r.i.c.e.

Apparently in the U.S. today is “Giving Tuesday” – last Thursday was Thanksgiving, Friday was Black Friday for shopping at stores, Monday has recently been claimed as Cyber Monday for shopping online, and bringing up the rear is Giving Tuesday, where perhaps you are supposed to give away whatever money is left to some worthy cause?

This is news to me, but it gives me an opportunity to remind you all that rice has been recognized as a 501(c)(3) public charity in the U.S. and donations to us may be tax deductible (depending on, for example, if you itemize your income taxes).

It also gives an opportunity to share some thoughts about the ethics, economics, and psychology of richer people giving money to help poorer people (or not doing so). I recently recommended Joshua Greene’s new book on this blog. Among other topics, he considers the question of why richer people in the U.S. (by which we mean middle class folks with college degrees and such) don’t give more to help poorer people in places like Darbhanga, Bihar. One factor that his experiments seem to suggest is geographic distance: we feel (and he literally means feel) less of an obligation to people living very far away.

But that cannot be the whole story: in the big cities of India, there are many well-off people who live lives as comfortable as many middle class people in richer countries. They live close to (and regularly see) very poor people, but they still do not give in the way that Greene recommends. Lest I be misunderstood, this is not just an Indian phenomenon: picture Cape Town, South Africa, for example, or anywhere where the global top 5 percent live near the global bottom 20 percent.

It’s also not clear exactly how much people should be giving. To some utilitarian philosophers, the answer is quite a lot: if you really can save babies' lives in rural India by donating just a few hundred dollars, or even a few thousand dollars, how can you justify spending the money on even – say – Mexican restaurant dinners and Diet Coke? Once you start thinking like that, it is easy to conclude that folks like me in the U.S. are required to give, well, quite a lot.

But that is not what strikes us as intuitively ethically true. Many of us feel that our ethical obligations are strongest in our local communities, not where the need is greatest. If somebody gives a lot to help the very poor, we feel, that’s great – but not required. If so, Greene would respectfully suggest – perhaps with a disarming joke – that the problem is with our intuitive feelings.

Of course, there are many types of “foreign aid” – some of them do no good, and some do a lot of harm. Here, I’m not talking about government aid, but the sort of giving we each might do on a Tuesday. Angus Deaton has done much to persuade me that the argument falls apart if you can’t save a life for a few hundred or few thousand dollars, and maybe you just can’t. Or maybe, sometimes, you can – perhaps only in special cases that we have an obligation to look for? The more uncertainty is involved, the further down our utilitarian obligations slide.

As you may know if you’re reading this blog, I help run a small non-profit, and interact with a lot of large non-profits and famous international organizations. It seems pretty incredible to me that your additional $1,000 would have any effect whatsoever on what these big organizations do. Budgeting decisions are lumpy, by which I mean nobody is waiting around in a jeep at the petrol pump for the phone call to tell them that your donation has arrived, so now they can buy the fuel to drive out to the village to bring the anemic pregnant woman to the blood transfusion center and thereby save her life. I imagine that some of these organizations have large rooms full of fundraisers, perhaps with a target to raise $10,000 a month, and your $100 lets one take a coffee break a few minutes sooner. But, maybe I’m wrong.

Sanitation seems like a promising example. It is preventative, has important externalities, has been thoroughly well documented to matter since the 19th century, and is quantitatively a really big deal in some places. I think it is very unlikely that building some latrines is going to backfire and badly hurt people (and unfortunately we can’t say that about everything non-profits do). In places with very high population density like UP and Bihar, statistics suggest that it may be the case that if you can get 5 largeish villages to reduce open defecation rates from 75% to 25%, then you can expect to prevent an infant death each year, on average. But those statistics may be wrong, and – more importantly – I don’t know that anybody really knows a good way to reliably achieve that reduction around here. If we find something, you’ll be the first to know.

So, helping people is hard, perhaps especially by giving money from far away. Concentrating even on cases of health aid and where we think outright harm is unlikely, I have argued that institutional constraints suggest that the probabilities of helping could be much lower than we think. Ethics asks that we consider expected benefits – not best case scenarios – so it matters if the probabilities of actually accomplishing anything are small.

But I haven’t argued that the probabilities are zero. Experimental economists and psychologists have shown that people are very bad at thinking about small probabilities: we round both a one-in-a-thousand chance and a one-in-a-billion chance down to zero, even though these are actually very different. As Derek Parfit argues in Reasons and Persons, this sort of misleading intuition could lead to bad decisions, when you have to factor in really big numbers.

For example, let’s say on Giving Tuesday you decide to sign up to give rice $100 a month. (Thanks!) Now, for this example, let’s imagine that hiring new people is hard and we’re already pretty busy. So, let’s say that there’s a 1% chance that your extra money causes us to take on an extra project, report, presentation, or evidence-based advocacy agenda.

What happens if you get lucky, and we do expand our activities? We write some more op-eds and show some more power-point slides. Policy-making is complicated; politicians and bureaucrats have a lot to worry about, and not all of them care about the infant mortality rate (although I’m pleased to report that many of the ones who work on sanitation in India do). Let’s say that our hypothetical policy research that you helped cause has a 1 in 10,000 chance of achieving some sort of minor change in central government sanitation policy in Delhi.

Multiply those together, and that’s a pretty tiny probability of success: one in a million, to be precise. But there are over a billion people in India. As long as the hypothetical policy change makes people at least just one dollar’s worth better off, on average, then in expectation your donation created more value than it cost you. With different numbers, it could start to look like people like you and I have a pretty compelling obligation to give.

The final answer, then, isn’t something we can reason out from an armchair. It requires knowing something about how good we humans and our organizations are at helping. My guess is that, on average, we’re less good at helping than some philosophers believe, but better at helping than many people with money to give realize (especially if we donate carefully). If so, there may still be some good giving left to do.

Happy Giving Tuesday!