Will a jobs squeeze choke America?

FT.com

One of the great constants in the world economy in the past few decades has been the consistently strong growth in the US labour force. This has given American economic performance a demographic head start compared with other developed countries. Not only has it been the main factor ensuring that US GDP growth has remained well above that in Europe, it has also injected flexibility and dynamism into the US economy. But all of that is now at risk. The US labour force suddenly stopped growing in 2008, and has been falling slightly ever since.

As a result of this sudden disappearance of growth in the labour force, the unemployment rate has fallen by 1.5 percentage points in the past two years. But it is doubtful whether this represents a genuine tightening in the labour market. More likely, the underlying growth in the labour force has been disguised by the fact that potential workers have been discouraged from remaining in the labour market by the shortage of job opportunities. Without this shrinkage in the labour force, the unemployment rate would have risen to more than 11 per cent by now. It is urgent to fix this problem before the labour market atrophies, as it did in Europe in the 1980s.

In the long term, the growth of the labour force is determined by the growth of the population of working age, and by the behaviour of the participation rate. In the US, the growth of the population has generally been about 1-1.5 per cent per annum in recent decades, partly driven by natural growth, and partly by net migration. In recent years, the population growth rate has fallen very slightly (to about 0.9 per cent per annum), but this does not account for the sudden change in the growth of the labour force.

The factor to blame has been the participation rate. Prior to the late 1990s, this rate had been rising continuously throughout the post-war period, driven mainly by a rise in the number of (mostly) married women who wanted to find jobs. Hence, the labour force grew even more rapidly than the population of working age. However, in the last decade, the participation rate for women has hit a ceiling. The reasons for this are not entirely clear, though we can all speculate about the change in social behaviour which might have caused it. In addition, young people have been choosing to spend longer in education, possibly reflecting the rise in the relative earnings of educated workers. These two factors taken together caused a gradual drop of around 1 percentage point in the participation rate from 1998-2008. (See this analysis from the Federal Reserve Bank of San Francisco.)

Since then, the participation rate has fallen precipitously, by over 2 percentage points in just three years. Long-term factors cannot explain this sudden change in behaviour. Since the break in trend occurred exactly when the recession began, by far the most likely explanation is a severe shortage of jobs caused by the depth of the recession. Disenchanted with a lack of job opportunities, people have simply stopped looking for work.

This has also happened in previous recessions, but the extent of the decline in the participation rate this time has been unprecedented in the post-war period. The graph below shows the behaviour of the labour force before and after the recession ended (June 2009=100), and compares this with the same stages in previous cycles.

It is clear that the stagnation of the labour force in the latest cycle represents a very different pattern from that seen previously. What are the consequences of this?

The first is that the measured rate of unemployment has fallen to levels which are far lower than would otherwise have occurred. In the past three years, about 4.8 million people have disappeared from the labour force instead of becoming 'unemployed', and this has allowed the unemployment rate to drop from a peak of 10 per cent in October 2009 to 8.5 per cent now. If these people had stayed in the labour force, instead of drifting away from it, the unemployment rate would now be about 11.3 per cent, and the whole debate on US economic policy would look very different.

The second consequence is for the Fed. If they view the decline in the labour force as genuine and irreversible, then they would have to conclude that the labour market has tightened considerably in the past two years. They might still believe that there is plenty of excess or 'cyclical' unemployment left in the economy, thus justifying their stance on monetary policy, but they would think that the labour market, which constitutes half of their dual mandate, is at least heading in the right direction. And they would be more disposed to worry about signs of a rise in core inflation, such as were seen last year

On the other hand, if they view the decline in the labour force as reversible, they would conclude that the amount of slack in the labour market is probably much larger than the unemployment statistics indicate. If the economy were to expand more rapidly, bringing the missing 4.8 million workers back into the jobs market, the rise in the labour supply would add about 2 per cent to potential GDP in the US economy. This would greatly dampen inflation risks during an economic recovery.

And if they do not act quickly enough to restore job opportunities to these missing workers, this 2 per cent of potential GDP might be completely lost. There would be a rising risk that the absentee workers would become too old, or too disenchanted, ever to return to active participation in the labour market. This is what happened in Europe after the recession in the early 1980s, when a cyclical rise in unemployment became permanently baked into the system for the next two decades. Economists call this hysteresis.

Americans are usually quite disdainful of the European economic model. Hysteresis is one feature of the European labour market which they should try hard not to imitate.

Copyright The Financial Times Limited 2012.