See my posting below
What Dani Rodrik Taught Me About Industrial Policy for background on this one. In a nutshell, when resources aren’t allocated efficiently, structural shifts in the economy to higher productivity activities can create economic growth. Policy interventions to promote these shifts are justified only when we can identify a market and/or government failures that is inhibiting these shifts from happening on their own.
What happens when we have identified a list of both market and government failures? Dani teaches us to target the “binding constraint” – i.e. the failure that most inhibits structural shifts. If we don’t target the most important constraint, policies will be ineffective since other failures will continue to inhibit investments.
Though I will admit I have not yet been formally taught how to find a binding constraint, I have a feeling that the exercise of actually doing so is completely unrealistic. Is it realistic to think policy makes will have the data and capacity to find the single binding constraint?
And also wouldn’t the binding constraint to be different for different sectors as well as different firms within a sector? I understand the notion of one constraint being most binding in aggregate. But if the binding constraint differs by sector and entrepreneur, then policies that don’t target the overall binding constraint could still create growth even if not the theoretically maximum possible amount.
Dani is always taking about second-best economists – those that consider the reality of market failures or political obstacles and realize that theoretically first best policies won’t be effective. My gut is telling me the idea identifying a binding constraint might be a first best approach, unrealistic when we consider the information that’s actually available to policy makers. Not to mention the politics themselves.
