Labour economicsChapter 4: Labour market equilibrium
Monopsony
Key facts
Key facts
- Course
- Labour economics, ECON 331
- Textbook
- Borjas, chapter 4
- Related
A monopsony is a labour market with a single buyer of labour. More generally, a firm has monopsony power whenever it faces an upward-sloping labour supply curve, so that hiring more workers means paying a higher wage. A monopsonist that pays everyone the same wage hires where the value of the marginal product equals the marginal cost of labour, which leaves it with fewer workers, paid less, than a competitive market would have.
Why hiring costs more than the wage
Section titled “Why hiring costs more than the wage”A competitive firm is one of many employers and can hire as many workers as it likes at the going wage: the supply curve it faces is flat. A monopsonist, such as the mine in a one-employer coal town, faces the market’s upward-sloping supply curve. To attract one more worker it must offer a higher wage, and if it pays everyone the same wage, it must raise the pay of everyone it already employs. The marginal cost of labour, , is the new worker’s wage plus the raise for everyone else, so it exceeds the wage.
| Workers, E | Wage needed to attract E workers ($ per hour) | Total labour cost, wE | Marginal cost of labour, | Value of marginal product, |
|---|---|---|---|---|
| 1 | 10 | 10 | 10 | 30 |
| 2 | 12 | 24 | 14 | 26 |
| 3 | 14 | 42 | 18 | 22 |
| 4 | 16 | 64 | 22 | 18 |
| 5 | 18 | 90 | 26 | 14 |
Hiring the fourth worker means paying that worker 16 dollars and paying the three existing workers 2 dollars more each: . The gap between and the wage grows with each hire, so the marginal cost curve lies above the supply curve and rises more steeply. For a linear supply curve , total labour cost is and
twice as steep as supply.
The nondiscriminating monopsonist
Section titled “The nondiscriminating monopsonist”A nondiscriminating monopsonist pays all its workers the same wage. It hires until the last worker adds as much to revenue as to cost:
That fixes employment at , point A in Figure 1. The wage is the lowest that attracts workers, read off the supply curve: . Three results follow.
- Too few workers. Competitive firms with the same curves would hire , where meets supply. The monopsonist stops short because each extra hire raises its whole wage bill.
- Pay below productivity. At the wage is below the value of the marginal product. The gap, , is the markdown. It is larger the less elastic the supply of labour to the firm.
- A deadweight loss. Workers between and would produce more than their reservation wages but are not hired. The triangle between and supply over those workers is lost.
- Elasticity of labour supply at 60 workers
- 1.3
- Employment, monopsony (competitive: 60)
- 40.0
- Wage, monopsony (competitive: $20)
- $15.00
- Markdown, VMPE − wM
- $10.00
- Deadweight loss (per hour)
- $100
The perfectly discriminating monopsonist
Section titled “The perfectly discriminating monopsonist”A perfectly discriminating monopsonist pays each worker exactly their reservation wage. Hiring one more worker then costs only that worker’s wage, because nobody else’s pay changes, so the supply curve is the marginal cost curve. The firm hires until equals the supply price, which is the competitive employment level .
- Employment is efficient, so there is no deadweight loss.
- Every worker earns their reservation wage, so worker surplus is zero and the firm keeps the entire surplus.
Perfect wage discrimination is mainly a benchmark. A firm rarely knows each worker’s reservation wage, and paying equally productive workers different wages is hard to sustain when pay is visible or set by a collective agreement.
- Workers hired
- 40
- Wage of the last worker hired
- $15.00
- Producer surplus (per hour)
- $600
- Worker surplus (per hour)
- $200
- Deadweight loss (per hour)
- $100
A minimum wage can raise employment
Section titled “A minimum wage can raise employment”A minimum wage changes the monopsonist’s marginal cost. Up to the number of workers willing to work at , everyone is paid whatever the firm does, so one more hire costs exactly . The marginal cost curve becomes flat at until it reaches the supply curve, then jumps up to the old . The firm hires where meets this new marginal cost curve.
- Minimum wage at or below . The law does not bind and employment stays at .
- Minimum wage between and . The firm hires everyone willing to work at the minimum wage. Wages and employment both rise, along the supply curve, and at the market reaches the competitive outcome.
- Minimum wage above . The firm hires where equals the minimum wage, so employment falls as the minimum rises and some people who want jobs cannot get them. Employment stays above until the minimum wage reaches the value of the marginal product at ; beyond that it falls below the monopsony level.
