Against 'equal value'

A tricky argument, but an essential one

Occasionally, a metaphor arises that’s too potent not to seize on and wring out, shamelessly. Lately that’s been Birmingham’s bin strikes, where an accretion of bad laws, poor governance and feuding unions has taken on pungent physical form: bin bags have littered the streets, to the discomfort and horror of the city’s residents.

That offered ample raw material for me to write a fairly feisty leader article for The Economist (not usually a publication to dwell on how local rats “feasted, and multiplied”)—arguing against ‘equal value’ pay rules, the underlying cause of the strikes and the billion-pound payouts that have bankrupted Birmingham.

“[Equal value laws] penalise employers for disparities between sometimes wildly different roles—binmen and cleaners, warehouse stockers and shopworkers—if judges can be sold on the idea that the work was of “equal value”. Never mind whether the market concurs.”

You can read the full leader here

Still, while there’s nothing wrong with plucking low-hanging writerly fruit every now and then (someone must!), I also thought there’d be value in working through the topic a little more methodically. Much of the public conversation about ‘equal value’ has been a mess¹ and—given the ongoing claims against most of Britain’s supermarkets, plus several more local authorities—the issue isn’t going away.²

That’s what the rest of this post aims to do. People familiar with the issue may want to skip the first section or two. For those skimming, I’d particularly recommend Section 4, on the mechanics of how equal value assessments work in practice, which gets to a level of gory detail I haven’t seen elsewhere in press coverage of the issue.

1. Why care at all?

But to start: a pitch for how this matters, in case any readers are (unfathomably) on the fence over whether to keep reading about equal pay, local government finances and the stealthy comeback of central planning in Britain.

In short: I’m not sure most Brits yet appreciate the sheer scale of these equal value cases and the chaos they’ve unleashed. Start with the claims against local authorities, which are generally further along:

• When all’s said and done, Birmingham will probably end up paying out at least £1.3bn, possibly more. In 2015, the city council flogged two exhibition centres and two arenas to private-equity buyers at fire-sale prices.³ Council taxes have doubled over the past decade and public services have been cut. Another round of assets sales is happening now.

• Two tranches of equal-value claims hit Glasgow, one for ~£500m in 2019 and another for ~£800m in 2023. That’s been paid for by selling (and leasing back) landmark buildings like the main art museum and the city council chambers, as well as a a big lift to council taxes and wider cutbacks.

• A case in Sheffield is ongoing, but could well cost hundreds of millions.

• North Lanarkshire has paid out about £100m

• There have also been claims worth tens of millions or more in Cumbria, Dumfries and Galloway, and Newcastle, plus ongoing cases in Sunderland, Fife, Brighton and Hove, and Dundee⁴.

And in the private sector, 70% of British grocery chains by market share are in the equal value crosshairs. Each of Asda, Tesco, Morrison’s and Sainsbury’s could be on the hook for a billion or more in back-pay, plus a hefty boost to their future wage bill (which, in a low-margin business like supermarkets, flows straight into prices). There are also smaller cases with Next, Boots, and more.

It’s worth pausing for a second to take in the colossal scale here. The ultimate value of the total liabilities, across the public and private sectors, could easily end up in eleven digits.

Given these claims have hobbled some of Britain’s most important cities for well over a decade⁵ , you’d hope that they’d be exceptionally well-founded. Virtually all press coverage, at least until the past year or so, has presumed so⁶.

That strikes me as utterly delinquent: a failure either to explain how unusual Britain’s situation is compared to peers internationally⁷, or how precarious the economic logic underlying the law is. Whichever side of the issue you ultimately fall on, the sheer scale of the mess means it’s worth seriously interrogating whether the chaos has been worth it.

2. Defining ‘equal value’

Before we get there, what exactly is ‘equal value’? Usually, when people think of equal pay cases, the situation that comes to mind is when someone is paid less than a coworker despite doing near-identical work, because of gender, race, or another protected characteristic. That sort of unpleasant but clear-cut discrimination still happens (especially when you include subtler instances like unfair denial of promotion and so on), but has thankfully gotten rarer.

