Labor Market Recap March 2025: The Calm before the Storm

The labor market added 228,000 net jobs in March, with revisions of -14,000 and -34,000 to January and february. The unemployment ticked upwards by 0.1pp to 4.2% and prime-age employment fell by 0.1pp to 80.4%, down 0.5pp from six months ago but still above its 2019 average. The payroll print came in above consensus and above our estimate, at least in part because government-related “layoffs” may not yet be showing up in the establishment survey data if DOGE-affected workers are still receiving severance pay still being counted as employed according to the BLS establishment survey (to be clear: it has always been defined this way). Average hourly earnings grew by 0.25% in March, bringing the year-on-year measure down to 3.83%.

This is not a very interesting labor market report. It is basically a continuation of the data from the previous month, and is already made stale by the tariff announcements. This is our last snapshot of the labor market before we really start seeing the effects of the new Administration’s policies on the economy. It’s a strong labor market with good wage growth and high employment rates. But, the labor market has been decelerating for a long time and there are genuine vulnerabilities. It’s not horrible, but it might not hold up to a trade war.

This Is (Was?) Still a Good Labor Market

In our opinion, there’s no better gauge of the labor market than prime-age employment rates, and I think it really still encapsulates where we are in this cycle. At 80.4%, the prime-age employment rate isn’t bad by any means. At 80.4%, it’s slightly below the pre-pandemic peak (80.6%) and above the 2019 average (80.0%). But, the local trend is the wrong direction: we’re down a full half-percent from the post-pandemic peaks of the middle of last year.

The decline in full-time employment continues its steady decline. There was a sharp decrease in the full-time employment rate last month, which we hoped was attributable to one-off weather effects in February. However, this was only partially reversed in March.

The percent of the population that are part-time for economic reasons (meaning they would prefer to work full-time, but can’t find work due to lack of demand) continues to rise. It’s still historically low, but over the last year it’s started to surpass 2019 levels.

One theme we have been highlighting is the sectoral composition of job growth, which has been heavily concentrated in “acyclical” sectors (such as education and healthcare and government) and not "cyclical" and interest-rate sensitive sectors such as manufacturing and professional and business services.

That concern still remains, but the last few months have seen a slight broadening in job growth as some of the acyclical sectors continue to slow and some cyclical sectors have started to pick up. Construction employment continues to show steady growth for now, and manufacturing employment is now moving sideways instead of contracting.

Meanwhile, acyclical sectors have exhibited a gradual decline in payroll growth as catch-up growth from the pandemic continues to wane.

We (and others) were concerned about the prospect of government and government-related layoffs showing up in this payroll print, but federal government employment fell by only 4,000 jobs, and is only down 15,000 since January. A note from the BLS suggests that many laid-off government workers are still showing up as employed in the survey because they are receiving severance pay. We expect federal layoffs to start to show up over the next few months, given the volume of announcements of federal layoffs.

Wage Growth Slows

Since the beginning of 2024, wage growth has been bouncing around 4%. Again, historically speaking, this is fairly good. But the local trend is towards slower growth. Average hourly earnings have now grown 3.83% over the past year.

Playing a large part in that decline was a decrease of 0.2% in the average hourly earnings of production workers in private education and health services (you can see that reflected in our core non-housing services wage index). It's unclear how much of this is an actual change in March: it turns out that for certain subsectors in private education, the BLS maintains constant average hourly earnings and average weekly hours constant throughout the year and updates them once a year with each benchmark.

Based on past data, it looks like this happens every March. So, this might just be the annual update playing havoc with the monthly wage growth numbers. By the way, for those of you paying to PCE Prices: education wages feed directly in to the price index for education non-profits, and that price index has historically shifted around drastically in March as well.

Will the low hiring, low firing regime continue?

For months, the labor market has been in a "low hiring, low firing" regime. Layoffs have been basically flat for years, while hiring has retreated to early-2010s levels. Quit rates have settled in around mid-2010s levels.

This month's data was no different—in fact, after revisions, there was no change to hiring, layoff, and quit rates. Even before the chaos of federal layoffs and tariffs, there was a question about whether or not the labor market can continue in this state. The stability of the labor market with such low hiring rates is going to become even more questionable under heightened policy uncertainty and tariffs—it's easy to see businesses switch from a labor-hoarding mindset to a "we need to get lean" mindset.

Where does the Fed go from here?

The Fed is in a very difficult position now. The announced tariffs are far higher than expected. Theyre staring down a cost shock, slower growth, heightened recessionary risk, and a stock market selloff all at the same time. And theyre coming into this with inflation uncomfortably above target already.

Balancing the dual mandate in the face of a supply shock is an age-old question, and one they faced just a few years ago. The Fed still has not figured out a way to deal with supply shocks, and its paramount that their framework review addresses this problem. For a start, they should check out our very own Skanda Amarnath's piece today on why nominal aggregates such as gross labor income can help them find their way through the fog.