A Slowdown in Hiring is Dangerous Even Without Layoffs

Three weeks after “Liberation Day,” the economic outlook is bleak. Stocks and the dollar are down since April 2nd while the bond market has exhibited volatility, and the soft data—especially in the manufacturing sector and for expected business conditions—have collapsed. The prospects for another era of high growth and productivity have dwindled as analysts, ourselves included, have started to put out recession calls.

The remaining shoe to drop is in the hard data, where we’ve yet to see signs of recession. In large part this is because the hard data comes out with a lag; we’ll get the first jobs day that covers post-Liberation Day data next Friday. It’s also the case that the hard data could hang on for a few months before the start of a recession, given the strong growth in 2024 and strength in household and firm balance sheets. For our part, we see a 75% chance of a recession materializing within the next four quarters with the modal onset most likely during the 3rd quarter of this year.

As we await signs of deterioration in the hard data, people will likely pay attention to announcements about layoffs. We’re already starting to see anecdotes about manufacturing firms laying off workers. This op-ed frames the recession script around layoffs: “If layoffs really start to pick up, consumer spending will nosedive and a downturn is almost certain.”

It’s understandable why layoffs come into focus during a recession. Layoffs are the most visible manifestation of the labor market falling apart, and the damage is obvious: someone had a job, and now they don’t. Layoffs are often announced in advance, and there are a number of data sources out there that track layoff announcements. Layoffs often happen in waves, rather than one-by-one.

Layoffs are important, but focusing on layoffs risks missing the other part of the equation: hiring. In fact, hiring activity is oftentimes even more important than layoffs when understanding the trajectory of employment and unemployment over the business cycle.

We can get a sense of the relative importance of job-finding and job-losing over the business cycle using the Current Population Survey data. I’ll skip the finer details, but I follow the methods from Shimer (2012), who uses the number of recently unemployed and changes in the number of unemployed to estimate rates of job-loss among the employed and rates of job-finding among the unemployed. In steady-state, the unemployment rate is the ratio of the job-loss rate to the sum of the job-finding and job-loss rate. I then calculate the contribution to changes in the unemployment rate from changes in both job-finding and job-loss rates.

Below, I plot the contributions of changes in job-finding and job-loss rates to the change in the unemployment rate during the 2001 recession using this method. The rise in the unemployment rate, which accelerated in mid-2001, was primarily driven by a collapse in the job-finding rate. In late-2001, the rate of job loss increased and contributed to the rise in the unemployment rate (you can also see this in the JOLTS data, which the BLS began to collect in 2001). However, the unemployment coming from job loss was relatively small and reverts by early 2002.

For another illustration, I plot two counterfactuals of the unemployment rate during the 2001 recession below. In the first counterfactual, I keep the job-separation rate constant at January 2000 levels, and compute the implied unemployment rate from the movements in the job-finding rate. In the second, I flip the exercise: I keep the job-finding rate constant and only allow the job-separation rate to vary.

The movement in the job-separation rate explains very little of the rise in unemployment during this period. Meanwhile, the fall in the job-finding rate almost entirely explains the rise in unemployment.

Of course, not all recessions are the same. Layoffs played a very strong role in the rise in unemployment in 2020, due to the sudden and unexpected nature of the onset of COVID and the rapid closure of many businesses.

If we are to see a recession in 2025, it seems likely that it will look more like 2001 than 2020, and that declines in hiring will be the predominant reason for a rise in unemployment. We are already hearing reports that businesses are hesitant to engage in hiring in this environment. The uncertainty around trade policy may also discourage layoffs, at least initially. Layoffs are difficult to reverse and hurt the morale of remaining employees, and businesses may not want to take that step if they believe there is a chance of policy reversing course. Firms that remember how difficult it was to find workers in 2022 may not want to lay off workers and risk repeating that experience.

Still, the lessons of past recessions are clear: a slowdown in hiring alone can drive a rise in the unemployment rate even without layoffs.