As mentioned previously, we believe that further economic development, and general human empowerment, is likely to be substantially net positive, and that it is likely to lead to improvement on many dimensions in unexpected ways. This post elaborates on the reasons we hold this view and the implications of it.
We haven’t done nearly as much empirical research on whether this view is appropriate as we would ideally like to, and in the future we may approach it with a more concerted research effort. For now, we’d point to the following as broad defenses of this view:
- Since the Industrial Revolution, it appears that quality of life has improved in nearly every measurable way. A simple illustration of this idea comes from a brief recent post we made showing broadly rising per-capita income and falling infant mortality in the developing world. A more thorough discussion is available in chapter 2 of From Poverty to Prosperity by Arnold Kling and Nick Schultz, from which we’ve excerpted the key tables. (Note that this chapter isn’t our “primary source” for this claim; we have picked up various perspectives on this question from GapMinder, general discussions, etc. and point to this chapter merely as a relatively accessible summary.) The Better Angels of Our Nature, a relatively recent book by Steven Pinker, provides a deeper and narrower investigation of the effect of these changes on violence.
- The developed world appears to be better off than the developing world on nearly every metric we can think of, such as life expectancy and reported happiness, nutritional status (particularly for children), civil rights and human rights, and education (including for women and girls in particular).
- In the past, there have been many concerns about new technology making the world worse in some way, but these generally don’t seem to have panned out. For example, sulfur dioxide emissions, which cause acid rain, spiked during the mid 20th century in the U.S. but now are at much lower levels, along with most other pollutants in the U.S. Persistent worries about the mass unemployment effects of automation also appear not to have panned out, though they continue to be raised.
We don’t believe that avoidance of modernity-related problems can be taken for granted. In many cases it may take place because of concerted efforts to improve regulation and societal norms, and such concerted efforts may be needed to deal with various issues today. However, we also think it’s worth noting that concerted efforts to make the world a broadly better place seem to have become more common and more viable as economic development has progressed. Environmentalism, multiple civil rights movements, and large-scale foreign aid are examples of positive social changes that have emerged in the last two centuries and appear stronger in the developed world than in the developing world today. We’d guess that increased wealth and improved technology often improves people’s ability to coordinate around, and concentrate on, movements whose effects go beyond their personal lives.
- One of the most compelling cases for a way in which development and technology can cause harm revolves around “global catastrophic risks” such as climate change and nuclear war. However, from where we sit today, improved technology and economic development seem at least as likely to play a major role in mitigating these risks (via e.g. cleaner energy sources and more efficient overall economic activity to mitigate climate change, and greater economic interdependence and more effective security to mitigate military threats) as to worsen them. We will write further on this point in the future. It may be true that we would be safer from global catastrophic risks if we had never had any economic/technological development, but a faster rate seems safer than a slower rate from here.
We wish to note that we do not embrace the explanations for improvement sometimes associated with observations such as the ones above, explanations that often focus on the role of free markets to the exclusion of other institutions. We believe much of the improvement we describe may be attributable to the actions of governments, activist movements, and voluntary altruism (including philanthropy) as well as economic exchange. We believe that the exact dynamics by which the world has improved aren’t fully clear. My own take is that the concept of broad market efficiency is important here: as people become wealthier, better informed about each others’ activities, and generally gain more abilities and options, they become more empowered and motivated to tackle problems that they previously wouldn’t have been able to work on (or would have viewed as less pressing than other problems). To give a simple example, whatever good GiveWell does will be creditable partly to the huge number of other world improvements that have (a) given us the wealth and security to start a new venture; (b) given us education and tools to do our investigations; (c) addressed other problems that might have occupied our attention instead; (d) produced technology to run our lives and organization efficiently and find our audience (who themselves have similarly benefited).
If our overall view on this topic is broadly correct, it has some important implications.
First, it implies that a substantial part of the good that one does may be indirect: the people that one helps directly (by e.g. funding distribution of bednets) become more empowered to contribute to society, and this in turn may empower others, etc. If one believes that, on average, people tend to accomplish good when they become more empowered, it’s conceivable that the indirect benefits of one’s giving swamp the first-order effects.
If true, this is yet another source of noise (beyond the many we’ve identified) in formal cost-effectiveness estimates, and another reason not to take these estimates literally. It also implies that helping people who are well-positioned to contribute to society and/or help others is particularly valuable, relative to e.g. simple reduction of suffering for people who are not well-positioned to help others.
Second, it implies that helping to address any problem is a possible path to addressing many other problems. For example, if one’s only goal is to improve women’s education, it’s conceivable that the best option for doing so is to fund distribution of bednets (and if one’s only goal is malaria control, it’s conceivable that the best option is to fund women’s education).
Thus, even if one is convinced that a particular issue is the “most important” one to work on, this doesn’t by itself establish that one should directly fund or work directly on this issue. The nature of one’s practical opportunities matters greatly. If issue X appears to be of paramount importance, but issue Y has far more appealing giving opportunities for reasons related to room for more funding, one should consider donating toward issue Y.
Of course, a strong project aimed at the “right” problem is likely to have more impact than a strong project aimed at the “wrong” problem, but this isn’t always the choice that a donor faces (particularly a low-information individual donor). The details of what opportunities one has on each front are crucial.
So far, GiveWell has focused on the “easiest” interventions to have confidence in, figuring that being confident of accomplishing some substantial good is better than giving in an uninformed way, even if the latter is aiming at a cause that seems more important than global health. Going forward, we expect to be able to assess other ways of giving, from funding political advocacy to funding scientific research. But we expect to continue to put a substantial weight not just on the importance of an issue but on its tractability and its room for more funding.