It’s Not That Complicated

It is a lesson that (development) economists need to relearn every generation: stuff is complicated and “best practices” don’t always work.  Keynes said it in 1951 in his Essays in Biography: “we are faced at every turn with the problems of organic unity, of discreteness, of discontinuity—the whole is not equal to the sum of the parts, comparisons of quantity fail us, small changes produce large effects, and the assumptions of a uniform and homogeneous continuum are not satisfied.” And long-forgotten development economist Albert Lauterbach emphasized a related point more than half a century ago: “the necessity of applying Western experiences and ‘rational’ methods only with great caution and with a real effort to understand the culture, value system and social structure of each population group that is to be exposed to a development policy.”  But in a well-meaning attempt to find universal solutions to (seemingly) intractable problems, the development industry does seem to like proclaiming new panaceas, however poorly the last panacea played out.

I’m a card-carrying member of the “stuff is complicated” party—it is one big reason why cross-country regressions haven’t taught us terribly much about the causes of economic growth.  If you need more convincing, read Lant and Justin’s brilliant paper noting that interventions “proven” to work by a randomized control trial in one setting don’t always have the same impact—or work at all—in another (start with Justin’s parable).

At the same time, it is important to emphasize that a counsel of complexity should not be taken as a counsel of despair for development—or even just for development agencies. First off, there has been immense global social and economic progress.  Child mortality worldwide halved over the last 20 years alone, and the number of low-income countries keeps on shrinking to the point that some aid donors are getting worried about lack of clients.  A further and underappreciated ubiquitous economic success: even countries with the fastest population growth have seen little evidence of declining returns to labor—in other words, everywhere has escaped the Malthusian trap.  Development economists may be unclear about all of the reasons why the world is getting healthier and wealthier, but apparently actual people worldwide have figured out a bunch of approaches that work.

Second, not everything is that complicated, even for aid agencies. As Lant and Justin make clear, concerns regarding external validity (that something which worked here works there) don’t apply universally. Boiling water at sea level in East Timor or East London will be about 100 degrees centigrade. Here are some things that aid agencies do or fund that work similarly across countries:

• vaccination research and rollout
• building roads according to international engineering designs
• building power plants according to international engineering designs

These things apparently work worldwide. Of course the power plants stop functioning for lack of fuel or spare parts, and the roads fall apart because of poor maintenance and overloaded trucks. Institutions matter.  And because of that, it turns out that none of this stuff guarantees per capita GDP growth in the slightest.  But I’d even suggest that some “institutional” solutions have a pretty good record worldwide: think avoiding hyperinflationary exercises in the printing of money or introducing competition in telecommunications provision.  There are still areas where there’s (broadly) a “right” way to do things. 

Of course there are also a lot of areas where context and complexity is central—anything that touches heavily on norms and culture and institutions.  And (again) in the aggregate it turns out norms and cultures and institutions appear to be really important to relative development outcomes at a given time.  But the third point is that, even there, “it’s complicated” shouldn’t be an excuse to give up.  Owen lays out a positive agenda for what complexity implies for development policy here.   And Lant and Justin have more to say in their paper and elsewhere.  In short, “it’s complicated” is a reason to try new things, partner not preach, and constantly learn.  Embracing diversity: it is a great strategy not just for migration policy and stock picking, but also for aid agencies.