The July employment report showing strong jobs growth and an unemployment rate below 5% would, on the face of it, suggest a US labour market in robust health. But many commentators remain unconvinced, pointing to the continued low labour force participation rate (LFPR) – the combined percentage of the adult population either employed or looking for work (unemployed).
They argue the low unemployment rate offers a flattering picture of the labour market because many would-be workers are discouraged to the point of giving up their job search altogether, thereby dropping out of the labour force and artificially lowering the unemployment rate.
So which narrative are we to believe – is the labour market tight or not?
The participation rate does have a cyclical component – discouraged workers drop out of the labour market to return when job openings and wages pick up.
But a look at the longer history suggests structural drivers are significant.
In particular, there are three long-term trends which have resulted in a structural decline in the LFPR this century:
- the ageing of the baby boomers;
- the end of a rising trend of female participation; and
- a 60-year trend decline in the participation rate of prime working-age males which has shown no signs of abating.
The demographic impact is very significant and can be highlighted by comparing participation rates by age cohort over time. From 2004 to 2014, the LFPR declined from 66% to 62.9%. The majority of this decline is accounted for by the fact that the share of the population aged over 55 – understandably a cohort with a low participation rate – has increased substantially.
If we hold the distribution of broad age categories constant between 2004 and 2014, the LFPR would instead be 65.0% rather than the 62.9% reported for 2014. This is still lower than in 2004, which suggests some ongoing cyclical slack especially as 2004 was not the peak of the previous cycle, but controlling for population ageing provides a much less dramatic decline in the LFPR.
Note: the percentages below differ from those in figure 3 above because they are a percentage of the civilian noninstitutional population rather than for the population as a whole.
The table also highlights some interesting trends over the past decade for different age groups. Bucking the overall trend, the participation rate has actually risen for people over 55 years. For young adults (<25 years), the participation rate has declined most sharply. Much of this is likely to be cyclical, as evidenced by a similar decline after the 2001 economic slowdown. But, in part, it may also reflect a very rational decision in favour of increased education among Millennials, given the high returns to investing in college education in the US, which appear to have increased.
Trends by gender
The other structural trends we observe are particular to gender: a long-term decline in the LFPR among men of prime working age (25-54) and an increase in female participation which peaked out at the end of the last century. The Council of Economic Advisers report offers an interesting analysis of labour force participation among prime-age males, which peaked in 1954 and has declined consistently to leave the US as a significant underperformer in this regard by international standards.
The report convincingly argues that low participation for this group mostly reflects inadequate demand for such labour, especially among low-skilled workers which is consistent with the relative decline in real wages this group has experienced.
Where does this leave us?
Economists should always be humble with their conclusions, even more so than usual when it comes to estimating whether unemployment is above or below its natural rate (also known as the NAIRU1), a concept notoriously hard to pinpoint. (That we will know with the benefit of hindsight is little comfort).
Simply looking at today’s rate of unemployment (4.9%) relative to recent decades would suggest the US is currently approaching full employment and at a level where one might expect wage growth to accelerate. Indeed, this now seems to be what is happening, albeit gradually.
Considering also today’s low participation rate, it would seem reasonable to conclude there remains some modest further slack in the labour market. But much less than the headline drop in participation might suggest. So long as the observed long-term structural trends in participation persist, we should expect the prevailing levels of US economic growth to lead to a further gradual decline in the unemployment rate and a pick-up in wage inflation. The participation rate meanwhile is likely to remain low judged by the 20-year average.
The above assumes structural trends remain intact. More optimistically, there is plenty of scope for improvement if somehow the right policies can reverse the trend decline in male prime-age participation. The experience of other developed countries may provide examples, where the reverse trend exists. With so many of America’s economic and social problems seemingly connected, anything that can encourage labour force participation, especially among the poor, less educated and disadvantaged would be a positive development.
1Non-accelerating inflation rate of unemployment