Saturday, September 19, 2015

Doing Good Better by Will MacAskill – review

I don’t give to charity much. It’s not for want of a compassionate spirit. Some of my greatest anxieties over the years have been whether I’m a good or, more frequently, a bad person. I’m sometimes inspired by the desperation of disaster appeals to part with the odd £50, but having worked for charities, and knowing people who work for them, and knowing a little about them, I’ve become incredibly dubious about the value of giving to them.

Will MacAskill is a bright spark. He’s a research fellow in philosophy at Cambridge, and an associate professor in philosophy at Oxford, and from his photos he looks about 17, so I’m going for bright (which I think we save for people we’re trying to patronise) rather than ferociously clever, but you can take your own pick. His new book, Doing Good Better, is full of great new ways of thinking, surprising conclusions and helpful advice, and it has garnered great reviews from other ferociously bright Oxbridge graduates such as the Guardian's David Shariatmadari.

While Doing Good Better is ostensibly about how best to contribute to charity (branded nowadays as "effective altruism"), it is more widely about how to make decisions in general and in this way it fits neatly into the nascent Freakonomics genre – lo and behold, there is a gushing quote from Stephen Levitt on the cover – in which writers apply economic concepts to unusual topics, and which, spiced with a slug of maverick thinking and a certain amount of journalistic sleight-of-hand, have sold shedloads. (Incidentally, Freakonomics as a term, I’ve recently realised, only really works pronounced by an American rather than a Brit, for whom Dreckonomics might work better, or possibly Feckonomics.)

MacAskill's book is replete with interesting ideas and eyecatching (and liberal-baiting) heresies: Fairtrade is rubbish; sweatshops are good; carbon offsetting is more than just nonsensical greenwash; working in the City (and then donating a portion of your wages) can do more good than digging wells in Africa; going into politics could do anything good at all.

The most essential tenet with which MacAskill sets out his stall is that we should do as much good as possible with our donations. Some good is not enough; we must strive to do the most good we can do. He then moves on to a set of questions we can ask when judging whether a charity, or activity, or job is going to provide that “most good”. The question of whether we must strive to find the most good is already a vexed one, and although it sounds superficially a useful idea, it begins to betray the huge number of assumptions that MacAskill rolls into his calculations once he moves on to making assessments. (I find it interesting that an ambitious high flyer should be so set on “the most” or “the best”.) As Amia Srinivasan in the LRB alludes to, however, this impulse to do the most good takes on its own momentum: if we’re into doing the “most” good with our money, why stop at 10% or 15% or even 50% of our income. Why should we have luxuries, such as shoes or toast? Why do we need chairs, when there’s people dying out there? We can sit on the floor. Do we even need to sit down at all? (Late add: read this)

It’s when we come to MacAskill’s framework for making choices that we face our own choice of how much to accept. He has his five key questions on altruism: how many people benefit and by how much; is this the most effective thing you can do; is the area neglected; what would happen otherwise; what are the chances of success, and how much success would there be – and in answering them, the full battery of Freakonomic jargon comes out to play: a rinse and repeat cycle of Qalys (quality-adjusted life year), marginal utility, expected value, counterfactuals and a lot of putting numbers to things that you wouldn’t imagine at first glance (or indeed second or third) could have numbers put to them. Micromorts – where the risk of an activity killing you is turned into a value – get an airing; taking ecstasy is worth one micromort apparently, since the risk of dying is one in a million; climbing Mount Everest gets 13,000. There’s an almost hilarious attempt to judge how much money an MP could influence in a career.

This quantification of the difficult-to-quantify harks back to John Stuart Mill and utilitarianism, trying to assess the greatest good for the greatest number. Putting numbers on things does help us make sense of the world. And you can be sceptical about it without necessarily writing the whole thing off. The early scientists might have found the idea of putting a number on the speed something travels as impossible, even undesirable; I mean, how would you even begin to think about going about it? So it would be churlish to say out and out that trying to assess different medical programs or charitable endeavours by measuring or predicting their outputs is so fraught and filled with gross assumptions that the numbers you get at the end, while they might help you feel like you can make a decision, are closer to a fantasy than a predictive map.

