See my posting below What Dani Rodrik Taught Me About Industrial Policy for a description on what allocative inefficiency is, where it comes from, and how industrial policy can improve resource allocation to create economic growth.

I felt it was important to make a couple points regarding the implication for overall allocative efficiency of policies justified by efficiency gains in specific sectors.  My last post explains how it might be easy to always justify government intervention for sectors of choice.  This post explains how in some cases, these policies could have a net negative effect on economic growth.

Allocative inefficiencies come from two sources: intra and inter sector inefficiencies.  Within sectors, we observe significant variation in productivity levels between firms.  Across sectors, well that’s pretty obvious.  We need to be weary of instances where these two are moving in the opposite direction, potentially leading to aggregate declines in total productivity and thus a negative effect of economic growth.

One example is trade liberalization.  Influx of imports can cause less productive firms in an industry to go out of business, which will improve efficiency within the sector.  But the question is: to where are these newly freed resources redirected?  While theoretically they go to higher productivity activities, if new industries don’t establish themselves due to market failures, they could remain idle and unemployed, or could be shifted to lower productivity sectors.  This is an obvious example of a net negative effect to the economy, even though productivity goes up in certain sectors.

A subtler example stems from my last posting.  While we might be able to identify an externality that justifies policy intervention to boost investment, what if the sector in question is less efficient than sectors that aren’t receiving policy support? Take for example Singapore, which heavily promoted high tech manufacturing in the 70s and 80s.  You could certainly justify this based on learning about costs externalities but if the productivity in the short run is lower than existing activities, which was the case in Singapore, the net effect to TFP will be negative.

The key point here is that when thinking about productivity gains from industrial policy, we need to make sure we’re thinking about from where resources will be shifted.  We can’t just look at any industry and justify intervention based on the presence of externalities or potential productivity gains in that sector.  We have to remember the big picture and ideally focus policy on the highest possible productivity sectors in the economy.  I have an unfortunate feeling that’s much easier said than done, especially when we are talking about promoting new industries whose costs are unknown.