Equal value is something else: looking across entire professions, not just person-to-person. Statistically, many jobs skew male (builders, factory workers, binmen) or female (nurses, cleaners, teachers). Under equal value, an employer can be on the hook for pay discrimination if a judge determines that the work done by one female-heavy category of employees is of the same value as that of a male-heavy category, but the former is paid less than the latter. Hence, cleaners get compared with binmen. Since within-sector gender disparities have shrunk (see chart below), these sort of cross-sector gaps now matter more in explaining the overall gender pay gap.

Source: The Times

Winning these cases doesn’t require proving sexist intent. Nor is the fact that a particular firm’s payscales track the going market rate for those jobs an adequate defence. Instead, everything hinges on whether judges decide that two roles are of equal value.

This idea has a long history in Britain. Famously the women sewing car seat covers walked out of the Ford plant in Dagenham in 1968 over equal value logic: that they were underpaid compared to the men in the factory, despite doing work that was equally necessary to get the cars on the road,

Equal value was legally enshrined in the 1980s, but the sorts of big class-action cases we see today only got going in the 2000s. I walk through the history in my leader.

“Britain did not originally recognise “equal value”; Barbara Castle, the Labour minister who introduced the Equal Pay Act in 1970, thought it too woolly. But it caught on across the English Channel, and was added to the law books in the 1980s, under pressure from Brussels. Trouble began in the 2000s, after a ruling in the late 1990s by the European Court of Justice, which let victors claim six years of back pay, making a win more lucrative. Alongside Britain’s relaxed rules about no-win-no-fee lawyers and higher levels of pay transparency, that proved a potent lure.”

Doing away with the equal value principle would be a momentous change. One lawyer that I spoke to compared it to “removing a limb” from British equal pay jurisprudence, so established the principle now is. Certainly, it would require primary legislation. After Brexit, Britain can do so without falling afoul of EU rules. But for the moment, that doesn’t look like a path the current government is especially interested in taking.

3. Should the market be the judge?

Now we get to the rub: are today’s equal value cases an admirable reckoning with injustice, or rent-seeking and economic distortion cloaked in the language of equality? That largely comes down to how you answer one question:

• What share of the unexplained pay variation between jobs, after experts have compared the day-to-day work, is down to implicit society-wide sexism, versus differences that the assessments have failed to capture but the market has?

A valid answer—and where I’ve landed—is that ascribing much of the unexplained gap to sexism was sensible in 1960s era of the Dagenham strikes, when overt sexism was rampant and workplaces were often totally gender-segregated, but pretty tough-to-justify today.

That isn’t, as some commentators theorise, because of ‘woke’ judges or a poorly-run process. If you read through the judgements on equal pay cases, the overwhelming impression that comes across is of well-intentioned people trying to do a fundamentally impossible job. And employment lawyers I’ve spoken to don’t see these cases as expansive judicial activism, but instead a pretty clear application of the law as written. Still, no matter how hard the courts try, it’s all-but-impossible to keep these decisions from being wildly subjective.

The best way to see that is by walking through a real-life instance of how these decisions are reached. The process quickly becomes a bizarre sort of judicial central-planning, where an outward rigor masks flim-flam underlying logic. Let’s work through it on its own merits, before getting to why the underlying philosophy of the approach is dubious too.

4. Litigating equal value claims

Here, I focus on the Asda equal pay judgement, which isn’t quite finished yet but is trundling toward a £1.2bn payout. The core claim was that shopworkers (more often female) were unfairly paid worse than warehouse staff (more often male).

The two sides battled out over how to compare the roles. Eventually, the court settled on an eleven-factor model. For each factor, every job gets banded between A (50 points) and E (10 points). The chart below shows the comparison between the top jobs at the shop (yellow) and warehouse (orange)—the court determined that the physical demands were higher in the warehouse, but that the emotional demands and broader responsibilities were higher in the shop.

