Contrary to what many of us were taught in Introduction to Economics courses in college and graduate school, markets are rarely perfectly competitive. In the case of the market for labor—a market that, it is worth noting, is fundamentally different from those where financial products or commodities are bought and sold—imperfect competition means that workers' pay is solely determined neither by their productivity nor by the forces of supply and demand.
In recent years, rising interest in the causes, consequences, and policy implications of imperfect competition has sparked a new wave of research on a framework known in economics as monopsony. Kate Bahn, the director of labor market policy and chief economist at the Washington Center for Equitable Growth, defined monopsony in a briefing to congressional staffers last month: "Monopsony can be narrowly defined as any time there is one or few employers hiring workers, so they have considerable power to keep wages low since those workers do not have a lot of other options for jobs."
But monopsony power, Bahn noted, is present not just when there are only a few employers in a particular labor market. Barriers that limit workers' ability to move from job to job, incomplete or asymmetrical information about jobs, and discrimination are all factors that can give employers an upper hand when setting wages and, with it, the power to underpay workers.
Hosted by Equitable Growth on March 23, the briefing, part of a series dubbed "Econ 101," introduced Hill staffers to labor markets under monopsony. During the presentation, Bahn first discussed how the assumptions that underpin the view that markets are perfectly competitive are rarely met. She then described the basics of the monopsony model and covered some of the empirical evidence on how monopsony affects the U.S. workforce and economy. And to wrap up, Bahn discussed the public policies that can counteract employers' wage-setting power, boost competition, and deliver broadly shared economic growth.
The implications of monopsony for U.S. labor markets
So, what are the implications of monopsony for our understanding of how labor markets work? And what does imperfect competition mean for the design of effective and equitable economic policy?
Consider the basic supply-and-demand model of the labor market. According to this framework, when governments enact a wage floor, they artificially set wages at a level that reduces firms' desire to hire workers. As demand moves away from equilibrium and up along the labor demand curve, people who would be willing to sell their labor at a "competitive" wage (the theoretical market rate) are unable to find work. In this hypothetical labor market, then, firms hire fewer workers than they want to, workers who would like to sell their labor cannot, and employment is lower than it would be under equilibrium. In economic terms, wage floors lead to socially inefficient outcomes. (See Figure 1.)
Further, under this model of perfect competition, employers are "price-takers," meaning that they must accept the market wage and have no way of influencing it. The assumptions that underpin the perfectly competitive model of the labor market—for instance, that all market participants have perfect information, that all workers are able to seamlessly transition from job to job, and that wages are an accurate reflection of all workers' productivity—also imply that workers are very sensitive to wages. In other words, an employer only needs to pay one cent more than its competitors to have an unlimited supply of workers.
At last month's event, Bahn noted, however, that "if workers face frictions when trying to find a job—if workers do not know what jobs are available or face constraints in what kind of position they can take—then employers do not need to only pay one cent more than their competitors to have an unlimited flow of workers."
Indeed, researchers document that various obstacles when trying to land a position, the concentration of a few employers in any given labor market, and employee preferences about job characteristics besides wages make workers much less sensitive to pay than the perfectly competitive model proposes. "In the actual labor market, employers do not have to offer pay raises or cost of living adjustments tied to inflation if they do not have to compete for workers," Bahn told congressional staff.
In addition, empirical evidence shows that these noncompetitive forces affect U.S. labor market dynamics, with monopsony power giving employers the ability to pay wages below workers' productivity. For instance, a team of economists at the University of California, Los Angeles, the Massachusetts Institute of Technology, and Burning Glass Technologies shows that more than 10 percent of the country's workforce is in labor markets where lack of outside options for their current jobs leads to employer concentration, depressing these workers' wages by at least 2 percent.
In other words, in parts of the country where relatively few employers are competing to hire, lack of competition pushes down average pay. Highly concentrated labor markets, another study finds, are especially pervasive in rural and less densely populated areas. (See Figure 2.)
Other studies highlight how one big employer can influence marketwide wages. For instance, research by economist Ellora Derenoncourt (now at Princeton University) and Clemens Noelke and David Weil at Brandeis University shows that as Amazon.com Inc. set a $15 minimum wage in late 2018, other employers in the same commuting zone had to increase wages. Amazon and other big and powerful employers, the authors find, can therefore act as a wage-setters rather than price-takers, pushing wages up or dragging them down depending on what their particular corporate policy is at a given time.
