Wednesday, September 23, 2009


Co-bloggers Arnold Kling and Bryan Caplan have been discussing Kling's "bizarre" macroeconomics.

Kling insists that the current recession (and perhaps all of them,) are due to a great "Recalculation." Sometimes, this amounts to a claim that there is a need to reallocate resources. Labor and other resources need to be shifted from some industries to other industries.

In macroeconomics, the unemployment associated with these sorts of shifts is called "structural unemployment." People are laid off in shrinking industries, and only very gradually find jobs in expanding industries. There can be a similar impact with capital goods, mostly involving a shift of existing capital goods to less valued uses, or even abandoning them as scrap, and only gradually producing new capital goods appropriate to the expanding industries. (The impact of "heterogeneous" capital goods plays a key role in the Austrian theory of the Business Cycle.)

Usually, the process of reallocation of resources is understood as being continuous, and the structural unemployment and losses from capital goods is one of the determinants of the potential output of the economy. For example, the higher institutional unemployment resulting from the sorts of labor market regulation common in Europe is a reason for their lower potential outputs. However, it is not difficult to see that a large shift in the composition of demand would result in larger shifts in resource allocation, higher structural unemployment, and slower growth in the productive capacity of the economy.

For monetary disequilibrium theorists, this would be described as a higher natural unemployment rate, along with slower growth in potential income. It is possible that in an extreme case, the productive capacity of the economy would shrink. However, as the adjustments are made, structural unemployment should fall, and the productive capacity of the economy should rise. Eventually labor will move, workers will be retrained, and new capital goods will be produced.

Monetary disequilibrium theorists, would insist, however, that this process could not possibly result in slower growth in nominal expenditure or nominal income unless it somehow impacts the quantity of money or the demand to hold money. And further, that if any changes in the demand to hold money are accommodated by changes in the quantity of money, then nominal expenditure and nominal incomes would not be impacted by these changes.

Suppose that such a policy exists. Nominal expenditure continues to grow, but the natural unemployment rate rises and productive capacity falls due to a need to redeploy resources. Then the shortages of goods whose demands are growing will result in higher prices for them. Substitution effects may dampen the decrease in the demands for the products of the shrinking industries, slowing the adjustment. The inflation--the higher prices--will also reduce real wages. This will dampen the losses in the shrinking sectors and so, their reductions in output. The lower real wages, as well as the higher nominal prices in the growing sectors will add to the profits there, both providing motivation and internal funding for a more rapid expansion.

As production expands in those sectors, rising supplies will push their prices back down. Any substitution effect that dampened the decrease in demand in the shrinking sectors is reduced. Real wages rise. That is no longer decreasing costs in the shrinking sectors as much. In other words, resources continue to be pushed out of the shrinking sectors. Further, the extra pull towards the growing sectors created by the temporarily higher prices and lower real wages also are smaller. In the end, the price level falls and real output rises, as do real wages. As a first approximation, the price level, real output, and real wages readjust to their previous growth path.

Looking at this scenario, it seems clear that avoiding monetary disequilibrium, and keeping nominal expenditure growing on its long term growth path, is the least bad option when there is a need for a reallocation of resources.


  1. >>Looking at this scenario, it seems clear that avoiding monetary disequilibrium, and keeping nominal expenditure growing on its long term growth path, is the least bad option when there is a need for a reallocation of resources.<<

    Is this just another way of saying that inflation causes fewer problems in light of nominal rigidities than deflation? Don't some of the of correcting factors you mention resulting in eventual reallocation of resources also occur (after some cost) where prices decline? Would it matter to you how much productive capacity was supposedly destroyed? If real output experienced a sudden 10% decline due to a dramatic shift in demand preferences, would 5% nominal GDP growth still be the least bad option?

  2. Dir:

    I favor a 3% growth path for nominal expenditure. This is equal to the the trend growth rate of productive capacity and should result in a stable price level on average.

    If productivity should rise faster than 3% for a time, then the effect would be deflation and a permanently lower price level---at least until there is a period of slow growth in productivity.

    If the increase in productivity were temporary, like an exceptionally good growing season, the the result would be deflation as real output grows, and then inflation when it returns to normal.

    Trying to create a shortage or surplus of money to manipulate nominal expenditures in order to offset these changes in the price level is an error.

    I don't think that deflation is always disruptive. For example, in the scenario in the original post, there was deflation as resources are sucessfully reallocated and real output recovers.

    I think it is obviously true that resources can be reallocated when nominal income falls. One of the processes is lower money wages and lower costs. The sectors that need to expand have their prices fall less, and those industries that need to contract have their prices fall by more. As nominal wages fall, the industries where prices have fallen less return to profitability first, and so expand at the expense of those industries that need to contract.

    Why anyone would consider it desirable to make these (or any) adjustments in an environment of monetary disequilibruim is difficult to understand.

    I think that if there was a really massive change in the allcation of resources, so that real output fell 10%, then maintaining nominal income growth would still be desirable. If, on the other hand, there is a permanent change in the trend growth rate of potential income, then a change in the growth rate of nominal expenditure would be in order.

  3. Thanks for responding. As you may have noticed, Kling added another entry expounding slightly on his idea today, and making explicit that in the short and "medium" run, changes in the supply of money are somehow neutralized by offsetting changes in the real demand for money. I also agree that it is hard to see exactly why he thinks this -- this is a more aggressive stance than merely claiming the Fed might for some period have trouble doing enough to counter an increase in the demand for money -- he seems to be claiming that an increase in the supply of money will itself result in an increased demand for money even if demand was not increasing to begin with.

    But taking that aside, it does seem like his view can be modestly restated as at least saying that it is merely difficult (not impossible) to overcome the putative inertia of how people value money because he acknowledges that strong enough action (a "regime change") could do the job of affecting nominal spending, the price-level, etc. He still doesn't go into detail on why he believes that growing money supply will be treated like a stock split, but maybe that's a sideshow and the real argument is more narrowly defined as "how hard is it to change nominal spending over X time period periods?" I think James Hamilton also makes this tamer version of Kling's argument in his response to Scott Sumner's article at Cato Unbound (analogizing nGDP futures targeting to Mars' orbit targeting).

  4. Dir:

    Thank you for you pointer to the new Kling post. I have been following Sumner's posts on Cato Unbound and the responses by Hamilton, Selgin, and Hummel closely.

    I find Kling's "pessimistic" timing theory peculiar. Changes in the quantity of money can only impact nominal expenditures after a lag long enough so that the impact of the increase in nominal expenditures only impacts the price level. A bit anti-fortuitous to my way of thinking.

    On the other hand, I think that really he believes that somehow nominal income changes dollar for dollar with productive capacity. Efforts to distrupt this happy market process, by changing the quantity of money, are offset by changes in the demand for money. If the Fed works really hard to disrupt the market, then it can cause inflation. Still, the natural state of the economy is for real output to stay at productive capacity and for nominal expenditure to adjust so this is consistent with the equilibrium inflation rate.

    If only it were true.

    To understand Kling's version of the liquidity trap, think about the impact of open market operations on both the quantity of money and the stock of T-bills that remain in the hands of the private sector.

    My view is that in the special case of an excess demand of both T-bills and base money, then open market operations in T-bills will not correct monetary disequilibrium. Kling, however, has occassionally suggested that open market operations in T-bills never impact nominal expnediture.