The annual ritual

Every January, a similar story does the rounds. According to the High Pay Centre, “the bosses of FTSE 100 companies will have made more money in 2026 before midday on Tuesday [6th January 2026] than the average worker will all year”¹. Median FTSE 100 CEO pay is £4.4m. This is a 113 times the £39,039 earned by the median full-time worker.²

The numbers are produced. The unions are wheeled out. The complaint is similar every year. The TUC’s General Secretary says that “while millions of low- and middle-income workers are still struggling with the cost of living, those at the very top keep helping themselves to a huge slice of the pie”, and calls for workers to be given a seat on executive pay committees.¹The High Pay Centre adds that “the idea that executives, as a class, are individually contributing over 100 times more in value than the workers they rely on is simply not credible”.¹

We want to make two points against this complaint. First, a wage is not a reward for what you contribute. Not for the worker and not for the CEO. Second, the worker and the CEO are not paid to do the same thing, and once you see what each is paid for, the pay gap stops being a moral puzzle.

A wage is not a reward

The union complaint takes it for granted that a wage should track what someone contributes, and that the scandal is that it fails to. Carers, delivery drivers and shop staff, the TUC says, “give more than they get back”.³ The implication is that, properly calibrated, a wage would reflect the work performed. The director of the High Pay Centre has put this directly: a maximum pay ratio would help ensure that all workers received “a fair reward for their contribution to business success”.⁴

This is wrong about what a wage is.

At first appearance, a wage is what an employer pays to get someone to turn up and work. From the worker’s side, it is the money they need to live: rent, food, transport, the rest. From the employer’s side, it is a cost and one of several inputs (materials, machines, workers) bought in order to produce something that can be sold for more than the inputs cost.

This sets the floor and the ceiling of the ordinary wage. The floor: across the workforce as a whole, wages have to cover what it costs workers to keep showing up; to eat, sleep, raise the next generation of workers. If they don’t, the supply of workers dries up, which is why the state steps in with minimum wages and welfare programmes. Not out of kindness, but to keep the labour force in working order.⁵ Whether any particular worker can actually live on their wage is, from the firm’s point of view, the worker’s problem.

The ceiling: a company buys or rents “inputs” (such as raw materials, machines, factory buildings) in order to produce “outputs” which earn more than the costs of the inputs added up, counted as profit. A firm hires people on the calculation that the what it can get out of them is worth more than the wages it pays them for it. That is the whole point of hiring anyone.

What the employer buys for the wage is not the worker’s output. It is their ability to labour for a given stretch of time – eight hours, a shift, a month. What actually gets done in that time is a separate question, settled by supervision, deadlines, targets, the threat of being replaced. Anyone who has worked a hard shift under a manager on their back, and an easy shift when no one was looking, knows that the wage paid was the same in both cases.

From the employer’s point of view, the same wage can buy a lot of work or a little. Hence the constant pressure on breaks, shift patterns, productivity targets and headcount: not arbitrary cruelty, but the routine attempt to widen the gap between what workers cost and what can be got out of them. The wage itself is the outcome of a power struggle over that gap, which is why it has to be fought over year after year.

So when the unions say that ordinary workers “give more than they get back”, they are correct wand they are describing the normal functioning of paid employment, not its corruption. A worker whose expected output was exactly equal to their wage would not be employed. The wage was never a reward. The worker’s “contribution to business success” is exactly this: being paid a little, working a lot, and the difference going to the firm.

The CEO is paid to do a different job

What the CEO⁶ is paid for is not what the worker is paid for. Once that is clear, the comparison with their pay starts to look strange – the same number, a sum of money paid for a period of work, used to compare two roles that are not the same kind of role at all.

The “ordinary worker” is hired to produce: to assemble, to drive, to care, to serve. What caps their wage is what they themselves can be made to produce in the time they are bought for. Pay them more than that and the employer loses money on them. This is why, for the vast majority of jobs, the calculation runs in one direction: wages down, intensity of work up⁷.

A CEO is not hired to produce. A CEO is hired to make the firm profitable, and to keep it profitable against every other firm trying to do the same. That means organising the people who do the producing: which lines to expand, which to close, how the workforce is structured, how breaks are timed, how performance is measured, how unions are handled, how technology is brought in to intensify the work or replace it. It also means everything that sits around production: which markets to enter and exit, which competitors to undercut or buy out, which suppliers to squeeze, how to position the firm with investors and lenders, how to lobby governments and regulators. The common thread is that all of these decisions are made on behalf of the firm’s profits, against the claims of workers, rivals and suppliers.