A study examining how the opening of Walmart Supercenters affects local labor markets captures how one firm's monopsony power leads to worse labor market outcomes. The research finds that Supercenters lead to a decline in countywide employment and earnings as Walmart undercuts wages through low-road employment practices that spill over on other local workers and industries—a dynamic that was mitigated in the counties that experienced an increase in the minimum wage and that would not take place under perfect competition.
So, how does our model of the labor market change when we assume imperfect competition? Under monopsony, the labor demand curve and labor supply curve intersect at both a lower wage and at a lower employment level than would be achieved under equilibrium. This results in socially inefficient outcomes, where the lost wages due to underpayment are kept as firms' profits, with workers making less than the value they contribute. Contrary to the perfectly competitive model of the labor market, in this model, higher wages also lead to greater equilibrium employment. (See Figure 3.)
In contrast to Figure 3 above, in a competitive labor market, if a firm tried to pay only "W monopsony" wages, then workers would go to other firms willing to pay a higher wage.
How policymakers can protect workers from monopsony power in the U.S. labor market
There are two broad areas of public policies that can counteract these forces and create greater balance between the power of employers and the power of workers: pro-competition policies and pro-worker policies.
Together, these policies will boost wages and increase job quality, increase competition and productivity, and drive broad-based economic growth.
To learn more about how monopsony affects the U.S. labor market, see the presentation slides from the March 23 congressional briefing here. For a write-up of a previous "Econ 101" briefing, on the effects of federal tax changes, see here.
A popular line on our recent surge of inflation is that an over-tight labor market has led to rapid wage growth, which in turn is forcing companies to raise prices. Higher prices will lead workers to demand higher wages, which will give us a wage-price spiral and soon lead to double-digit inflation.
While this was a story that plausibly fit the data in the 1970s, it is very hard to make the wage-price spiral fit the current situation for a simple reason: the wage share of income has fallen sharply since the pandemic.
As can be seen, the wage share of corporate income had been recovering gradually from the troughs it hit following the Great Recession in 2014. However, we see a sharp reversal in 2021, with the wage share falling from 76.1 percent to 73.7 percent, a decline of 2.4 percentage points.
Perhaps some economists can tell a story where rapid wage growth is driving inflation even as the wage share of income is falling, but I’m not that good an economist. This still looks to me like a case where supply-side disruptions, associated with the reopening from the pandemic and the war in Ukraine are driving inflation.
This view is consistent with the fact that year-over-year inflation in the European Union was 7.5 percent as of March. The EU countries did not have as big a stimulus as the United States and by most measures its labor market is not as tight.
 Wage share actually refers to all labor compensation, Line 4, NIPA Table 1.14. Corporate income is the sum of labor compensation and net operating surplus (Line 8).
In the general telling of popular history, the Reagan years were a period of a booming economy and general prosperity. Reagan was of course re-elected in a landslide. And, for the only time since Calvin Coolidge, he was succeeded by an elected president of his own party.
In spite of the celebration of the Reagan economy, most workers actually lost ground in the 1980s. Their wages did not keep pace with inflation. And, unlike the current situation, where a war is pushing up the world price of oil and other commodities, in the eighties world oil prices declined sharply from peaks reached following the Iranian revolution.
The inflation adjusted average hourly wage was 1 cent lower in January of 1985, the end of Reagan’s first term, than it had been when he took office in 1981.The rate of decline accelerated in Reagan’s second term so that in January of 1989 it was 1.7 percent lower than when Reagan took office. This means that over his two terms in office, rather than sharing in the gains from growth, workers actually lost ground.
This history provides a notable contrast to the current situation. The media were happy to completely ignore the reality of declining real wages throughout the Reagan era. By contrast, they are happy to jump on a drop in real wages in the last year due to the reopening from the pandemic (the jump in inflation is worldwide) and the war in Ukraine.
The focus on inflation in reporting has been so intense that most people tell pollsters that they think we lost jobs last year, even though it was the strongest year for job growth ever. I’m an economist, not a psychologist, I don’t know how people form their views of the economy. But I can say that the media have not been giving an accurate picture of the economy, nor one that is consistent with their reporting during the Reagan era.
