« Categorizing Has Consequences | Main | Sneaking in Connotations »

February 19, 2008


Isn't this just a special case of the general rule that we are more generous with those close to us?

Because we "give" to poor people for two reasons, and two reasons only.

1. To feel good about ourselves and how we in the rich world "take care" of poor people. (We don't, obviously, but that's not the point. Most people actually believe foreign aid is helpful.)

2. As a government funded support to domestic companies. With WTO, EU, NAFTA and other trade rules, governments can't really just give money to companies anymore. "Foreign aid" is the easiest way to circumvent that problem.

A far weaker reason, but probably not completely irrelevant, is pure rasicm. "These poor foreigners really don't know what's best for them, so it's a good thing we can direct them".

Err.. what about the Bennet entail in Pride and Prejudice, not to mention all the tales of the wealthy making their offspring miserable by imposing conditions in them inheriting?

Paul, I'm not talking about the amount of the gift but about the strings attached.

John, I'm talking about real people today.

Joe, why do we feel better about strings for Nepal than for nephews?

Wait a second, Robin. In your example, you're not giving your nephew money "with no strings attached." You're offering to pay for college. You're funding a specific project. The contrast you're trying to make doesn't ring true.

Let me ask you: why would you decide to pay for college instead of just giving your nephew an equivalent pile of money? Some of those reasons may help explain why people pay for aid projects rather than sending equivalent piles of money.

Nigella Lawson? http://www.guardian.co.uk/commentisfree/2008/feb/01/comment.money

How about this?

We know our aid to Nepal is a drop in the bucket compared to what they need to bring them up to our level. If we give them money, then, we will have no idea how it's spent, and their level of material well-being will appear not to improve at all. But if we build them a bridge, we can point to the bridge as something *we* did for them, and we can see how the bridge increases their well-being.

That is: as Joe T. says in comment 2, we give to feel good about ourselves. This gives us something concrete to point to.

Here's a possible explanation: We have a certain set of values and we would like our charitable donations to support those values. In general, our close relatives share many of our values, so we can give them money with relatively few constraints. On the other hand, random strangers may not share our values, so we need to impose constraints, in order to ensure that our values are supported. Those cases in which we impose constraints on our relatives are exactly those cases where we are not confident that the relatives share our values.

I second Ping here. Maybe we would have a better society if we just gave the nephew money, but we don't even come close to doing so. Anyway, a very large fraction of our foreign aid goes to the purchase of things that are not consumer goods. As givers of vaccines, for instance, we respond to the externalities of immunization, not just to the direct effect on the recipient.

"Why do we feel better about strings for Nepal than for nephews?"

Well, we don't. We don't care about strings either way. What counts is:
1. That it seems like we do something. Then it doesn't matter if there are strings attached or not.

2. That what we actually do benefit domestic companies. Then strings are completely mandatory and the whole point of the endeavour.

Part of it is probably fear of corruption: give a poor country $100m, much of if may be stolen by officials. Build 1000 wells, and there are 1000 wells. Or at least that's the theory.

Aside from the aid being seized by the government (even if you gave cash payments to individual Nepalese, the government could respond by increasing effective rates of taxation), cash aid also creates increased dangers of citizen rent-seeking, e.g. criminal entrepreneurs have greater incentive to acquire fungible cash than bednets. By targeting a narrow class of goods like vaccines for a particular disease one may flood the market, preventing major rent dissipation.

When we give, why do we interfere so much more with distant poor, and interfere so little with those close to us?

One reason is that we have a lot more indirect pressure on those close to us; just giving the money to someone close to us pressures them to use it in ways we'd approve of (the words "son, make us proud" is a very powerful chain).

As for why we interfere with the distant poor, the reasons mentioned here are good. I'd add two more: there is a strong meme that aid doesn't work, or that the money is wasted. In that situation, one who doesn't want to see their donations wasted (ie most people) will either refrain from giving or give only with stringent strings attached.

The second is that our valuations of things (for ourselves, but especially for other people) does not follow the market. A mosquito net may cost the same as a bottle of whiskey, but we don't value them identically for others. Hence, strings. (extreme example of that: we tend to plan for the long term for others much more than we do for ourselves)

Finally, I'm surprised that no-one asked the question: is giving aid with strings attached more effective than with no strings, or not?

or, indeed, whether mosquito nets with strings attached are more effective or not.

When we give, why do we interfere so much more with distant poor, and interfere so little with those close to us?

How do you know that?

When we give, why do we interfere so much more with distant poor, and interfere so little with those close to us?

You trust people you know more than you trust people you do not know.

The "you" in the first paragraph is an individual. The "we" in the second paragraph is government. It doesn't get more apples and oranges than that. Granted, the "we" in the second paragraph might be a nonprofit group, but it's still a group.

For one, I will second Ping. I don't see how funding a college education is "no strings" money, especially since you yourself mentioned GPA and no off semesters. On the other hand, in both cases what we want to do is control the way our money is spent. In case of nephew we want to make sure that the money goes to give him a college diploma (hence paying tuition instead of giving him money) and college education (hence minimal GPA restriction). In case of remote countries we want to make sure that something useful to the actual poor people gets done, a hospital built, food bought etc. and that these money don't buy a villa in the Caribbeans for some government official.

My issue with shooting free money down the pipe is accountability. In underdeveloped regions, there is likely to be overt corruption, and I have a (reasonable, I think) suspicion that string-less money will find itself funding some despot's harem instead of improving access to medicine or potable water.

If I'm funding my bright nephew, let's say for the sake of argument that I really do just give him a wad of cash, I can expect by virtue of my high opinion of him, that he'll act wisely with it. If I don't have a high opinion of his fiscal responsibility then sure, I'll put it in a silo for college.

...just like I'll put Nepal money into a special "potable water" silo because I don't have a high opinion of Nepal's ability as a region to keep a local official's grubby paws off of it.

Without sounding too heartless, could it be that the poor have proved themselves inept with money, while our promising nephew has not. And I would second Ping, paying for college isn’t exactly like handing over cash.

On an unrelated topic, is Cuba the ready for Futarchy? I would think it would be best to experiment with a government that was A) in transition and B) very bad to begin with.

See Bruce Bueno de Mesquita on foreign aid.

Are you suggesting something along these lines?

If you drop cash on a bunch of Nepalese peasants, what do you think they are going to do with it? Truck on over to Walmart? Order from Amazon?

I'd like to challenge the notion that we treat poor people in our nation with less interference than the Nepalese. Have you ever seen a welfare eligibility determination system? They are notoriously complex, requiring tons of programming effort to adjust every time Federal and State regulators adjust aid policies to the poor. No conditions that we impose on foreign aid approach the conditions we impose on our own poor.

I'm in favor of our current policy of imposing greater restrictions on the domestic poor, given the sums of money involved.

Our "Official Development Assistance" is the best figure for total non-military foreign aid. It was $19 billion in FY04: http://digital.library.unt.edu/govdocs/crs/permalink/meta-crs-7336:1

In FY00 our total spending on social welfare was about $1 trillion. Excluding entitlement programs, about $430 billion of this was targeted to the poor: http://www.libraryindex.com/pages/72/How-Much-Does-Nation-Spend-on-Welfare.html

John, I'm talking about real people today.

The entail critique is more legitimate than you think in that the US at least doesn't recognize entailments beyond one generation (and I think the same is true of GB today).

We don't really know how many real people today would put such things in their wills if they were enforceable. But many real people certainly *did* so back when it was.

The comments to this entry are closed.

Less Wrong (sister site)

May 2009

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30