Moral Marginalism: The Crisis of Foreign Aid

2 years ago Alger Mag Editor 0

Picture: Jess Seldon/Department for International Development

Picture: Jess Seldon/Department for International Development

There is something about the question of foreign aid that seems to bring out the utilitarian in everyone. These days, “aid” – be it public or private – is almost always discussed in the context of measuring its effectiveness. How many lives in Liberia can be saved per dollar of mosquito nets? Can de-worming keep more Kenyan children in school than providing textbooks? This kind of cold calculus, which I suspect most people would be reluctant to apply to ethical decisions in their personal lives, is pervasive in international development circles. Although it is no large sum, there is something about the flow of money to the global South that leads many thinkers and policymakers to demand a higher standard of “rigor” in their moral accounting. Development donors’ insistence on “Impact evaluation” is not just a management practice; it has become basis for an entire new ethos of giving, epitomized by the Effective Altruism movement.

The move toward quantifying the impact of development projects is far from a bad thing, of course. Though the statistical techniques that underlie impact evaluation (henceforth IE) are often misapplied or naively interpreted, they’re certainly an improvement over blindly funding projects based on vague metrics. There are good reasons why IE has been adopted by nearly all of the large development agencies. What’s more, thinking about one’s own personal giving through the frame of maximizing impact – the basis of the Effective Altruism movement – can be a step toward living a more intentional, ethical life. However, underlying both trends is a philosophy that I believe is misguided at best and counterproductive at worst. I call this philosophy moral marginalism, and while it’s not new (in fact, it’s a more than a century old), it’s stronger now than perhaps it has ever been.

Moral marginalism, in short, is the notion that the only thing that matters when making ethical decisions is the measurable impact of the last dollar or hour of effort that is spent. The average doctor in the US may save hundreds of lives throughout their career, but the marginal (last additional) doctor in South Sudan will likely have a greater impact, even if they save fewer people – there is no shortage of doctors in the US, but vanishingly few in conflict areas like South Sudan.

The Effective Altruism (or EA – bear with me, as development professionals love their acronyms) movement, founded by young philosopher William McAskill, seeks to create an orderly framework for measuring the marginal impact of individuals’ choices. The concept seems straightforward in the doctor example above, but when comparing different well-being outcomes, different ways of giving aid, etc., the task becomes considerably thornier. Working out how to weigh a set of diverse moral  “goods” – should we consider people’s well-being individually or consider distribution among the population? Should we care about people’s range of possible actions or just realized outcomes? How should we even quantify “well-being”? –  is a problem that has hounded utilitarians through history.

McAskill’s solution to this bind is to create a standardized measure of well-being called “quality-adjusted life years” (Qalys, of course). Qalys are based on surveys of self-reported well-being from people with a variety of ailments. A “baseline” year of life is one Qaly. A year lived with untreated Aids is 0.5 Qalys; a year with AIDS on anti-retrovirals is 0.9. A year of blindness is 0.4 Qalys, while a year on kidney dialysis is 0.56. Already it should be evident how this framework can lead to questionable decisions – as Amia Srinivasan points out in her great piece on EA, does this imply that is better to cure a 20 year-old of blindness than give a 40 year-old AIDS patient antiretrovirals, all else being equal? Moreover, as Amartya Sen has written, self-reported well-being can be a problematic measure for interpersonal comparisons.

But let’s take as read McAskill’s yardstick for goodness and move on to his prescriptions for a virtuous life. This is where he has attracted the most controversy – famously, he suggests that it is better to work in a lucrative profession like finance and give most of your money to a high-impact, low-overhead charity (like the Against Malaria Foundation) than it is to, say, work for an NGO abroad.

I won’t dwell on the pernicious social implications of McAskill’s ideas, which Srinivasan explores much more ably than I could –

“[EA’s] utilitarian calculations presuppose that everyone else will continue to conduct business as usual; the world is a given, in which one can make careful, piecemeal interventions. The tacit assumption is that the individual, not the community, class or state, is the proper object of moral theorizing.”

Instead, I want to focus on what the EA movement (and philosophies like it) has meant for practical policy-making in international development.

Impact evaluation – usually conducted through randomized control trials and similar quasi-experimental methods – is an ideal tool for realizing effective altruism’s goal of quantifying moral good. It’s little surprise that McAskill cites studies from J-PAL – the leading organization doing randomized control trials in development – to build his case for EA. However, this framework greatly constrains the kind of questions one can ask.

Randomized control trials (RCTs; sorry) are a good way to evaluate the individual-level impact of programs with clearly defined outcomes, like health treatments. Yet some of the most critical issues in developing countries are precisely the opposite – big, interrelated structural problems like lack of land tenure rights, under-collection of taxes and poor public infrastructure. These issues will never conform to the logic of comparing treatment and control groups, yet it is that logic which now drives the priorities of large institutional donors. USAID, the World Bank, the Gates Foundation and others have come to insist on RCTs as a “gold standard” of effectiveness for any projects they fund.

This feeds into the dysfunction that Srinivasan points out in her piece –

“The more complex the problem effective altruism tries to address – that is, the more deeply it engages with the world as a political entity – the   less distinctive its contribution becomes.”

While there is nothing wrong with quantitative methods of evaluation, if NGOs and governments let those methods drive the questions they ask and the programs they run, they miss something vital about development: The role of just institutions. In fact, EA can be used to elide the West’s responsibility for unjust power structures: 

“More worrying than the model’s inability to tell us anything very useful once we move outside the circumscribed realm of controlled intervention is its susceptibility to being used to tell us exactly what we want to hear.”

Indeed, EA’s popular success can largely be attributed to its easy compatibility with the logic of global capitalism – it’s little surprise that a philosophy that can reconcile doing good with making a fortune has become a favorite among tech donors like Sergey Brin and Larry Page (who hosted an EA conference at Google’s headquarters), Bill Gates and Mark  Zuckerberg. There’s no need to challenge the US’s destructive agricultural trade policy, for instance, when you can blame poor aid project management for the underdevelopment of foreign farmers instead.

Moral marginalism strips away all questions of collective responsibility in favor of an atomistic view of the individual. There’s no sense in feeling moral qualms about going into finance, McAskill suggests, since someone else will surely take your place if you don’t. There’s no sense in pushing for basic labor rights in the global South, tech giants say, since doing so will surely shift business elsewhere  – and shouldn’t those Bengalis be happy to have jobs at all? McAskill’s view of social welfare may be more Rawls than Pareto, but both his “scientific” benevolence and the dog-eat-dog ideology of entrepreneurial capitalism are grounded in a fundamentally individualistic notion of ethics. It’s for this reason that his ideas have found purchase amongst both moneyed private philantropists and DC think tankers alike. 

In short, moral marginalism reduces justice to an accounting problem. We can, and must, insist on something more revolutionary, and this begins with how we choose to measure the “good” we are doing. Perhaps aid agencies and institutional donors should base their efforts on Sen’s capability approach to ethics, in which freedom from domination in all its forms is considered, rather than just health outcomes. Doing so would necessarily require a broader epistemic framework for impact evaluation than statistical measurement alone, but that change within the discipline of economics is long overdue.

By: Anonymous