Tuesday, January 19, 2016

Oil and the paradox of toil

I think it was in the fall of 2009 that I first stumbled upon what I coined the "paradox of toil", see e.g. a 2010 WP version here, sort of playing on the old paradox of thrift, that if everybody tries to save, there will be less aggregate savings.
The point was really pretty simple. It was that in economies stuck at the zero lower bound (ZLB), then what really matter is aggregate demand not aggregate supply, so you want to focus on things that increases demand. If the problem is that people are not buying enough things, it's not really going to help adding production capacities if the existing ones are not even being used (if a farmer that can't sell all his products, what is the point of adding a new tractor or a new worker?).
There was another perhaps somewhat more subtle point the paper made, hence he paradox in the title. It was that if you increase factors of production under these conditions -- e.g. if everybody wakes up and want to work more -- then you may in fact end up reducing equilibrium production -- i.e. everybody works less in equilibrium.
The logic is straight forward in most monetary models, and since then has been found to apply pretty broadly (when stating the paradox I left it open if this was a bug or a feature of New Keynesian models). At the ZLB the real interest rate is too high, i.e. it is above the natural rate of interest. Stuff that increases supply will tend to reduce marginal costs of firms (e.g. wages in the example above if everybody wants to work more) and thus trigger deflationary pressures, thus increasing the real interest rate, which is exactly what you don't want because the natural rate of interest is negative and the central bank can't cut rates. Higher interest rate reduce people incentive to spend and invest. So there you have it: Everybody wanting to work more then implies less actual work in equilibrium. This is the paradox of toil.
I talked about many other examples in my old paper that may have such an effect, most notably temporary drop in oil prices. Looking at the world today, it hard to think that the recent fall in oil prices has been particularly helpful for the world economy. People have typically focused on the disruption these drops have on the oil producing firm, and the banks that are lending them. Surely, these consideration are quite important.
The paradox of toil, however, points to a more general problem, namely that these fall in oil prices are generating an overall pull downward in current and expected inflation, thus effectively making monetary policy more restrictive around the world by increasing real interest rates. It seems that inflation expectation are falling across the board in industrialized countries, at least by some measures. It is hard to see how further deflationary pressure are helpful at this stage. So perhaps the current collapse in oil prices will prove to be a nice laboratory experiment for the paradox of toil (although as always in economics, there are a lot of things happening in the same time, so it's not that we have a clean identification here).

PS. These events remind me that perhaps it is time for me to actually try to publish this paper. I think the path this paper took is a classic example of the dis-functionality of our profession (and to some extent a reflection of my own procrastination). I submitted it to what we refer to as a top five journal way way back -- where it sat for about two years -- ultimately to be rejected on first round. As I got it back about two years later, there was already a modestly lively literature that had grown about the paradox, in order to prove it or disprove it in different setting, ranging from things like arguing non-linearities matter and make the paradox disappear, that the price setting assumptions in the paper was driving it, or on the empirical side that seasonal fluctuations, oil price variation or earthquakes proved it wrong. Ironically, I think some of that work was even published in the journal that rejected the first paper in the first place. But in any event, the enormous lags we have in the profession in terms of reviewing (which I am afraid I contribute to as a slow reviewer myself), just meant that when I finally got it back I clearly needed to address the growing literature the paper had generated (some of which I had in fact contributed to and published myself by analyzing contractionary effect of tax cuts (2010 NBER Macro Annual), the expansionary effect of the New Deal (AER) and structural reforms in Europe (JME)) and I felt most objections could be addressed in a straight forward way. But at that point, I had several other things cooking, so I never got back to it. Perhaps it is time now.

No comments:

Post a Comment