At one level, that isn’t totally mad. Directionally, customer-facing shop work probably does demand more emotional nous, while jobs at the warehouse are more physically taxing. But dig deeper, and the implicit assumptions rapidly start to look rather wild.

For one, for this approach to be remotely defensible, the eleven categories would have be be roughly mutually exclusive and collectively exhaustive (MECE, as the consultants say)⁸. I challenge anyone to argue that there is no overlap between “emotional demand” and “mental demands”, or “physical skills”, “physical demands” and “working conditions”.

And even if that hefty bar gets met, then comes the question of how to weigh the categories. Implicitly, the judgement has chosen to treat them all equally, but there’s no real reason to think that “responsibility for assets” and “physical skills”, for instance, are equally important. The judgement is extraordinarily sensitive to that assumption; fiddle a bit with the weights and it’s not difficult to reach whichever conclusion you want.

But add up the raw scores as they’re laid out and you get a score of 460 for the warehouse team-manager and 440 for the shop section-leader. Then, it gets wackier. During the case, there was a dispute over whether responsibility over “customer goodwill” should have been added as an extra category in the schema. The compromise was to include it as a tiebreaker category. The other tiebreaker category is “marginality”, which adjusts the scores for cases where a role just scraped into a higher category.

Those two adjustments, worth 40 points, knock the shop section-leader into the lead, at 480. But the tolerance band for equal value was set at 20 points, and so the jobs end up being deemed of equal value.⁹ The table below summarises how the process worked. This was repeated across the judgement for dozens of pairwise comparisons.

5. Bad on paper

Even on its internal logic, this approach looks flimsy to me: its hand-waving pseudo-rigour is sensitive to impossible-to-justify assumptions, and vastly unlikely to pin down the differences between two jobs well enough to make it possible to confidently ascribe the remainder to sexism.

But worse, none of these attributes are static or objective; each person stepping into the job market has their own preferences. For some, physical labour might be a deal-breaker; others might not mind at all. Delightfully, markets are a social technology that’s spectacularly good at finding the right wage to sell just enough people on doing a job, and parcelling out roles and salaries accordingly.

It is worth being cautious: the market-fundamentalist view that any sort of pay discrimination is incompatible with a market economy (a sort of no-arbitrage logic, where if one employer undervalued female workers’ productivity, another would scoop them up) clearly goes too far. But in a context where outright sexism has abated a good bit, and when the jobs in question are gender-integrated, just not perfectly 50-50, an automatic presumption that unexplained salary variation must be down to ambient sexism is difficult to sustain.

One anecdote supporting that: Next, the clothes retailer which lost an equal-value judgement last year, tried in 2021 to get shopworkers to transfer to the warehouse. They made the offer to 25,000 workers. 44 indicated interest, 9 tried a taster week, 7 switched jobs, only 4 lasted beyond a year. If the jobs were indeed interchangeable, this should have been a far more popular choice.

Reading through the Next judgement, another piece that jumped out was the notion that keeping costs down was a legally illegitimate reason for pay differences. In a competitive market economy, this, again, strikes me as utterly wild—keeping costs down as far as possible, and ideally passing those on to consumers in cheaper clothes, is exactly what a business like Next ought to be doing.

“The aim was just a financial one; the respondents could have afforded to pay a higher rate of basic pay to retail staff as some competitors did, but paramountcy was given to keeping costs in the labour budgets to their minimum.”

Digging deeper, though, I also think there’s a real philosophical incoherence to the entire notion of ‘value’ that underpins the equal value framework. A traditional equal-pay dispute comparing two people doing similar work doesn’t rely on determining their value: the issue is purely whether they were performing the same tasks, irrespective of those tasks’ worth.

