If You Lose Your Job, The Next Gig May Take You Longer To Land

When examining the economy and society, patterns of data are crucial. They help forecasters to estimate what is likely to happen in the future. The more accurate the forecast, the better governments, companies and individuals can anticipate how they should respond. When expected patterns are off, there may be trouble coming, and that is what’s happening today.

Past Performance And All That

The typical warning that past performance offers no guarantee of future results is apt. And yet, broad, repeated rhythms and movements are usually important. If there were no hope of the past offering insight, even the greatest experts would likely throw their hands up into the air. There would little way to know how to map the influence of forces into a picture of upcoming activity.

In economics, there is usually a good amount of previous experience on which to base future predictions. No promise that they will be correct, but there is frequently a directional coherence. When patterns change significantly, it’s time to consider whether underlying dynamics have changed, and how long the new pattern might continue.

Recessions And Unemployment

There appears to have been a major shift in the pattern of how long people remain unemployed.

A graph of the percentage of unemployed people who have been jobless for at least 27 weeks, which is just over half a year, helps explain the change.

Below is a graph from the Federal Reserve Bank of St. Louis using data from the Bureau of Labor Statistics that was updated earlier this month. The next update is scheduled for Friday, Oct. 3. The title of the data set is “Of Total Unemployed, Percent Unemployed 27 Weeks & over.” The vertical gray bars are recessions.

Since the start of 1948 there have been upward and downward shifts in the portion of the unemployed that remain without a job for at least 27 weeks. Up until 2020, the pattern was that long-term unemployment would shoot up in every recession. The percentage would peak sometime after the end of the recession. Then the percentage would begin to drop, often to half the last peak. When the next recession came, the rate would once again climb.

There are two patterns in the data. One is a trend line growing in scale. The graph below uses data from the BLS in an Excel spreadsheet to create a trend line.

The percentage of long-term unemployed has been growing. What is telling, outside of unusual spikes, is the lower end of what is sustained. A recent low was 16.6% in June 2023. Compare that with January 1949, when the low was 3.3%. Over time, even with volatility, it has become more common for people to be unable to find work for longer periods of time.

That is disturbing enough. Still, the second pattern of the relation to recessions has understandable logic. At the start of a recession, the percentage of long-term unemployment goes up. Sometime after the end of the recession, the percentage of long-term unemployment decreases to a new low before the next recession. Since the mid-1970s, the new low has been higher than the previous one.

What Changed

Now comes the unusual signal from the data. The recession in 2020 saw a spike and then a drop. However, the next upward swing began without an intervening recession — a total departure from everything that had happened since at least January 1948.

One possible explanation is that there had been another recession since the one that began and ended in 2020; however, it is unlikely. As a 2022 Newsweek article reported, for the previous six recessions, the average time for the National Bureau of Economic Research to officially call a recession was 234 days. That stretched out to 366 days after the end of the 2008 Global Financial Crisis. That’s a little more than a year. It’s currently been five years and five months since that 2020 recession; the chance of another recession with economic growth and, until recently, a decent labor market seems unlikely.

Another possibility is some underlying crisis has kicked up since mid-2023. Again, that seems unlikely without additional indications.

More convincing is that corporations are finding ways of cutting the number of employees they have, whether through layoffs, fake job listings or using artificial intelligence to eradicate an increasing number of jobs. As the Financial Times reported, there is a “jobpocalypse” in which entry-level jobs for university graduates are disappearing with “the rise in big graduate employers really cutting the number of jobs they’re offering.”