A New Era of Endless Labor Shortages? A Critical Analysis of McKinsey’s New Report

By Thomas Ferguson and Servaas Storm / Institute for New Economic Thinking

Every so often a publication comes along that more or less perfectly captures the Zeitgeist of world business elites. So it was on the 26^th of June when the McKinsey Global Institute issued a new report: “Help Wanted: Charting the Challenge of Tight Labor Markets in Advanced Economies.”

The message its few pages of charts and text deliver is dire indeed: “Labor markets in advanced economies today are among the tightest in two decades, not merely a pandemic-induced blip but rather a long-term trend that may continue as workforces age.”

This constraining labor market, the report claims, “means forgone economic output,” since employers are not able to “fill their excess job vacancies.” “Companies and economies,” the authors warn, “will need to boost productivity and find new ways to expand the workforce. Otherwise, they will struggle to exceed—or even match—the relatively muted economic growth of the past decade.”

Less than two weeks later, on July 5, the United States Department of Labor released its “Employment Situation Summary” for the month of June. Lost in a wave of mostly upbeat press coverage were two striking facts utterly incongruent with the McKinsey report’s claims. Firstly, unemployment had risen to its highest point since November 2021: 4.1%. Even more awkwardly, Long Term Unemployment among the unemployed had spiked to 22.2% — more than five percentage points above its February 2023 low of 17.5%.

The measure of unemployment highlighted in the June release is the “narrow” measure, sometimes referred to as “U3.” It excludes many workers who we and many other analysts have long insisted should also be counted, such as discouraged workers and those working part-time because they could not find a full-time position. At no point during the pandemic did that “wider” measure, “U6,” in Labor Department parlance, ever fall below 6.7%. Since July of 2023, it has fitfully risen to 7.4. This means that, in June 2024, there are more than 13.1 million American workers willing to work but without a job.

These data are for the United States, but the situation in other developed countries is comparable; indeed, in many countries, the numbers are even more adverse to the McKinsey view. The authors of the report acknowledge that labor markets in some big countries, such as France and Italy, are not very tight, but they minimize the discrepancies, claiming that those are in fact tightening, too, just more slowly.

Rates of unemployment across developed countries differ and bounce around, as one might expect. But the June OECD figures (mostly reporting figures for April that OECD tries to make as comparable as possible), indicate that unemployment rates in most of developed countries run above current US rates and are trending up.

So how does McKinsey justify its conclusion that labor markets are tight and likely endemically so? The answer is simple: It ignores actual unemployment and instead builds its case in terms of data for “job vacancies.”

We have observed before that resorting to data about job vacancies became popular among mainstream economists as the much-touted Phillips Curve relation between unemployment and inflation broke down. And that a close look at US data raises deep doubts about these data’s reliability over time.

The emphasis McKinsey places on these numbers makes it worthwhile to reopen the case, so consider Figure 1. This plots the “job openings rate” and the “quits rate” since the beginning of this century. The first of these is computed by dividing the number of job openings by the sum of employment and job openings; the second refers to the percentage of workers who voluntarily leave their jobs each month. We are interested in the job openings rate because that is closely related to the job vacancy ratio, which is the ratio of job openings to unemployed workers.

Before 2010, the relationship between vacancies and quits was relatively stable. Between 2010 and 2019, however, the job openings rate began to surge relative to the quits rate. During the pandemic the relationship between the two indicators diverged markedly: as job openings rocketed upward, the quits rate remained rather steady.

Similarly, the job openings rate and the hiring rate have grown apart, as is shown in Figure 2. Both moved closely together from 2001 to early 2020, but then the job openings rate almost doubled, whereas the hiring rate held fairly steady.

These disparities raise further doubts about the reality of the sudden rise in the number of posted job openings. We could as well have pointed to the job-finding rate of the unemployed or any other labor market measure to show the very exceptional behavior of the job openings rate during 2021-2024.

Figure 3 shows just how unusual the situation was. It presents a “predicted job openings rate,” which we computed assuming that the relatively stable relationships between the job openings rate and the quits rate (Figure 1), and the hiring rate (Figure 2) that existed during January 2001-December 2019 persisted during 2020-2024.[1]

It can be seen that the “predicted job openings rate” closely tracks the actual job openings rate from January 2001 to December 2019, before diverging dramatically from the actual openings rate from January 2021 to May 2024. The “predicted job openings rate” peaks at 5.7% in…