Wednesday, March 19, 2008

Tyler Cowen on Deflation

While grading papers yesterday, I listened to Tyler Cowen's EconTalk on monetary policy with Russ Roberts. Listening to Tyler's calm, soothing voice as he discussed all things monetary made the frustrations of grading all the more bearable. The discussion is definitely worth your time.

Amidst the Cowen-induced calm, however, I did have a momentary lapse into the land of angst when the topic of deflation was brought up. Roberts asked Cowen what would happened if the monetary base were frozen in a growing economy. Cowen correctly replied that there would emerge deflationary pressures. He then proceeded to say, though, that such a development would be a destabilizing outcome since people are not capable of handling a world of falling prices. Owing to "human irrationality", Cowen claimed worker morale would suffer if laborers got a 3% nominal wage cut even if the price level fell 4% in an expanding economy. Moreover, maintaining inflationary expectations among workers provides an "easy way to trick people into a wage cut" if needed.

Wow--is this really Tyler Cowen or his evil twin Tyrone Cowen? I certainly did not expect this response from Tyler. Let me begin my reply to whichever Cowen it may be by stating up front I do not advocate a freezing of the monetary base (although something like that happened after the U.S. Civil War). I am, though, open to a monetary policy rule that allows for productivity changes to be more fully reflected in the price level--George Selgin's Productivity Norm rule comes to mind--because doing so may actually improve macroeconomic stability (see here). Such an approach to monetary policy would also imply mild deflation at times, so Cowen's critiques still apply and must be examined.

So, Cowen believes people are subject to a form of money illusion that only allows them to be fully functional when prices are rising. In short, people are too dumb to distinguish between nominal and real values. This understanding is surprising for someone who champions a more libertarian view of the world, one where individuals are capable of making choices--such a distinguishing between nominal and real values--that improve their welfare. Empirical evidence suggests Cowen should take more seriously his libertarian instincts on this issue. Michael Bordo and Andrew Filardo in their article "Deflation in Historical Perspective" review the literature on this topic and conclude

... that such notions of downward wage inflexibility that were formed during the Great Inflation may in fact be regime-dependent. It is possible that once a low inflation or moderate deflation environment were to become more familiar, the past psychological aversion to downward nominal, rather than real, movements would become less of a constraint.

Even stronger evidence for regime-dependent framing of expectations comes from Christopher Hanes and John A. James in their AER article, "Wage Adjustment under Low Inflation: Evidence from U.S. History" From their conclusion:

We have looked for evidence of downward nominal wage rigidity in the nineteenth-century United States, using data that allow clear comparisons between historical and modem patterns. We find no evidence of downward nominal wage rigidity in the historical data, especially when they include the various monetary regimes of the 1860's and 1870's. One interpretation consistent with our results would be that the modern wage floor reflects employers' fear of damaging employee "morale" by violating social norms and concepts of fairness (as described by Bewley, 1999) rather than a fundamental preference on the part of workers. Unlike fundamental preferences, social norms can change with the economic environment. Under monetary regimes delivering very low trend inflation, such as the postbellurn deflation, a norm that enforced downward nominal wage rigidity could become costly for individual employers and employees, as well as for society as a whole… It seems safe to conclude… that U.S. historical experience fails to support the proposition that downward nominal wage rigidity is a fundamental feature of employment that prevails under any circumstances.

Cowen's view, then, of downward nominal wage rigidity is more representative of the inflationary times in which we live than a universal truth.

As I mentioned above, there are also reasons related to macroeconomic stability that should make Cowen more open to deflation, at least the benign form. That is why I like George Selgin's Productivity Norm rule. (In case you missed it, here are my postings on the relationship between macroeconomic stability and benign deflation.) From what I can tell, Cowen's colleague Bryan Caplan takes this latter point more seriously. Sometime they should get together and discuss George Selgin's "Less than Zero".

Update: G. Selgin makes an excellent point in the comments section--there is no need for a downward wage adjustment under the productivity norm. The decline in the price level will guarantee a rise in the real wage. Thus, the issue of whether wages are downwardly mobile is really inconsequential with productivity norm.


  1. I'm sure that the first sentence of your post will not freak out your students (too much). :-)

  2. Gabriel,

    I realy do enjoy teaching and most of my students. However, grading non-multiple choice tests--as I was doing while listening to EconTalk--can be a mind-numbing task.

  3. Even if Cowen were right, and wages were resistent to downward nominal adjustments, would it be so awful if workers experienced an increase in real wages as a result? One of the stories of the past three decades has been the flatlining or loss in real wages, which I would argue has contributed to the consumer debt binge. Let workers share in the productivity gains of the past decades!

  4. But we should be careful about dismissing money illusion. I seem to remember a paper on the stock market and inflation by Campbell and Vuolteenen concluded that the stock market underperformed during inflationary times because people extrapolated cash flows from low inflation times and discounted them using the (then) current high nominal interest rates

  5. David neglects to observe that under a productivity norm there's no need for general downward adjustments in nominal wage rates; and in doing so he seems to confuse the argument for such a norm with an argument for more rapid deflation. Tyler, on the other hand, may be wrong in supposing that a frozen base regime would result in deflation exceeding the rate of productivity growth. Whether it would depends on the nature of the banking regime. With sufficient deregulation, growth in the money multiplier and broad money stock could be sufficient to prevent any general need for downward wage rate adjustments.

  6. Excellent points G.Selgin. I should have been more careful in the original post. I made an update to the post that reflects these points.