The language of "growth diagnostics" and of "removing binding constraints" is becoming so commonplace in multilateral agencies and donor organizations these days that I sometimes wonder whether we have not unleashed something out to the real world before its time. The trouble is that what I see being implemented in practice is often the rhetoric and not the substance.
That is because the framework cannot be applied mechanically and requires an inquisitive, detective's mind-set. You need to use economic theory and evidence judiciously to look for a series of clues that will identify the most likely suspect. So while the approach comes with a decision tree (reproduced below), which probably accounts for its good reception in policy circles, it is different from just checking a series of boxes--which is what is often done. There is an element of craft in doing the diagnostics right, but it is a craft solidly based on economic science.
Here are some of the hallmarks of a successful growth diagnostics exercise:
- moving downwards in the decision tree, rather than upwards or sideways (you move by eliminating constraints, not by considering them all one by one);
- working off at least an implicit model of what drives (or will drive) growth in the economy;
- looking for the tell-tale symptoms that a given constraint binds (if growth is constrained by saving, the economy should be running against its external balance constraint, interest rates must be high, and borrowers must be chasing lenders rather than the other way around; if the constraint is human capital, the skill premium must be rising while returns to complementary factors remain depressed)
- looking for clues that the hypothesized constraints are consistent with recent growth experience (i.e., did growth boosts occur when those constraints were relaxed? a constant cannot explain a change);
- using firm-level surveys critically, cognizant that complaints do not always accurately identify binding constraints (businesses may complain about access to finance when the real trouble is that they cannot document profitable projects; or respondents may be the established firms that do not represent the most dynamic part of the economy);
- locating successful firms or sectors and tracing their success either to their low intensiveness in the hypothesized constraints or to special circumstances that allowed these activities to overcome the constraints;
- combining cross-national benchmarking, firm-level surveys, and aggregate macroeconomic data in an eclectic manner as the nature of the question demands.
That the framework has to be applied with care and in an economically sophisticated way rather than mechanically is also the main message I take from some of the critiques that have begun to appear (see for instance the papers by Dixit and by Aghion and Durlauf). The framework does not economize on inputs (the thoughtfulness required to reach decisions), only on outputs (the list of things that we recommend governments should do to get growth going).
Encouragingly, there are some quite good exercises along these lines, showing how the framework can be used to good effect and to shed light on the requisite policy agenda. See for example the studies on Egypt by Klaus Enders, Bolivia by Sara Calvo, and Mongolia by Elena Ianchovichina and Sudarshan Gooptu.
I know from my own experience that there is still a lot that we need to learn about how to do this right. To those who say that the framework is difficult to implement in practice, my answer is "right, it is indeed hard to determine policy priorities, but this approach at least forces you to confront those difficulties in a systematic way."