With Beveridge down, the job market is moving down a different path

The job market remains in very good shape and continues to surprise on the positive side. This is also shown by the latest data. On this occasion, I would like to raise a problem that economists around the world are currently facing and that affects the interpretation of what is happening and may happen in the labor market. The question is whether there is a shift or even a break in the famous Beveridge curve, which indicates that the cooling of the labor market must be effected by increasing unemployment. It is possible that structural changes in the labor market mean that the market can only cool through a decrease in the percentage of job vacancies, rather than an increase in unemployment, which may mean that deflation may be less costly than in the past.

Data first. The Central Bureau of Statistics (GUS) reported on Thursday that the registered unemployment rate was 5.2% in April. Compared to 5.4 in March (a year ago in April it was 5.6 percent). Thus, the decline was deeper than indicated by the preliminary data of the Ministry of Family and Social Policy. Of course, there is a seasonal effect, but after removing it, we get a very stable or even slightly declining unemployment picture. Let us recall that according to the macroeconomic forecasts of the Polish National Bank, unemployment was already expected to grow *. Polish GDP is falling year on year, so in theory fewer people are needed to work and we should see more layoffs. However, the practice looks different. What’s going on?

Economists in the United States are puzzled by a similar conundrum. There is also the central bank projecting an increase in the unemployment rate in response to the economic slowdown, and nothing of the sort has happened. I have written about this phenomenon many times and predicted that eventually we would see this increase in unemployment. Maybe it was wrong.

The belief that cooling the economy and the labor market should cause the unemployment rate to increase is based on a theory called the Beveridge curve (named after the English economist). Show it on the chart. She says that as economic activity decreases, companies reduce the number of new hires, which lowers the vacancy rate (the ratio of job vacancies to the number of jobs), and at the same time increase the number of people laid off, which increases unemployment. Therefore, in the economic cycle, the vacancy rate and the unemployment rate follow a downward curve—the vacancy rate falls and unemployment rises during recessions, and vice versa during accelerations.

Of course, this line around which the labor market moves is not static. The matching efficiency between people looking for work and firms looking for employees means that the curve can move higher (higher vacancy rate for a given unemployment rate, poor market match, firms can’t find people) or lower (lower vacancy rate for a given unemployment rate , the matching is better, companies find workers quickly). These up or down shifts occur for structural rather than cyclical reasons – for example, due to technological changes that may reduce the suitability of workers for the jobs offered.

Many economists, myself included, were convinced that we would follow a flat curve in this cycle – the economy would slow, companies would reduce the number of vacancies while at the same time laying off some people. Lawrence Summers and Olivier Blanchard – two very well-known economists working in the United States – wrote in mid-2022: “Fighting inflation will require a reduction in the number of job vacancies, as well as an increase in unemployment. The Central Bank’s hope to reduce vacancies without increasing unemployment contradicts historical empirical evidence.

However, not everyone was convinced of this. Fed economists Andrew Figura and Chris Waller argued in the fall of 2022 that the job market is in a unique position — the pandemic has made it so bad to match employees to job offers, increasing the vacancy rate without causing unemployment to fall. In fact, companies had a lot of trouble finding employees and often posted several advertisements for a single job. The demand for labor was high, and the supply was weak, which led to higher wages and sustained inflation. According to these economists, a cooling of the labor market may lead to a decrease in the vacancy rate, but not necessarily an increase in unemployment. Companies will reduce the number of job offers, wage growth will slow, but the unemployment rate will remain low. The efficiency of matching people to jobs will improve. In other words, cyclical and structural change overlaps, with fluctuations in demand there is an increase or decrease in the matching of workers to jobs.

For now, the optimists are more right. In both the United States and Poland, we see a decrease in the number of jobs offered, but not an increase in unemployment.

There is a particularly interesting phenomenon in Poland. For several years now, the labor market has been moving along a different trajectory than the typical Beveridge curve – the vacancy rate and the unemployment rate have been positively correlated since 2018. This would mean that all changes are structural in nature, and there is an increase and decrease in the quality of job matching. Market cooling translates into restricting search by firms and making it easier to match offers and employees, but not to an increase in unemployment.

This should be an argument for stronger monetary tightening, since the social cost of inflation is lower. NBP President Adam Glabinski said in April that inflation can only be brought down faster with a big increase in unemployment. Well, it hasn’t budged yet, it’s below NBP forecasts, and the labor market is just adjusting through changes in job vacancies.

* NBP analyzes the unemployment rate as measured by surveys and not the records of employment bureaus. These measurements show a different result. However, their changes are almost always identical.