-being.
Donating money to charities to save children (with e.g. bed nets against malaria) results in a strong increase in total well-being, but its influence on average well-being is not clear. There is some small evidence that saving lives results in fewer unwanted pregnancies and fewer births (because more people implies more economic transactions and productivity, which results in more investments in e.g. family planning and more economic activities by women who as a consequence delay or avoid pregnancies). This might result in an increase in average well-being that is perhaps offset by an increasing competition for scarce resources. So the overall impact on average well-being is unclear.
Donating money to charities to prevent unwanted pregnancies (with family planning) results in a strong increase in average well-being, but its influence on total well-being is not clear. Investments in family planning save healthy life years (of mothers and children who are already born) at a cost of €50 to €100 per DALY. This is a bit less cost effective than the bed nets, but it is in the same order of magnitude, much higher than most other health interventions. This might result in an increase in total well-being that is perhaps offset by a loss of well-being due to the fewer births. So the overall impact on total well-being is unclear.
In summary: saving children implies increasing total well-being (and an unknown impact on average well-being), preventing unwanted pregnancies implies increasing average well-being (and an unknown impact on total well-being).
In population ethics there is an important debate whether we should promote total well-being or rather promote average well-being. The former results in the repugnant conclusion (preferring a huge population where everyone has a very low but still positive well-being), the latter results in a reverse repugnant conclusion (preferring a very small population where everyone has a high well-being and preventing the lives worth living if those lives have a lower than average well-being).
So what should an effective altruist do? I suggest to promote donations to both kinds of charities, to save children and to prevent unwanted pregnancies. Here is why. I believe that our preferences for total or average well-being are based on our moral intuitions, and both intuitions to increase total or average well-being are equally valid. Both principles to improve total well-being and average well-being are equally coherent, so there is no objective method to decide between these two principles. If you have an intuition to improve total well-being, you should realize that you could have been born with the intuition to improve average well-being instead. There is no objective preference. It would be arbitrary to say that your own intuition and your preference is the right one. So you should be tolerant towards the effective altruists who have the other intuition. You should value both total and average well-being.
The figure below illustrates how total and average well-being are affected by the two interventions. The horizontal axis corresponds with the improvement in average well-being, the vertical with total well-being. An intervention to save children lies somewhere in the blue area: a clear increase in total well-being but an uncertain influence on average well-being. An intervention to prevent unwanted pregnancies lies in the yellow area: a clear increase in average well-being but an uncertain influence on total well-being.
If you donate only to save children, you increase total well-being but you risk decreasing average well-being (although that risk might be small, because saving children might also improve average well-being and investments in family planning). Conversely, if you only donate to prevent pregnancies, you risk decreasing total well-being (although that risk might be small, because preventing unwanted pregnancies also saves healthy life years of mothers and children who are already born). To maximize the likelihood to end up in the right upper quadrant, it is best to distribute donations to both charities.
Note: there is in fact a third possible population ethic next to total and average utilitarianism: the person affecting view (see the population ethics trilemma). This results in a third axis in the above figure, vertical to the other two axes. When it comes to increasing the person affecting well-being (minimizing total maximum complaints or losses of utility), both saving children and preventing unwanted pregnancies are beneficial, with saving children through distributing bednets perhaps being a bit more effective. So when it comes to the person affecting view, there is no conflict between the two interventions. The arrows in the above figure all point in the direction of the third axis, towards an increase in person affecting well-being.