Take a single example. A CEO signs off on a new system – handheld scanners on the warehouse floor, monitoring software in the call centre, productivity dashboards in the office – that measures the work as it is done. Every task is timed, every gap between tasks logged, every pattern in the data fed back into the next round of targets. Over the next year, the same workforce produces noticeably more for the same wage bill. The slack that any human pace of work contains, the seconds between tasks, the small, previously unmeasured pauses, has been found and taken back.

The CEO has not produced more widgets. The CEO has reorganised the conditions under which other people produce them, so that more comes out for the same money going in. That is the service being bought. Multiply across a year, and across the scale of a large firm, and you can see why this kind of work can be paid for at a different order of magnitude from the work it organises.

This is why the cap that applies to the ordinary worker’s wage does not apply here. A worker’s pay is limited by their own output. A CEO’s pay is limited by other people’s output – by how much extra can be squeezed out of the workforce they direct, and by how well they position the firm against its rivals. An executive who increases the surplus extracted from ten thousand workers by a few percentage points has, from the company’s point of view, “earned” the equivalent of hundreds of those workers’ wages, and can be paid accordingly without the calculation going wrong.

The structure of executive pay reflects this. CEOs are not paid by the hour. A large share of their income comes as share options, performance bonuses, and long-term incentive plans tied to profit targets, share prices, or returns to investors. This is not a quirk of generosity. It binds the executive’s own income to more profit for the firm: the more they wring out of the operation, the more they personally take home. “Ordinary wages” are structured the opposite way: as a fixed cost the firm tries to hold down regardless of how well the business is doing. What they are paid in is money. What they are paid for is different.

None of this is because companies are sentimental about executives. They would prefer to pay less for them, as they would prefer to pay less for anything. They compete for them, they fire them when they underperform, they design pay packages to keep them on task. But the calculation is structurally different. For the “median” worker, the wage is a cost to be held down: every pound paid is a pound the firm would rather have kept. For the executive, the wage is an investment in a larger return wrung out of everyone else. Paying more can pay off.

So what is wrong with the union complaint?

The standard form of the complaint is a ratio. The average FTSE 100 CEO is paid 113 times the median full-time worker’s salary²; at the extremes, the figure runs higher still. The chief executive of Mitie was paid 575 times the median pay at the firm in 2023-24.⁴ Twenty years ago the figure was lower; in the 1960s, lower still. The ratio is offered as a measure of how far things have gone wrong, and the implied remedy is to bring it back down through pay caps, worker representation on remuneration committees, profit-sharing schemes, or simply public shaming⁸.

The historical movement of the ratio is real and worth taking seriously. The calculation that produces it – workers bought as cheaply as possible, surplus extracted, executives paid to maximise the extraction – has not changed. But how that calculation comes out at any moment depends on conditions. When unions are strong, workers can hold the floor of their wages higher and slow the intensification of work. When unions are weak, the same calculation produces a wider gap.

That the fight has to be fought at all is the point. The wage is not a measure being honoured or violated; it is whatever the two sides can force it to be.

The problem with the ratio is not that the numbers are wrong. It is that the two numbers are not measuring the same thing. A worker’s wage and a CEO’s compensation package are both denominated in pounds, but they are paid for opposite roles in the same operation; one to be squeezed, the other to do the squeezing. Pointing to the “gap” between them does not explain why it exists.

The union complaint gets several things wrong at once. It treats the wage as a reward, when it is a cost. It treats the gap between worker pay and CEO pay as a moral failure. It is the predictable result of two calculations the same firm makes at once: one to minimise the cost of labour, the other to maximise the surplus extracted from it.

It implies that if only remuneration committees were better designed, or workers were given a seat at the table, the pay structure could be made to reflect “fair” contribution. It cannot. The whole arrangement runs on the gap between what workers cost and what can be got out of them. That is what a capitalist firm is for.

The unions are not wrong that workers give more than they get back. They are wrong about what to conclude. What their members suffer from is not that executives are greedy. It is being on the wrong side of the same calculation that pays those executives so well: being bought as cheaply as possible, worked as hard as possible, for as long as their employer can get away with. The CEO is not a parasite on this arrangement. The CEO is the person paid to enforce it. Changing who fills that role, or how much they are paid, leaves the arrangement intact.

The union complaint objects to the result of the calculation without explaining what produces it. What produces it is the arrangement itself: a firm whose business is to extract more from its workers than it pays them, and an executive paid well to make sure it can.

The union’s remedy is then based on this misunderstanding. Rebalancing the pay structure of an endeavour that exists, in the first place, to extract a surplus out of workers; not for the lavish lifestyle of the CEO (this is rather a by-product) but for the production of profit.