Of course, in the real world, we do these kind of gross assumptions all the time, and it has to be better than other means of making these decisions: asking a priest or consulting runes or taking acid in a desert cave and waiting to see what the wall paintings tell us. But as MacAskill mentions when he’s talking about Fukushima nuclear disaster, and as we also saw in the subprime mortgage fiasco of 2007, small assumptions (in those cases, that a tiny chance of failure equalled zero, which is fine and dandy until it suddenly doesn’t) when rolled into other assumptions and cooked in a lovely Qaly-marginal utility-expected value stew, can turn out quite convincing but entirely wrong results.

The expected influence of an MP is a great example of how nebulous the calculations can get. MacAskill divides up British government spending into portions based on who he thinks influences where the money goes. At the end of this, he decides that an MP can influence £8m. Because he’s made a lot of conservative assumptions on the way, he feels pretty confident about this figure. But conservative or not, those assumptions carry a huge forward weight on them. A slight error in one has a dramatic effect on the end figure. And do backbench opposition MPs really think they influence £8m a year? Isn't there a huge variation between them that makes the number meaningless?

This sort of thing is what I call, in the manner of MacAskill and Freakonomics, the spiral of assumptions. From tiny assumptions, colossal mistakes are made. MacAskill’s so confident in his calculations that he even rolls how uncertain he is into them. But how certain is he about how uncertain he is? We’re into known knowns and unknown unknowns, and look how that worked out for Rumsfeld.

There’s something quite neoliberal in the assumptions MacAskill has made – not just when he tells people to get jobs in the City in order to donate their wages, quite forgetting, in his calculations, to take on board the possible damage that a career in the City could do to those you’re supposedly doing it in order to help (he has apparently subsequently addressed this). The two key assumptions underlying neoliberalism seem to be to be: the best marker of value is price, and the best decider of price is the market. It’s in the game of sticking numbers on things. Sticking numbers on things that are hard to quantify seems like a sensible enough thing to try – you can always adjust a value if you later find out you’ve got it wrong. But when it leads you to take for granted all sorts of things that are speculative at best – and at its worst, to screen out those things that you can’t jimmy up a value framework for – then you might find the results are not what you expected. You might even be doing bad.

I hesitate to get too up in MacAskill's grill – as Kant used to say – mainly because he is a philosophy don at both Ox and Bridge and I am, well, not. But there's a couple of points in which I feel he's on shaky ground. The first goes back to doing the most good, which automatically means doing good in places such as Ethiopia, where your money goes a lot further. But there's a lot to be said for proximity – giving to causes both close to your home and your heart – and giving in such a way that you can see the results. Charity begins at home, after all. And there's something curious about people who state in all confidence that they care no more for their family, or neighbours, than strangers across the ocean. Something not quite right.

The second point is slightly too dull to go into at length here, but is to do with carbon offsetting and whether preventing further rainforest destruction (as the charity he recommends does) can really count as offsetting your continuing carbon output (at a time when scientists are calling urgently for "negative emissions", preventing other further emissions while adding your own seems the quintessential "greenwash"). But whether or not it does – I don't think so – the charity he recommends, Cool Earth, sounds great, and definitely seems worth giving some of your money to.

Doing Good Better is filled with similarly great recommendations and ideas, but there's something awry in the ultra-rationality and seemingly arbitrary valuations. It's good, in other words, but he could have done better.

This just in: While I was attempting to construct this robust intellectual critique, I chanced upon a recent article by MacAskill, W and MacAskill, A (wife? sister?) that attempts to use the familial utilitarian-logic-and-reason method to show that lions shouldn’t be allowed to hunt wild animals. With its abundance of reason and surfeit of sense, it manages to come off as a sort of masterful Voltarian satire; the kind of thing, so they would teach you in philosophy class, after which utilitarianism was never heard of again.