Saturday, July 16, 2011

More on the Long Run

This post is little more than free-association but after yesterday's post and doing some more reading I've been thinking a bit about what people mean by the long and short run.  I won't pretend to be consistent about this myself, but I thought it might clarify things a bit (I'm not sure whether I'm trying to clarify things for myself or readers of yesterday's post, bear with me) if I explained a bit about what I had in mind by the terms long vs. short run.

Short run is the most obvious, its the time period that we have good data on and can be reasonably sure of the impacts of our actions, generally several months to a few years.  Short run policies would be things like a WPA-style program of public works or playing around with interest rates.  With these things we can be pretty certain of the immediate short term effects.

Medium term is where things get hazy, but we have some reasonable models (or failing that, metaphors and analogies) to predict the impacts of our short term actions.  So we can create a government infrastructure bank or change rules regarding, say, housing, with a goal of changing investment patterns.  Generally, we're thinking 5 - 10 years out and a lot of academic bickering over whether or not the right model is being applied.  In many cases we can be reasonably sure that the impact of any intervention will be in the direction we want it to be, but we are often unsure of its magnitude.  So we can be reasonably sure that an investment in early childhood education will result in better educated more valuable workers later on, but we can't be sure that it will be good value for our money relative to other investments.  On a policy level, the medium term can also be thought of as the primary level where relative power positions matter, based on these policies there will be meaningful and fairly predictable movements in the relative positions of say China, the US, and Europe.  This matters for diplomacy and international negotiations of various sorts, the current short term statuses of these states (such as the economic downturn in the US) don't do a lot to alter the diplomacy (though they do some) but the impacts of these short term shocks on predictable medium-run concerns do matter.

The long term is where we don't really have the tools to influence or predict the consequences of actions, other than in some trivial or commonplace ways.  In the long run, individual policies don't matter, it's the cumulative impact of dozens or hundreds of policies that determine outcomes.  In the long run, we often see the unanticipated consequences of actions that looked sensible in the short and medium terms.  To give an example (this is drawing from a paper I wrote for a US economic history course as an undergrad, first thing that sprang to mind but I confess in advance there may be some inaccuracies in the following since I simply don't trust my memory that far back) we can look at the decline of the US steel industry.  In particular, US companies invested heavily in very capital intensive plants that represented the best technology on offer at the time.  However, organizational and technological developments ended up undermining this, the Japanese focused on a different type of steel plant that involved less upfront costs and ultimately proved more competitive (though their lower wages didn't hurt, but they retained competitiveness as wages rose).  This led, among some other factors (to expand on my earlier warning, I am vastly oversimplifying this story, the basics of US steel plants being locked into much larger facilities and thus relatively inflexible does stand up however) to a relative decline in the role of US steel production.  Of course, in the long term, US steel production did recover and we have a pretty good steel industry today.

However, within this long run there are many short runs, some of those short runs still impacting us today.  In particular, at its nadir, the decline in US steel production led to mass plant closures and unemployment among US steel workers, leading to blighted areas such as Gary, Indiana.  In the long run, of course, everything looks OK, US steel production became profitable again and recovered.  But, in some of these short runs, there was a lot of human misery caused because the owners of the US steel companies couldn't foresee the long-run shifts in production and operations, lives were lost, individuals ruined, etc.  All this would have been avoided if somehow the owners and managers of US steel companies had foreseen future shifts and not invested in these plants.  But, they couldn't.  The tools for making these predictions simply don't exist.  With few exceptions there is simply no way to know whether or not our plans for long run growth are those of the US or the Japanese, both evidently thought they knew what they were doing, but in the long run only one turned out to be right.

Which brings me to the kind of long run planning I believe we can do.  We can plan in the long run for a series of short runs, that we know will consist of periods of growth, periods of contraction, and periods that are just ho-hum and unremarkable.  We can plan in the long run to have the flexibility to address the different challenges of these periods.  When there's a downturn, when our long run plans go bad, we can have plans in place so that there's flexibility to address blight in cities like Gary.  We could have had plans to help people relocate and to wind down those plants more slowly.  What we can't do is make long run policies that will let us be the Japanese steel industry rather than the American, life's just too complicated for that.

This is what drives me crazy about the current calls for austerity.  It reminds me of stories about Ethiopian or Somalian famines where grain was stored in the country but wouldn't be dispersed (a more apt comparison would have been if they were refusing to take loans to buy more grain, but alas, the real world analogies I can think of don't always precisely match my rhetorical purposes).  The purpose of long run policies for growth is to address the kind of short term that we are currently in.  The call for austerity is like taking grain to store for a later famine, during a famine.  It's complete madness.  Undoubtedly policies pursued today will have unanticipated long term effects, many of them negative.  But anyone that says they know what these will be is selling snake oil.  Neither the history of business or governments is replete with institutions that were able to adequately plan for their long term growth.  Rather, the survivors bias is towards those institutions that act admirably in cases where the short run was uncongenial to them and that were able to survive them relatively well, to bounce back strongly when times were better.  Long term planning that goes much beyond simply hedging our bets for the unavoidable thin times is simply beyond us, and the more I hear about the long term in the midst of a terrible economy the more I despair.  Which isn't to say the long term, in the sense of things we do have somewhat under our control such as quality of labor force or infrastructure, shouldn't be thought of or invested for, but we do need to acknowledge that long term planning is really thinking about a series of short runs rather than as something truly distinct.  Human life occurs in the short run, the long run is simply an abstraction to organize our thinking and should not be privileged at the expense of the real suffering we see around us, especially since we have such little grasp of what influences things in the long run.*

*As a brief aside, look at the wide disparity in policies pursued in the developed world and the relative comparability in life styles across them.  To the extent that policies are impacting this, we are really only talking about very marginal differences.  However, we do see some really sharp short term fluctuations, such as the unemployment rate or the murder rate in the US relative to Europe in the 90s.  The degree of difference between the similar long run trajectories under different policy regimes and the short run noise that seems to occur is something that I think needs to be considered in these discussions.  Of course, at extremes long runs can matter, compare the developed world to Zimbabwe, but the set of policies under discussion, while containing a great degree of variance, isn't to such an extent that we're talking about a reversion back to arbitrary rule and cronyism.  The long term stakes of individual policy outcomes current in political debate are really quite small, but loom quite large in the short term for the individuals impacted by them.

No comments:

Post a Comment