It was Friedrich Engels (1945) and Karl Marx (1847) who first described the unemployed as a ‘reserve army’ whose purpose was to keep wages down. Ever since, economists have used the unemployment rate – defined as the number of unemployed divided by the labour force which adds the unemployed to the employed – as the key measure of labour market slack. They have plugged unemployment rates into wage equations expecting to see a negative partial correlation. And for over a century, the unemployment rate was negatively correlated with wage growth.
The fact that unemployment is at historically low levels has therefore led some policymakers to conclude that labour markets are currently very tight and there is potential for wage-push inflation to add to other inflationary pressures in the economy. Central banks have concluded that they need to respond by hiking interest rates to suppress demand in the economy. A major macro puzzle, though, was why wage growth remained benign in the period 2010-2020 as the unemployment rate fell from 6% to below 4%. The answer was there was more labour market slack than shown by the unemployment rate.
However, in a new paper (Blanchflower et al. 2022), we show unemployment no longer performs the role it once did. Since the Great Recession of 2008, unemployment rates have been uncorrelated with wage growth in the US. Instead, two other labour market measures are key: (1) the under-employment rate, which is the percentage of workers working part-time who would like longer hours; and (2) the non-employment rate, which is the percentage who are jobless in the adult population.
The under-employment rate has been negatively correlated with wage growth in the US since the 1980s, with the size of the effect rising since the Great Recession, whereas the non-employment rate was not statistically significant in wage growth equations prior to the Great Recession but has become strongly so since then – effectively swapping places with the unemployment rate. This is apparent in panel wage equations for the US which include state and year fixed effects and lagged wages. Figure 1 gives a sense of the way wage growth closely tracks non-employment and under-employment rates in recent years among private sector production and non-supervisory workers in the US, who constitute around three-quarters of private sector workers in the country.
Figure 1 Underemployment, non-employment and weekly wage growth of PNSW
Economists have also traditionally expected wages to rise when the vacancies-to-unemployed ratio (the Beveridge Curve) is high. But we find no role for the vacancies-to-unemployed ratio, or V:U ratio, in wage equations for the US.
These results are a little surprising to economists, many of whom were brought up on the seminal work of Layard, Nickell and Jackman (1990), who assumed that non-employment rates should not enter wage equations because those who are not actively seeking work are unlikely to compete for waged employment, and thus will not lower wages. However, Marx and Engels made no such distinction between the unemployed and the non-employed more generally in their discussions of the ‘reserve army of labour’.
The implication for policymakers is that high non-employment and high underemployment are indeed additional measures of labour market slack, pushing down on US wages. A substantial portion of those American workers who became inactive should not be treated as gone forever but should be expected to spring back into the labour market if demand rises to create jobs.
What’s more, in contrast to most other Western industrialised countries the non-employment rate has been rising in the US. The only other exceptions are Greece and Spain, both of which have double-digit unemployment rates. The US non-employment rate is up 3.6 percentage points since 2005 to 31% of the adult population, increasing the reserve pool of labour. The implication is that the labour market is not as tight as analysts have thought, helping to explain the absence of strong wage pressures and the continued decline in real wage growth.
The bigger the non-employment and underemployment rates, the larger the pool of available labour. The question is whether the non-employed are ready to spring back into the labour market when jobs present themselves. It seems they are, as they are impacting wage growth and there is no sign the vacancy rate is. The obvious conclusion is they are discouraged workers when offer wages exceeded their reservation wages.
What changed in 2008 to bring this about? We can only speculate but it seems plausible and indeed likely that the Great Recession scared workers and non-workers, reducing their bargaining power, hence lowering the non-accelerating inflation rate of unemployment (the NAIRU).
In addition, the underemployment rate reflects low wage pressure from within organisations, helping to suppress insiders’ bargaining power. The fact that the underemployed have been unable to find sufficient hours, has kept the wages of other workers down, presumably within their own firms.
Now that we have shown that, in the years since the Great Recession, the reserve army of labour is far bigger than indicated by the unemployment rate, the question is: how will policymakers and central bankers respond?
Blanchflower, D G, A Bryson and J Spurling (2022), “The Wage Curve After the Great Recession”, NBER Working Paper No. 30322
Engels, F (1845), The Condition of the English Working Class in England.
Layard, R, S Nickell and R Jackman (1990) Unemployment: Macroeconomic Performance and the Labour Market, Oxford University Press.
Marx, K. (1847), “Wages”, in Works of Karl Marx, Vol. 6, Max/Engels Collected Work.