- Workers hired
- 52.0
- Wage
- $18.00
- Change in employment from 40
- 12.0
- Willing to work but not hired
- 0.0
- Employment, competitive market
- 60.0
- Employment, monopsony
- 52.0
- Wage the monopsonist pays
- $18.00
This is one explanation for studies that find little or no job loss from moderate minimum wage increases: where employers have monopsony power, a minimum wage moves the market toward the competitive outcome. The broader evidence is in minimum wages.
Worked example
Section titled “Worked example”Use the table above: the wage needed to attract each worker, and each worker’s value of marginal product.
-
Nondiscriminating monopsonist. Compare with for each worker: 30 against 10, 26 against 14 and 22 against 18 favour hiring; for the fourth worker 18 is less than 22. The firm hires 3 workers and pays each $14, the wage needed to attract three.
-
Competitive benchmark. Competitive firms hire while covers the wage needed: the fourth worker (18 against 16) is hired, the fifth (14 against 18) is not. Employment is 4, at a wage of at least $16.
-
Perfectly discriminating monopsonist. Marginal cost equals each worker’s own wage, so it hires the same 4 workers as the competitive market but pays them $10, $12, $14 and $16.
-
Minimum wage of $16. The first four workers each cost $16, so is 16 up to four workers; the fourth worker’s of 18 covers it. A fifth would need $18 and a $2 raise for four others, a marginal cost of 26. The firm hires 4 at $16: both the wage and employment rise.
-
Minimum wage of $24. Only the first two workers have of at least 24, so employment falls to 2, below the monopsony level.
Exercise: the linear case in Figure 1
Supply is and . Total labour cost is , so . Setting gives , and the wage from the supply curve is . At , , so the markdown is 10 dollars. A competitive market sets : at . The deadweight loss is dollars an hour. With a minimum wage of $18, everyone willing to work at $18 is hired: .
Evidence
Section titled “Evidence”Nurses at veterans’ hospitals. The US Nurse Pay Act of 1990 replaced a national pay scale for nurses at Department of Veterans Affairs hospitals with wages tied to surveys of local pay. About two-thirds of VA hospitals had been paying below the local market and had to raise wages at once. The size of each raise depended on how far the old national scale had left a hospital below local pay, not on any change in that hospital’s own demand for nurses. Staiger, Spetz and Phibbs used these imposed wage changes to measure how a hospital’s nurse employment responds to its own wage. They estimate the short-run elasticity of labour supply to an individual hospital at around 0.1.1 A competitive employer faces an infinitely elastic supply curve; an elasticity near 0.1 means each hospital faced a steep supply curve and had substantial monopsony power.
The elasticity in a difference-in-differences study is the ratio of two differences. Suppose wages rose 12 percent at hospitals the law affected and 4 percent at comparison hospitals, while nurse employment rose 3 percent and 2 percent. The difference-in-differences change in the wage is 8 percentage points, in employment 1 point, so the elasticity of labour supply to the firm is . A value near zero signals monopsony power; a very large value signals a nearly competitive employer.
Many employers, still monopsony. Monopsony power does not require a single employer. Anything that stops workers from switching employers freely, such as search costs, moving costs, specialized skills or attachment to a workplace, gives each firm an upward-sloping supply curve. Manning’s review of the evidence concludes that employers have considerable monopsony power.2 In online labour markets, Dube, Jacobs, Naidu and Suri estimate labour supply elasticities to individual employers of around 0.1.3 Azar, Marinescu and Steinbaum find that the average local occupational labour market in the United States is highly concentrated by antitrust standards, and that more concentrated markets post lower wages.4
Common mistakes
Section titled “Common mistakes”Check yourself
Section titled “Check yourself”How does a nondiscriminating monopsonist choose employment and the wage?
Compared with a nondiscriminating monopsonist facing the same supply and curves, what does a perfectly discriminating monopsonist do?
A labour market has a nondiscriminating monopsonist. Which minimum wage leads to the highest employment?
Sources
Section titled “Sources”-
Staiger, D. O., J. Spetz and C. S. Phibbs (2010). Is there monopsony in the labor market? Evidence from a natural experiment. Journal of Labor Economics, 28(2), 211 to 236. doi:10.1086/652734. ↩
-
Manning, A. (2021). Monopsony in labor markets: A review. ILR Review, 74(1), 3 to 26. doi:10.1177/0019793920922499. ↩
-
Dube, A., J. Jacobs, S. Naidu and S. Suri (2020). Monopsony in online labor markets. American Economic Review: Insights, 2(1), 33 to 46. doi:10.1257/aeri.20180150. ↩
-
Azar, J., I. Marinescu and M. Steinbaum (2022). Labor market concentration. Journal of Human Resources, 57(S), S167 to S199. doi:10.3368/jhr.monopsony.1218-9914R1. ↩