A single buyer of labour: supply as the average cost of labour, the marginal cost of labour above it, employment where MRP = MCL, the wage read from supply, the exploitation gap, and why a minimum wage or trade union can raise both pay and employment. Through the NHS pay award of February 2026.
On 12 February 2026 the health secretary announced that the 1.4 million NHS staff in England on Agenda for Change contracts, nurses, midwives, paramedics, porters and more, would get a 3.3% pay rise from April. The NHS Pay Review Body had recommended it; the Department of Health had earlier told the review body it could afford no more than 2.5%; the previous year’s award was 3.6%. The Royal College of Nursing called 3.3% an insult, having walked out of the review body process, and pointed out that it was below the inflation rate at the time.
Nurses in England have, in practice, one employer. A nurse unhappy with 3.3% cannot cross the road to a rival hospital group offering 5%, because there is none of any size: the NHS employs the great majority of nurses in the country. Agencies, private hospitals and emigration are the exits, and all are costly.
What happens to wages and jobs when there is only one buyer of a kind of labour?
Source: Written ministerial statement HCWS1340, 12 February 2026; Royal College of Nursing, 12 February 2026; NHS Employers, 12 February 2026.
A monopsony is a market with a single buyer. In a labour market it is a single employer, or one so dominant that it sets the wage rather than taking it: the NHS for nurses, the Ministry of Defence for soldiers, a mine or mill in a one-company town, or in a weaker form a large supermarket chain in a small town.
The monopsonist faces the whole supply curve of labour: to hire more workers it must offer a higher wage, and, because it pays everyone the same, the higher wage goes to the workers it already has. So the marginal cost of labour (MCL), the cost of one more worker, is greater than the wage: it is the new worker’s wage plus the rise paid to all the others. The supply curve is the average cost of labour (ACL), and MCL lies above it.
A profit-maximising monopsonist hires where MRP = MCL and pays the wage on the supply curve at that employment, which is below MRP. The gap between what the last worker adds and what they are paid is monopsonistic exploitation.
Because the wage is below MRP, a minimum wage or a trade union can raise both the wage and employment, up to the competitive level. A union facing a monopsonist is a bilateral monopoly: the outcome depends on bargaining.
On 12 February 2026 the government set the pay rise for 1.4 million NHS staff in England at 3.3%, after telling the pay review body it could afford no more than 2.5%.
No other employer in the country could decide the pay of 1.4 million people in one statement.
Why can the NHS?
The NHS is a monopsony: one buyer facing many sellers. A nurse who wants a higher wage has few alternatives short of leaving nursing or the country. A single buyer, like a single seller, has market power, and uses it to pay less than the value of the work. That is what this lesson draws.
Start with the benchmark. The demand for nurses is their **marginal revenue product**, falling as more are hired. The supply of nurses rises with the wage. With many competing employers the market clears where they cross: 12,000 nurses at £36,000, Lc and Wc. Each hospital hires until the last nurse adds exactly her wage to revenue.
Check: In this labour market, if there were many competing employers, how many nurses would be employed and at what wage?
With many employers each takes the wage as given and hires until MRP equals the wage, so the market clears where demand (MRP) meets supply (ACL).
Now a single employer. The supply curve is what it must pay per nurse: the average cost of labour. But hiring one more means a higher wage for every nurse already on the payroll, so the marginal cost of labour is above the wage, and the gap grows with the workforce. On the diagram **MCL** rises twice as steeply as supply.
Teacher tip: With a straight-line supply curve from the wage axis, MCL has the same intercept and twice the slope, like MR under a straight demand curve. Draw it that way and the geometry takes care of itself.
Check: The NHS employs 8,000 nurses at £28,000. To hire one more it must raise the wage to £28,002 for everyone. The marginal cost of that nurse is:
The monopsonist pays a single wage, so the cost of the extra worker includes the raise given to the whole existing workforce. MCL lies above the supply curve.
The NHS hires while a nurse adds more to revenue, or to patients treated, than she adds to the wage bill. That is MRP = MCL, at **Lm = 8,000**, where the 8,000th nurse is worth £44,000 and costs £44,000 at the margin. Fewer than the competitive 12,000. The single buyer restricts employment, as a single seller restricts output.
Check: The monopsonist maximises profit by employing nurses up to the point where:
A worker is worth hiring while she adds more to revenue than she adds to total labour cost. For a monopsonist that cost is MCL, not the wage.
