Businesspeople in the U.S. have been complaining for more than a year about how hard it is to hire anybody. Their public-sector counterparts have had it worse, though, and it might not just be a pandemic thing.
To put it in numbers, nonfarm private-sector employment in the U.S. was 885,000 higher in August than before the pandemic in February 2020, according to the U.S. Bureau of Labor Statistics. (That’s seasonally adjusted; without the adjustment it was 3.3 million higher.) Government employment was still down by 645,000.
This actually understates the private-public divergence. Last month, the BLS reported that its annual benchmarking of payroll jobs numbers against unemployment insurance tax records showed private-sector employment to be an estimated 571,000 higher than reported as of last March and government employment 109,000 lower. These numbers won’t be finalized and incorporated into the monthly jobs data until January, though, so just be aware that the government jobs declines are even bigger than shown in the charts here.As currently reported, the job losses have all been at state and local governments, with federal employment rising ever so slightly. (The short-lived government employment increase in the second half of 2020 apparent in the chart above was all about temporary federal census jobs.)
Most of these job losses came early in the pandemic, when state and local income and sales tax revenue plummeted and elected officials feared it would keep falling. Instead, both bounced back quickly while property taxes, the other main source of state and local revenue, never took much of a hit. The estimated $513 billion in seasonally adjusted state and local revenue in the first quarter of this year is 17% higher, adjusted for inflation, than in the fourth quarter of 2019.
So it’s not as if state and local governments need to reduce headcount. But even before the pandemic, government agencies were struggling to keep workers and hire new ones in the face of strong competition from private employers, and as of July there were 1.1 million job openings at the local, state and federal level — almost double the number just five years ago.
Shortages of public school teachers have been receiving the most attention, but local governments have also been struggling to hire enough police, firefighters, paramedics, garbage collectors and others. Worker “shortages” are, of course, generally an indicator that wages aren’t high enough, but governments have less flexibility than private employers do in adjusting pay and benefit packages to changing labor market conditions. A recent survey of state and local government human resources staff members by the MissionSquare Research Institute found that only 44% thought the wages they offered were competitive with the overall labor market. In 2016, the first year the survey was conducted, 61% did. That’s for jobs that during the pandemic have often been especially dangerous and stressful, diminishing their attractiveness relative to much private-sector work.
Health and retirement plans are another matter, with 85% of those polled saying the benefits their states or localities offered were competitive. But retirement plans that don’t kick in for decades and often aren’t portable from job to job may be less appealing than they once were given that layoffs early in the pandemic and during and after the Great Recession made clear that state and local government jobs aren’t necessarily forever.
That Great Recession retrenchment marked something of a turning point for state and local government employment, which peaked as a share of nonfarm payroll employment in the mid-1970s but held more or less steady from then until around 2010. Here’s employment at all three levels of government as a percentage of nonfarm payrolls, which hit its all-time high of 19.4% in July 1975 and was down to 14.6% in August — about where it was in the summer of 1957. The interruptions in that decline have come mainly from recessions, when private employment tends to fall faster than government employment.
The federal government is the smallest of the three in terms of employment and has followed a different trajectory from the other two. Its relative size peaked during World War II, fell more than 50% from 1957 to 2000, and has held more or less steady since (the little spikes every 10 years are for censuses).
In numbers rather than percentages, federal employment excluding the U.S. Postal Service was 2.3 million in August and has been in that general vicinity since the late 1960s. The postal service, which I’ve excluded here because the non-postal workforce seems like a better reflection of what we mean when we talk about the federal government, employed 600,000 in August, down from a peak of just more than 900,000 in 1999. Active-duty military members aren’t included in the BLS jobs data, and if they were the decline shown above would be much more pronounced — there are about 1.3 million now, down from 2.8 million in 1957 and 3.5 million in 1968. The intelligence agencies aren’t included either, but with employment estimated at around 100,000, they wouldn’t make a big impact.
Also missing are the many non-federal employees whose paychecks are funded by federal dollars, but Paul C. Light, a professor at New York University’s Wagner Graduate School of Public Service, produces estimates of their numbers. He’s quite confident of the contractor totals, which are calculated from detailed government procurement inventories, but says there’s “a little more squish” in the grant-employment estimates.
Increases in defense spending mean greater numbers of government contractors, which mostly explains the ups and downs in the chart, as defense spending rose a lot during the administrations of Ronald Reagan, George W. Bush and Donald Trump, fell during those of Bill Clinton and Barack Obama, and was close to flat during George H.W. Bush’s presidency. Depressed private-sector job totals during or in the wake of recessions reduced the denominator and thus boosted the percentages in 2010 and again in 2020 and 2021, but even with that, the total federal employment in the 2021 fiscal year was smaller relative to overall employment than in the 1980s and early 1990s, and probably smaller than it was in the decades before then.
Still, by this metric federal employment is nearly as big as local government employment, some of which is funded by federal grants, and is significantly higher than it was 20 years ago. But is that too high? Too low?
It is interesting that the long decline in government employment as a share of total employment has coincided with an era of general dissatisfaction with government, although I’m not sure which was the cause and which the effect (probably a little of both). Federal spending hit a peacetime record of 31.3% of gross domestic product in the 2020 fiscal year, so it’s not as if government has shrunk in that sense. Washington’s regulatory reach hasn’t exactly receded either — the Federal Register, where new regulations are published, had twice as many pages in 2021 as in 1978. But maybe that’s the problem: We’re asking government to do more without hiring enough people to do it right.