But once you get to comparing different jobs, you do need that common currency of value. And given the entire substance of these claims is that market wages aren’t correctly reflecting that value, you need a basis that isn’t monetary to compare them. That ends up quite quickly getting you to a fairly dodgy sort of Marxian ‘labour theory of value’ logic. The very idea that wages can or should reflect something deeper and more essential that an employer’s willingness to pay for a particular service (and an employee’s corresponding willingness to sell their time) is, conceptually, rather a messy can of worms to be opening.¹⁰

6. Worse in practice

All that gets to the illogic of equal value as an idea, and how shaky it is as a foundation for billion-pound city-paralysing payouts. But actually implementing these rules runs you into yet more trouble. There are four specific issues worth calling out.

1. Persistent distortions. Judicial rulings can’t change the underlying labour market facts of which jobs workers deem more desirable. When an employer loses, they have two choices. Either they raise the pay of the worse-paid job, as the private sector mostly seems to be doing. That becomes a semi-arbitrary pay subsidy for one group, which the rest of society pays for in higher retail prices. Or, as cash-strapped councils often prefer, they pull down the better-paid role. That causes bigger issues: retention can become a problem, and efforts to bridge that gap (as Birmingham did, with one-off bonuses, new jobs and the like) can invite new equal value claims, forcing bigger payouts.

2. Job evaluations (JEs), a shadow payroll tax. The best way to get around equal value claims is to proactively run job evaluations, which regularly audit a firm’s workforce for gender discrepancies. These aren’t a perfect legal shield but do help quite a bit. The most compelling counter-argument I’ve encountered from equal value defenders is that employers just need to get better about running JEs and the issue would mostly go away. There’s some truth to that, but it’s also worth playing it out—doing so means an ongoing, perpetual subsidy by employers to law firms and HR consultants. I’ve had a decent go at digging around to get a sense of the running cost of keeping JEs up-to-date, and there’s vanishingly little public information out there. Even the cheapest software solutions cost £5-10k per year. At a guesstimate, conducting a full, bespoke evaluation for a big employer probably costs around two orders of magnitude more initially (i.e. a one-off cost of low single digit millions), and one order of magnitude more in ongoing costs (hundreds of thousands annually).¹¹ This functions as a de facto payroll tax on employers, only the receipts go to consultants not schools and hospitals.

3. The scale of back-pay. One reason why equal value cases in Britain have been so crippling is the generous six-year back-pay window. Even if you take the view that equal value rules are reasonable, an alternative approach would be to focus on changing forward-looking pay, rather than saddling public bodies with vast liabilities for historic disparities.¹²

4. Outsourcing. Employers can sidestep equal value rules by outsourcing hiring. That’s why claims mainly hit Labour and SNP councils, who find that sort of thing politically uncomfortable. But having businesses decide whether to outsource entirely on the basis of equal-pay rules isn’t exactly ideal. Now, the government wants to extend equal value to cover outsourced workers. That would entrench the principle at a time when it’d be much better to dial it back. Doing so also raises all sorts of further issues. Might subcontractors then be forced to set separate bespoke payscales for every employer they work with, depending on who else was in that employer’s workforce?

Source: Labour’s campaign materials. Government plans here.

7. A political landmine?

Undoing equal value would be messy, and requires primary legislation. But the toughest issue is even starting the public conversation. For obvious and laudable reasons, pay equality strikes most people as a good thing. But the result is that we’ve kneecapped local government for over a decade in cities across Britain, and not had an open conversation about why.

One line of thinking goes that this is all too much of a landmine. Belatedly (and expensively), employers are learning to deal with the equal value regime with some combination of JEs and outsourcing. And as painful as adding another compliance hoop is, the worst is possibly behind. Frankly, as someone who’s much more worried about the real-world consequences of these rules than any culture-warrior crusades, I can see the merit in that outlook.

On reflection, though, I do also think inaction would be a mistake. Partly that’s because the ongoing costs really aren’t that trivial—as I’ve laid out above. But partly because the very fact that this is a tricky issue makes the case for a proper public conversation more urgent.

