Pent Up Wage Deflation

August 29th, 2014 5:34 am | by John Jansen |

Via the FT:

US economy: ‘pent up wage deflation’ blues

“Pent up wage deflation” is an unfamiliar and somewhat abstruse term dropped into the economic lexicon last week by Janet Yellen at the annual Jackson Hole conference. Originally coined by researchers at the Federal Reserve Bank of San Francisco, the term is destined to be widely discussed because it is clearly influencing the US Federal Reserve chair’s thinking. If it exists, it would explain why wage inflation seems abnormally low, given the recent rapid drop in unemployment, and that could eliminate one important reason for keeping US interest rates at zero per cent for the “considerable period” promised by the central bank.

Ms Yellen is right to be aware of the concept, and to keep it under review, but in my view the Fed is unlikely to shift in a hawkish direction solely because of it. This blog explains the theoretical and empirical reason why this is the case.

(Warning some of these arguments are quite intricate – skip to the end if you want to avoid the economic debate and just want the policy implication.)

Early this year it became clear that the Fed was placing increasing weight on US wage inflation as an indicator of whether it should be tightening monetary policy. The doves have been arguing that low wage inflation is a reliable signal that considerable slack still exists in the labour market, despite sharp falls in the official unemployment figures.

With wage inflation at 2 per cent, underlying unit labour costs in the US economy have been rising at no more than 1 per cent a year. This is only about half the rate which would normally be associated with the price inflation target of 2 per cent.

The doves have therefore concluded that an increase in wage inflation would actually be a good thing, and they want to delay rate rises until they see “the whites of the eyes” of inflation pressure in the form of rapidly rising wage inflation.

This reasoning would, however, be blown out of the water if the relationship between wage inflation and economic slack has itself started to behave abnormally. Under these circumstances, the Fed might be misled by the low rate of wage inflation, and might keep interest rates too low for too long.

As Paul Krugman of The New York Times has pointed out, it is intellectually honest for Ms Yellen to recognise this possibility. And this is where pent up wage deflation enters the picture.

There is plenty of evidence that there is great resistance in the US labour market to cuts in nominal wages, even in recessionary periods with very high unemployment. In such periods, a very large proportion of workers (16 per cent in 2011) experience unchanged nominal wages, instead of the wage cuts which would be required to attain equilibrium in the labour market.

These downward wage rigidities imply that overall wage inflation remains higher in recessionary periods than the rate that would emerge if nominal wage levels were fully flexible below zero for all workers. The illustrative graph above suggests that this probably happened during the Great Recession from 2009-11. (In the graph, actual or observed wage growth in zone A was higher than the underlying wage growth which would have occurred in the absence of nominal wage rigidity.)

The opposite applies during recoveries (zone B). As demand rises, and unemployment falls, there is no need to raise nominal wage levels for those workers who had avoided wage cuts during the recession. Therefore unemployment can fall, reducing the amount of slack in the labour market, without this being signalled by any rise in wage inflation.

This is what Ms Yellen is now starting to worry about. It would suggest that low wage inflation may not in fact be providing a reliable signal about labour market slack, implying that policy might be tightened too late.

So does pent up wage deflation mean that US interest rates should be rising soon, after all? There are two reasons for saying no.

The first reason is theoretical. We have seen that under conditions of downward wage rigidity, wage levels are higher than they should be during recessions. This causes unemployment, which can be regarded as “involuntary”, given the rigidities in the wage-setting process. The economy is clearly not in equilibrium at the end of the recession.

During the subsequent economic recovery, the low level of wage inflation enables the labour market to return towards equilibrium by raising employment levels. Only by the end of the process is the real wage level, and the employment level, restored to full equilibrium.

It would be unnecessary and damaging for the Fed to intervene in this equilibrating process by raising interest rates before it is complete. If it did this, the Fed would simply be accepting a higher level of unemployment than is needed to hit the inflation target, which is clearly counter productive.

The end of the equilibrating process is signalled by a rise in wage inflation, which admittedly might be more abrupt than it would be in the absence of wage rigidities. Ideally, the Fed should only act when this is about to occur (allowing for the normal lags between monetary policy and the economy).

The second reason is empirical. If pent up wage deflation is a significant factor, it would be expected to have the greatest effect in those parts of the economy where the recession had been most severe. According to Jan Hatzius at Goldman Sachs, there is no evidence that this has been the case. In fact, the reverse is true. Wage inflation in the recovery has actually risen slightly more in those industrial and geographical sectors of the US economy where wages had been hardest hit during the recession.

As Mr Hatzius concludes:

This is the opposite of what the pent up wage deflation hypothesis would predict… Our results are hardly the last word… but our preliminary takeaway is that the pent up wage deflation hypothesis is not a good reason to abandon our view that the continued weakness of wages signals continued labour market slack.

In the absence of further research, the Fed doves are likely to agree with this conclusion. Newfangled arguments about “pent up wage deflation” will not be enough to make them abandon their dovishness.

Be Sociable, Share!

Post a Comment