The US unemployment rate has dropped from 9.8 per cent to 9.0 per cent in the last two months, and there have been signs that private sector employment may soon be rising at about 200,000 per month. Admittedly, this improvement is still a very minor one compared to the massive deterioration in employment which occurred in 2008-09, when 8.5 m jobs were lost in the economy. But at least the change is now in the right direction. (See this earlier blog.)
With the labour market beginning to improve, some members of the FOMC are contemplating an early tightening in monetary policy. Indeed, the markets now expect the Fed to raise short term interest rates by 1 per cent in the next 18 months. If this goes much further, it could undermine the strength of risk assets, and possibly also of the economy itself. So it is crucial to ask whether there really is a genuine case for the Fed to become concerned about the tightening of the labour market.
For most members of the FOMC, the key to this question is the relationship between the actual unemployment rate and the theoretical rate which would be consistent with a stable rate of inflation (usually known as the natural rate or the NAIRU). The gap between these two numbers is an estimate of the cyclical unemployment rate. Other things being equal, when this gap is positive, inflation should be under control, and monetary policy can remain easy. This is what the current graph looks like:
Using the latest CBO estimate for the NAIRU, the “cushion” of cyclical unemployment remains exceptionally high at almost 4 per cent of the labour force. This is higher than in any previous post-war recession (except in 1982-83), so there appears to be no case for monetary tightening, at least from this source.
However, estimates of cyclical unemployment have recently been squeezed from two directions. First, the actual rate of unemployment has been dropping very sharply, with the decline over the past two months being the largest since 1958. Second, some recent estimates of the natural rate have been much higher than earlier estimates, suggesting that there is less room for the unemployment rate to fall before inflation kicks in. This is what is causing the hawks on the FOMC to express concern.
To start with the recent decline in the reported unemployment rate, much of this has come from the large number of people who are dropping out of the labour force. The graph below shows the 12-month change in unemployment, along with the contributions to this change coming from the rise in employment, and the rise in the number of people who have exited the labour force (both plotted with the scale reversed).
It is clear that the decline in unemployment has been accompanied by a very large number of people choosing to drop out of the labour force. This number is much larger than the usual leakage which occurs as retirement and schooling reduce the participation rate. It implies that there is a sizeable group of potential workers who could be induced to rejoin the labour market if the number of jobs really started to improve. This effect would hold inflation pressures down if employment growth accelerates.
Turning to the natural rate, I recognise that many economists are dismissive of the entire concept, because they believe that it is unobservable, and subject to enormous measurement error. However, the importance of the concept cannot be dismissed, if only because it plays a central role in the thinking of many members of the FOMC on both sides of the policy debate.
Recently, there have been several studies suggesting that the natural rate is higher than the CBO’s latest estimate of 5.2 per cent. Perhaps the most important of these is the recent study by the normally-dovish San Francisco Fed, which suggested that the natural rate might be as high as 6.7 per cent. If this is correct, then it would reduce the implied level of cyclical unemployment to 2.3 per cent, and the Fed’s margin of safety would be much narrower than suggested above. A handful of hawks on the FOMC clearly lean towards this line of thinking.
However, this is not the majority view on the FOMC. In the minutes of the January meeting published this week, the committee estimated that unemployment would in the long term converge on a rate which varied between 5.0 per cent and 6.2 per cent. Furthermore, following a special focus on this topic during the session, even the hawks conceded that any increase in the natural rate may in fact be temporary, and would disappear when the emergency extension of unemployment benefit comes to an end. Most of the members of the FOMC would therefore need to change their assessment quite radically before they would be willing to contemplate an early increase in interest rates.
I conclude the following. The US labour market has started to improve gradually, but not as much as is indicated by the recent decline in the unemployment rate. Such was the depth of the recession that the absolute condition of the labour market remains very depressed, and some discouraged workers might be willing to rejoin the ranks of active job seekers if employment growth accelerates. Meanwhile, the natural rate may have increased more than the CBO believes, but part of this is probably temporary. Therefore the safety margin of cyclical unemployment remains quite wide.
The only thing which would really change this conclusion would be an unexpected rise in the core rate of price inflation. The January CPI figures were a bit worrying on this score, and I will return to this important risk in the near future.
