Alex L'aventurier

We do not think often of economists as detectives, as we tend to leave exciting puzzles to those with the more exciting professions. The truth is that economists love mysteries, and that these mysteries pose many unanswered questions. In this realm of mysterious phenomena there is one puzzle that continues to confuse economists in recent years: the wage growth mystery. This is a popular name for something that has been happening in the US and in other (mostly European) economies since the Great Recession of 2007-2009: although wage growth is supposedly negatively associated with the unemployment rate, recent data fails to correspond with such prediction. In other words, while the unemployment rate has ‘recovered’ to a pre-recession low, the growth in wages remains stagnated, much to the discomfort of politicians and policymakers.

For example, a labor market report made by the US Bureau of Labor Statistics showed that the American economy created 156,000 net new jobs in August 2017 and 2 million new jobs over the year beforehand. This has pushed the unemployment rate in the US below 4.5%, close to what is known as the natural level of unemployment (NAIRU), or full employment. Wage growth, given this unemployment rate, would be expected around 3.5%, but instead was around 2.5%. A recent report indicated that this mystery continues to baffle economists in the US, months after this report was published. Although these numbers do not seem like a big deal, in fact, the divergence between unemployment and wage growth contradicts one of the most fundamental macroeconomic models known to economists. The Phillips curve graphs a negative relationship between unemployment and inflation, wages included; if unemployment is back to its ‘normal’ level, why aren’t wages growing faster? A variety of explanations was given by economists in recent years in order to solve this problem: a poor inflation environment, the ageing of the baby-boomer employees, irregularities in the business cycle – all of which do not offer a definitive solution. However, a possible solution for this mystery may lie in the variables of the mystery itself, that is the changing definitions of unemployment.

The definition of unemployment, according to the Bureau of Labor Statistics, is pretty simple. An unemployed person is someone who does not have a job and has been actively looking for one in the past four weeks. A simple definition, but also quite a restrictive one. It turns out that by defining unemployment so narrowly, various groups in the labor market are excluded and left undefined. Most notably, people who are long-term unemployed and people who are involuntary part-time employed. The former describes those who have given up looking for work and are, consequently, not officially considered unemployed; the latter describes those who are constrained in their number of working hours and would prefer to work full-time. Even though we would probably consider these individuals unemployed, they are simply excluded from the statistics of unemployment. What is even more interesting is the fact that both types of phenomena, long-term unemployment and involuntary part-time employment, continue to claim a large share of general unemployment more than ten years after the Great Recession. Economists have recently shown that the share of the long-term unemployed and involuntary part-time workers increased during the recession and remained high during its aftermath, even though general unemployment decreased.     

What does this mean? It means that we inadequately measure unemployment to construct the model relating unemployment to wage growth. It means that while general unemployment seems to have gone down in the US since the recession, various groups in the labor market have remained unemployed over time. People who are long-term unemployed, for example, find it harder and harder to return to the labor market as time goes by – they may become discouraged, or stigmatized by employers, or lose important skills. The Bureau of Labor Statistics estimates the percentage of total unemployed, including all marginally attached workers and involuntary part-timers as a percent of all labor force, to be approximately 8% as of November 2017. Some economists, like in this paper, concluded that with an updated statistic of unemployment, there is in fact a negative association between unemployment and wage growth. Or, in other words, a reconstructed measure of unemployment can in fact explain the weak growth in wages. Although statistics may vary, it is evident that both long-term unemployment and involuntary part-time unemployment maintain a strong influence on labor market dynamics and suggest a hidden slack unaccounted for by the unemployment rate.

A possible conclusion is that alternative measures of unemployment should be considered when attempting to solve the wage growth mystery, even though the traditional measure of unemployment serves the politicians who claim that the economy has recovered since the recession well. Macroeconomists will benefit from a closer cooperation with labor economists if they wish to reflect the current situation in the labor market more accurately. A change in definition implies a further change in welfare schemes and in monetary policies; this is because the recovery from long-term unemployment and involuntary part-time employment is considered slower than from commonly defined unemployment. We must also consider the societal implications of excluding many de-facto unemployed workers from the official definition of unemployment. As underemployment increases, discouraged workers become even further discouraged and detached from the labor force, while part-timers find themselves in a poverty trap enabled by the lack of definition for their situation. Whatever discussion we are having, it is important that we continue to have it: the Great Recession influenced almost every economy in the world with devastating consequences for many. It is crucial that we continue to study the causes of such recessions and explore future adequate strategies for recovery.