The two sides’ incentives to discuss equal value are totally askew: supporters get the adrenaline-rush of pushing for equality and justice and sceptics (at least for those of us unwilling to go down the woke-bashing route) have to settle for bewailing central planning. And, much as I wish the British public shared my visceral discomfort towards overzealous dirigisme, my consistent experience has been that they do not. (Although: if any pollsters are reading this and want to run some questions putting a few equal value scenarios to voters, do get in touch. My sense is the core intuition that cashiers and warehouse workers do vastly different jobs is one that cuts through, and I’d love to test that.)

But in the meantime, equal value remains bad, messy and distortive policy. Worse, the government seems set on expanding its scope, as does the European Commission¹³. So, as tricky as the issue is, it’s a conversation that needs to be had.

1. With the notable exception of my colleague Millie Wood’s excellent reporting on the issue for The Economist, which has really informed my own thinking. Her articles are here and here, and she’s also worth following on Twitter.

2. Nor, frankly, is it entirely clear-cut; I have thoughtful friends on the other side of this argument, who just have different intuitions on how well labour markets work and how effectively the law can remedy social ills.

3. Technically, it sold a leasehold—the council remains the freeholder. Troublingly, the initial PE buyers were able to resell the assets for around £800m a few years later, more than doubling their money. Feel free to draw your own conclusions on whether the council was taken for a ride.

4. Observant readers will notice that these are all left-wing councils, usually Labour or SNP. That isn’t a coincidence, as we’ll get to.

5. Lots of the councils involved also made a hash of things in many other ways—Birmingham blew tens of millions on a botched IT transition, and could have limited their liabilities by settling sooner—but the core issue is whether councils should have been in this position in the first place.

6. I’ve failed to surface an op-ed in a mainstream publication questioning whether ‘equal value’ was a sensible way to approach pay equality from before 2023 or so. Some may well be out there, but vanishingly few! Credit to The Spectator, who flagged the issue in 2023. And it does now feel like the dam is breaking (hopefully The Economist’s coverage over the past few months has helped there)—we’re now seeing ‘equal value’ getting gently questioned in The Times, too.

7. America’s federal equal pay laws wouldn’t accommodate these claims (some state-level laws get closer, but remain more tightly-drafted); EU rules would (and indeed come from the same source as British ones), but in practice claims of this scale haven’t come up—mostly due to a combination of legal funding norms (fewer no-win-no-fee lawyers) and less salary transparency. But, as I touch on at the end of the the leader, that may change in the EU when the Pay Transparency Directive comes into force in 2026, depending on exactly how member states decide to implement it.

8. At least one ex-MBB consultant who reviewed a draft of this post confirmed to me that this was “not MECE”, in their view.

9. Not to moan, but briefly to moan: the 300-page judgement pdfs make these scores exceptionally hard to pull out (they’re spread over dozens of pages in the document), making proper outside scrutiny much harder. To illustrate that point: all three of o3, o4-mini-high and Manus made mistakes when asked to pull out the sub-scores comparing the top store and warehouse jobs—I ended up having to go through and manually extract them myself. At this stage having documents that are confusing enough to flummox a top-end LLM is frankly rather impressive!

10. Society’s developed a helpful set of norms to separate monetary compensation from any sort of deep intrinsic sense of value—hence ‘clap for carers’ during Covid, or the fact that plenty of people do things like pursue a PhD rather than make the most money they possibly could in the private sector. Trying to yoke value to money so explicitly breaks those down, and gets you to some nasty messages when roles are not deemed to be of ‘equal value’. That, as David Blagden has written, also points to quite a potent Left-flavoured opposition to equal value rules. Do we really want a society where courts are ruling that bankers are more valuable than nurses?

11. Anyone with reliable inside info on the costs here, please do get in touch!

12. My understanding is that Australia, for instance, generally takes this approach. I’m not deep enough in the context there to make any grand pronouncements, though. Any readers who know more, please direct me to where’s best to read more on this.

13. An entire separate post could be written on the ways the EU Pay Transparency directive could import these problems to the rest of Europe. But I’ll leave that topic to someone else!