What are remittances used for?

Here is a very thorough paper on remittances in Bangladesh. There's a lot of great info in it but I find Table 9 particularly interesting. I've written about remittances before here, and part of the point was that it was weird to expect poor families in developing countries to react to the receipt of remittances by investing all, or even most, of them. Understandably, low income people, just like anyone else, or hell, even more so, when faced with an increase in their income want to spend some of that extra money on consumption and things that make life worth living. The corollary being that all that worry about remittances inducing supposed "moral hazard" in recipients - making them work less or resulting in higher consumption rather than investment - is misplaced and, frankly, a bit patronizing. It demands of low income families in poor countries that they be more miserly and puritanical - save and work more (relative to their level of income) - than the people who have the fortune to live in richer parts of the world ever are (I mean, come on, look at US saving rates!).

But the data on remittances being way sketchy - mostly because a lot of them flow through unofficial channels, because poor countries which receive the inflows don't have the resources to keep track of them, and some of other statistical issues - it's actually pretty hard to say what remittances are being used for. That's where Prof. Siddiqui's paper comes in. Bangladesh is unique in that it actually has a government agency (BMET) which tries to keep close track of the inflow of remittances as well as, to some extent, on how these are used (and to its credit as a government agency, a significant part of its purpose seems to be to actually to figure out how best get the hell out of the way and just let the money flow). The main shortcoming here is that the remittances which BMET keeps track of are only the official ones sent by "short-term" migrants (mostly to select Middle East countries). So it misses both the remittances sent by migrants that wind up in Western countries as well as quite substantial flows sent through unofficial channels (and, roughly, estimates are that only about 50% of total remittance flows are through official channels). Still, even though the determinants of remittance flows may differ by host country and type of employment, it's fairly unlikely that their uses are affected by it. In other words there's no reason to expect that Table 9 in the paper does not accurately represent, at least approximately, the general use of remittances in Bangladesh. Here it is:



Here's my own quick aggregation into some basic categories:



What the table shows is the following:
1. To the extent that remittance receiving households spend them on consumption, it's mostly "basics" consumption. Almost 21% is spend on food and clothing. Only 6% is spend on other consumption (like furniture). And then 10% is spend on social ceremonies - weddings and the like. So out of total portion spent on consumption only 16.7% (that 16.7 of consumption, not total remittances) is spend on what could even remotely be considered "conspicuous consumption". Most of it is on necessities. And I hope nobody here's gonna begrudge anyone else a decent wedding. (Note: the amount of "conspicuous consumption" may be understated here since some of the unofficial remittance flows are in-kind)

2. The biggest investment categories are in land and migration. But this just means that the remittance receiving households are investing in the opportunities they are faced with. There probably aren't that many investment opportunities - for whatever reasons - in what folks in rich countries think of as "business". You invest in what makes sense in the economy you live in.

(Side note: I think this is a particular reflection of one of the themes in Dani Rodrik's book - for some countries the problem is not that savings rates are too low, but rather that saving rates are low because there's nothing to invest in. As a result, one would expect this tobe different for other countries. Different country, different recipe.)

And for people in countries which rely heavily on remittances, the investment with the highest return is to... send more people abroad to send back more remittances. Land (as well as construction on this land), on the other hand, is probably more of an investment in security. So taken together, these households are not just investing a lot, but they are investing wisely - they're diversifying their portfolios; spend some money on the high return and perhaps high risk investment of more migrants, and spend some of the money on the safe but low return of land.

3. Investment in "human capital" - health and education - is fairly low, though non-negligible. It's about 6% of the remittances. But here I can't help but shrug my shoulders. So what. The whole human capital thing has been more of a development-community/advisor fetish rather than something for which concrete evidence of return has been well documented (for example, see this classic). Given that these returns to human capital are fairly ethereal (and maybe ephemeral too!), 6% of remittance received spend on it is probably about right on the margin.

So the bottom line is that remittances are probably delivering lots of benefits and being used wisely, though not necessarily in the way that rich country advice-givers and would be do gooders would like. Tought nuggies.