Labour Markets

No firm wants labour for its own sake — it wants labour because labour produces something the firm can sell. That's derived demand: demand for a factor of production that exists purely because of demand for the final good or service it helps make. The spec names four influences on how much labour a firm demands for a particular occupation: demand for the final product (derived demand itself), the productivity of labour, the price of the product, and the wage rate relative to the price of capital — and every one of those four is really just a different lever on the same underlying quantity, marginal revenue product (MRP_L): the extra revenue a firm earns from employing one more unit of labour, equal to marginal product multiplied by the extra revenue each unit of the resulting output earns. A profit-maximising firm keeps hiring as long as the next worker's MRP_L exceeds the wage — that worker adds more to revenue than to cost — and stops exactly where MRP_L falls to the wage rate.

This is the same marginal product that drove marginal cost in the Costs lesson, just read from the other side of the same coin. There, falling marginal product (once diminishing returns set in) pushed marginal cost up. Here, that same falling marginal product pushes marginal revenue product down as more labour is hired — which is exactly why the demand curve for labour slopes downward. It isn't a separate empirical claim about labour markets; it's the same diminishing-returns fact from Costs, restated in revenue terms instead of cost terms.

Elasticity of demand for labour (spec 1b) turns on how much a wage change actually moves the quantity of labour hired. Two of the clearest, most reliably-tested factors: the proportion of a firm's total costs that labour makes up (a higher proportion makes demand more elastic — a wage rise now moves total costs by more, so the firm responds more), and how easily labour can be substituted with capital (easier substitution makes demand more elastic; the real Jun 2024 mark scheme states the inelastic case directly, in its wage-differentials indicative content — 'if labour cannot be replaced by capital, then demand for labour is more inelastic and wage rate is likely to be higher'). A third factor, the price elasticity of demand for the final product, is itself a derived-demand effect one level up: if consumers barely change how much they buy when the product's price rises, a firm can pass a wage rise straight through to price without losing much output — and without needing to cut back much on the labour that makes it.

The spec's own numbering runs on here in a way worth noting precisely: '1. The demand for labour' has parts (a) and (b); '2. The supply of labour' starts straight at part (c), not (a) — factors influencing supply of labour to a particular occupation: the size of the population, net migration, income tax rates, the level of welfare benefits, government regulations, and trade unions. A country experiencing sustained, large-scale net emigration — more workers leaving to work abroad each year than arriving — sees exactly this kind of leftward shift in its domestic labour-supply curve; a rise in welfare benefits (which raises the return to not working, at the margin) pushes supply the same direction, for a different reason.

Elasticity of supply of labour (spec 2d) mainly turns on how quickly people can actually respond to a wage change with their labour, which is mostly a story about skill and training. An occupation that takes years of qualification, licensing, or apprenticeship to enter has inelastic supply almost by construction — a wage rise this year can't manufacture a fully-trained specialist by next month, however attractive the new pay looks. An occupation needing little training has far more elastic supply, because workers can move into it quickly when the wage rises. Geographical mobility matters too: an occupation where workers are willing to relocate for the right wage has a more elastic supply than one where they aren't.

The total supply of labour to the whole economy — not just to one occupation — depends on the participation rate: the proportion of the working-age population that is actually in the labour force, either employed or actively seeking work. A rise in the participation rate (more people choosing to work or look for work, at a given wage) shifts the economy-wide labour supply curve right, independently of anything happening to any single occupation's wage. Two distinct concepts sit on the other side of this: unemployment is being in the labour force (available and looking) but not currently employed; under-employment is being employed, but for fewer hours than you'd take at the going wage, or in a job below your skill level — a worker counted as 'employed' in the headline statistic can still represent unused labour capacity the participation-rate/unemployment split alone doesn't capture.

Put demand and supply together and, in a genuinely competitive labour market — many employers, many workers, nobody big enough to move the wage by their own hiring or job-seeking decisions alone — the wage is set exactly where MRP_L (demand) meets the supply curve. 'Competitive' is doing real work in that sentence. The rest of this lesson is about what happens to that wage once it stops being true on either side of the market.

The Qm<Qe, Wm<We outcome derived above is the headline mechanism, but the real Jan 2023 mark scheme — a 20-mark essay asking candidates to evaluate the impacts of monopsony power on employees and suppliers, using Walmart as the named context — credits a second KAA harm channel alongside it, genuinely distinct from the wage effect: "monopsonies may pay lower wages and provide poorer working conditions compared to a competitive labour market." Working conditions aren't just the wage effect restated: an employer facing no rival bidder for its workers' labour has weaker pressure to compete on non-wage terms too (hours, safety, flexibility, benefits) for the same underlying reason it has weaker pressure to compete on pay — workers with no comparable alternative employer have nowhere else to take either complaint.

Two further points from the same mark scheme cut the other way, and both are genuine evaluation rather than the harm restated with 'however' attached. First: "job security is likely to be higher than in more competitive markets" — an employer with no rival bidding for the same workers has less reason to make them redundant when demand dips, since there's no competitor waiting to hire a laid-off worker away, and no bidding-war risk to defend against by keeping headcount lean. Second, genuinely separate from the trade-union case in the drill below: "the firm may not be the only employer of workers therefore they may pay higher wages than a pure monopsony" — real labour markets are rarely a textbook-pure monopsony with literally one buyer, and where a second or third employer of similar workers exists nearby, even a dominant employer faces some competitive pressure on wages. The DEGREE of monopsony power, not just its presence or absence, is itself something a strong evaluation weighs.

Spec item 3(c) names a specific non-competitive case: wage setting in the public sector/state-owned enterprises. The link to the diagram just derived is direct — a national health service, a state railway, or a single national employer of a specialised public-sector role is very often the dominant, sometimes the only, buyer of that particular kind of labour in its area, which is a monopsony position by definition. A real, verified exam question confirms the mechanism carries straight over: a June 2022 question describes a country where a competitive labour market for nurses became a monopsony once a universal healthcare system made the state the dominant employer of nursing labour, and asks what happens to wages and employment as a result — the correct answer is that both fall, exactly the Qm<Qe, Wm<We outcome the diagram above derives. Being a public-sector buyer doesn't exempt an employer from the ACL/MCL mechanics; it just changes what the employer does with the buying power that mechanism hands it.

But the mark scheme is careful not to assume this employer then behaves exactly like a private profit-maximising monopsonist: the real, verified point is that "many monopsony employers are governments that do not profit-maximise and may seek to prioritise other objectives such as well-being." In practice this usually means wages are set administratively — through a pay-review body, a national pay scale, or a budget negotiated politically — rather than by solving MRP_L=MCL directly.

That doesn't mean market forces are irrelevant to a public-sector wage, though. A government that sets pay too far below what comparable private-sector roles offer will struggle to recruit and retain staff — real recruitment and retention data is exactly what a pay-review body is normally asked to weigh. The honest statement of the mechanism is that public-sector wage-setting substitutes an administrative, budget-constrained process for the direct MRP_L=MCL rule, while remaining indirectly disciplined by the same labour-market forces the rest of this lesson derives — not that it escapes them entirely.