This is for production and non-supervisory workers, a group that comprises roughly 80 percent of the workforce.
Many of us have highlighted both the strong pace of job growth and the drop in unemployment in March. This is great news. We are now looking at a labor market that is as strong as at any point in the last fifty years.
This should be cause for celebration, but all the Republicans have said they don't give a damn about people getting jobs, the issue is inflation. And, most media commentators seem to agree. Hey, what difference does it make if someone can find a job, what about the price of gas?
I'm not sure how much gas people who don't have jobs can buy, but in any case, there was some good news about inflation in this report as well. The scary story about inflation being pushed by inflation hawks like Larry Summers, is that we are facing a wage-price spiral.
In this story, we are not concerned about just a one-time increase in the inflation rate. The real problem is that the inflation rate will continue to increase unless the Fed raises interest rates enough to give us a severe recession. So, the choice is either an inflation rate that spirals ever upward or a stretch of high, maybe even double-digit, unemployment.
On this front, the March data did indeed have good news. First and most importantly, there is some evidence that wage growth is slowing.
The average hourly rose by 5.6 percent over the last year, but it increased at a just a 5.1 percent annual rate comparing last three months (Jan-March) with the prior three months (Oct-Dec). There is a similar story with the pay of production and non-supervisory workers. Their average hourly wage increased 6.7 percent year over year, but rose at just a 6.0 percent annual rate comparing last three months with the prior three months. Pay for production workers in leisure and hospitality, which had been soaring, slowed from a 14.9 percent year over year increase, to an 8.3 percent annual rate comparing last three months, with the prior three months.
The wage data are erratic, and the picture may look very different next month, but the evidence in the March report is that wage growth is slowing, not accelerating. That is not consistent with the wage-price spiral story we keep hearing.
Other data in the report also suggest some weakening of the labor market. The length of the average workweek fell by 0.1 hour in March. This left the index of aggregate hours unchanged, in spite of the 431,000 jobs created in the month. This is consistent with the labor shortage becoming less severe.
The length of the average workweek has been consistently higher than the pre-pandemic level over the last year. This is consistent with a story where employers, facing difficulty getting new workers, have their existing workforce put in more hours. The shortening of the workweek in the March data could indicate that employers are facing fewer difficulties in hiring.
The other item in this report suggesting some weakening of the labor market is the drop in the share of unemployment due to people who voluntarily quit their jobs. This percentage dropped from 15.1 percent in February to 13.0 percent. This is an important measure of labor market strength, since it indicates the extent to which workers are sufficiently confident of their labor market prospects that they are willing to quit a job before they have another job lined up.
As I always note, these monthly data are erratic, and April may tell a different story, but the March data are consistent with some weakening of the labor market. To be clear, a labor market where workers cannot count on pay increases and quit a job they dislike is not good news. But the concern that the labor market was too strong, and would lead to serious problems with inflation, was real. The March report suggests this is less likely to be the case.
I will also add two points that are often overlooked in the inflation discussion. There has been a big shift from wages to profits in the last two years. The profit share of national income has risen by 1.1 percentage points between 2019 and 2021. This is not consistent with the wage-price spiral story told by inflation hawks. If the profit share is increasing, then wages clearly are not driving higher prices.
The other point, is that productivity growth has actually sped up over the last three years, compared to the prior decade. It has average 2.3 percent annually from the fourth quarter of 2018 to the fourth quarter of 2021. It averaged less than 0.8 percent annually from the fourth quarter of 2010 to the fourth quarter of 2018.
This is the opposite of the story we saw with the 1970s inflation. In the 1970s, productivity growth slowed to just over 1.0 percent annually after rising at a 2.5 percent annual rate in the prior quarter century. Workers had long become accustomed to substantial real wage gains year by year. That was no longer possible in the 1970s, with productivity growth slowing to a crawl.
Anyhow, I continue to stress the shift to profits in the pandemic and the uptick in productivity growth as two important differences between the current situation and the 1970s. We'll see how things play out in the quarters and years ahead, however the key point here is that the March jobs report not only had very good news on employment, it also had encouraging news on inflation. The latter point has gone mostly unnoticed.