Then the wage. The NHS needs 8,000 nurses and the supply curve says they will work for £28,000, so that is what it pays: **Wm**, read down to supply, not across to MRP. The 8,000th nurse adds £44,000 and is paid £28,000. The £16,000 gap is monopsonistic exploitation: the single buyer’s version of a monopolist’s mark-up.
Check: Having chosen to employ 8,000 nurses, what wage does the monopsonist pay?
The employer pays no more than it must. The supply curve says 8,000 nurses are available at £28,000, so that is the wage.
Here is the result that surprises people. Set a minimum wage at £36,000. The NHS can no longer hire cheaply at the margin, but it also no longer raises everyone’s pay to hire one more: the minimum wage **is** the marginal cost of labour, flat, up to 12,000. It hires until MRP = £36,000: 12,000 nurses. The wage floor raises **both** pay and employment, to the competitive level.
Check: A minimum wage of £36,000 is imposed. The monopsonist now employs:
With the wage fixed at £36,000 the employer no longer raises everyone’s pay to hire one more, so MCL is flat at £36,000 up to the point where supply would demand more. It hires until MRP = £36,000: 12,000.
A trade union does the same job as a minimum wage: by refusing to supply labour below the negotiated wage it flattens MCL. The RCN against the NHS is a bilateral monopoly: one seller of labour facing one buyer. The wage lands between the two sides’ positions, depending on bargaining strength. The pay review body exists to find that point without strikes.
Check: The Royal College of Nursing negotiates a wage of £40,000 with the NHS. On the diagram, employment is:
At £40,000 the employer hires where MRP = £40,000, which is 10,000. A union can raise both wages and employment against a monopsonist, up to the competitive wage.
The wage is read across from the MRP curve at Lm, giving £44,000. That is what the last nurse is worth, not what she is paid. The monopsonist pays the least that attracts 8,000 nurses: £28,000, on the supply curve.
Using a diagram, explain how a monopsony employer such as the NHS determines the wage and level of employment, and evaluate whether a higher minimum wage or stronger trade union would reduce employment in such a market.
A monopsony is a market with a single buyer. The NHS employs the great majority of nurses in England and sets their pay nationally, so a nurse who wants a higher wage has few alternatives: it is close to a monopsony buyer of nursing labour.
Because it faces the whole supply curve, hiring one more nurse means offering a higher wage, and since it pays a single rate the rise goes to every nurse already employed. The marginal cost of labour therefore lies above the supply curve, which is the average cost of labour, and rises faster. A profit-maximising monopsonist employs where the marginal revenue product of labour equals the marginal cost of labour, at Lm, and then pays the lowest wage that attracts that many workers, read from the supply curve: Wm. On the diagram Lm is below the competitive employment Lc, where MRP meets supply, and Wm is below both the competitive wage and the MRP of the last worker; the gap between MRP and Wm is monopsonistic exploitation.
A minimum wage or a union-negotiated wage set between Wm and Wc changes the employer’s calculation. The wage can no longer be bid down, so the marginal cost of hiring one more worker is simply the fixed wage, up to the point where supply would require more. The employer hires until MRP equals that wage, which is further along the MRP curve than Lm. Both the wage and employment rise; at the competitive wage, employment reaches Lc.
**However,** this depends on where the floor is set: above Wc employment falls back along MRP, though it may still exceed Lm. It also depends on how much monopsony power the NHS has. Nurses can move to agencies, private hospitals or abroad, so the supply curve is flatter than the model assumes and the exploitation smaller. The NHS is budget-constrained rather than profit-maximising, so low pay shows up as vacancies and waiting lists rather than profit, and a pay rise unfunded by the Treasury could mean fewer posts. The evidence after the UK minimum wage was introduced in 1999 and raised sharply after 2016 found little job loss, which supports the monopsony view of many low-wage labour markets. A moderate rise in the nurses’ wage is therefore unlikely to reduce employment and may raise it; a large one, without extra funding, might.
NHS pay award: written ministerial statement HCWS1340, 12 February 2026; Royal College of Nursing statement, 12 February 2026. The curves on the diagram are for an invented labour market, chosen so the numbers are clear. Diagrams drawn with Rough.js and perfect-freehand, both MIT licensed.