Possibly the most important debate about the US economy over the last couple of years has been focused on the issue of unemployment. Since the financial crisis, the unemployment rate has stayed stubbornly high and is presently close to 8%. From a policy making view, it is vital to understand whether this high unemployment is the result of structural or cyclical factors, as this distinction entails very different policy measures.
Briefly, the structural view of US unemployment goes along the lines that the financial crisis damaged the potential of the economy to produce goods. For example, this view might say that the construction boom before the crash meant that too many construction workers were employed and now they are essentially useless to the economy. Alternatively, the cyclical story is that the unemployment is a demand side phenomenon, rather than the supply side story outlined above. This story says that the financial crisis did not damage the potential of the economy in a major way, so there is now a large output gap – in other words, the economy is performing below potential and it’s the lack of aggregate demand that is the primary cause.
If the story is one of lack of demand, the correct response is stimulus, either from the government or the Federal Reserve (there is a big debate here about the merits of each). If the story is a structural issue, then stimulus cannot be used as it will lead to high levels of inflation. This touches on a theoretical economic concept known as the ‘NAIRU’, or the non-accelerating inflation rate of unemployment. This is the level of unemployment where inflation will remain stable – which is ideally where the Federal Reserve would like inflation to be. If unemployment is pushed below this level, say through government or central bank stimulus, then accelerating inflation will result. If unemployment is below this level, there will be downward pressure on inflation.
So, basically, if unemployment is high and there’s an output gap, then stimulus will not lead to runaway inflation. If unemployment is high, but there’s no output gap, then stimulus will lead to inflation.
Has the NAIRU changed?
A recent analysis by Rand Ghayad and William Dickens has presented evidence that supports the lack of demand story. They have used disaggregated data on the unemployment-vacancy relationship, the Beveridge Curve, to show that the high level of unemployment is not due to a skill mismatch between the skills that workers have and the skills that employers want.
How did Ghayad & Dickens do this?
Ghayad & Dickens used the Beveridge Curve, introduced as follows:
The Beveridge curve – the empirical relationship between unemployment and vacancies – is thought to be an indicator of the efficiency of the functioning of the labour market. Normally when vacancies rise, unemployment falls following a curved path that typically remains stable over long periods of time. When vacancies rise and unemployment does not fall (or falls too slowly) this may be an indication of problems of structural mismatch in the labour market leading to an increase in the lowest unemployment rate that can be maintained without increasing inflation (the NAIRU).
The concern with the Beveridge Curve is it’s present state, seen below. Since September 2008 the curve has shifted out, rather than moving back up the red line as unemployment has fallen slightly. This has been interpreted as a structural issue in the labour market, but cleverly, Ghayad and Dickens looked at disaggregated data to show this interpretation to be premature.
The diagnosis of a structural issue via a skills mismatch was not supported when Ghayad and Dickens disaggregated the data by length of unemployment. The following two charts show the Beveridge Curve for those unemployed for less than 27 weeks and those unemployed for greater than 27 weeks. The first chart reveals a consistent Beveridge Curve, with no signs of this break. The second chart, for those long term unemployed, shows quite strikingly that the break in the overall Beveridge Curve is being driven by the data from the long term unemployed.
Further emphasising these findings, Ghayad and Dickens also found that the break seen in the aggregrate Beveridge curve was also evident when looking across age groups, industries and type of occupation. The conclusion from Ghayad and Dickens is, therefore, a very important one:
We conclude that any explanation for the change in the vacancy unemployment relationship must account for its pervasiveness across different industry, blue-white collar occupations, age, and education groups, its concentration among the long-term unemployed and its absence from the short-term relationship.
The key point here is that if the labour market was suffering predominantly from structural issues such as a skills mismatch, the break in the Curve should have been evident in the curves for shorter durations of unemployment. Similarly, if it was skills related, some industries should have been suffering while other would have been benefiting. But no such pattern was found across industries or types of work.
Looking ahead, the concern is skill atrophy
Economists have a term called hysteresis, which basically refers to the scarring of the economy’s long run potential from prolonged unemployment. This is the real concern right now. If the lack of demand persists for too long, the labour market issues can turn from cyclical to structural. This can be seen in the following diagram, which I have adapted from Ghayad & Dickens.
If demand side policies are successful, such as the Fed’s promise to keep interest rates low until a target level of unemployment is hit, then the Beveridge curve may return to it’s long run pattern. If they are not, or are inadequate, then the break may become permanent as the economy is structurally damaged and a new Beveridge Curve results.
There’s still the issue of the participation rate
Finally, there is also an ongoing concern about the participation rate, seen below as a ratio of those employed against the population. A clear shift has occurred since the Great Recession and the worry is that if this trend is not reversed, this could also undermine the long run productive capacity of the US economy.
 It’s not a skill mismatch: Disaggregate evidence on the US unemployment-vacancy relationship, 5 January 2013.