BRITAIN has one of the widest gender pay gaps in Europe. For every pound that men earn, women make 80p, and the disparity has moved little in 15 years. Vowing to end this “scandal” within a generation, in 2015 the then prime minister, David Cameron, pushed through a policy long resisted by businesses. Organisations with 250 or more employees would have to publish the gap in hourly pay between men and women. April 4th 2018 was the deadline for the first wave of this annual exercise. The results aren’t pretty.

The 10,000 employers that filed results revealed an average median pay gap of 12%. The Economist Group, our parent company, reported a gap of 29.5%, more than double the average for the media industry*. Some, including Ryanair and Jefferies investment bank, admitted that they paid women less than half what they paid men. Firms also published the share of men and women in each income quartile, with most proving top-heavy with men (see chart).

The data do not adjust for employees’ different roles, so chief executives are compared directly with secretaries. Mean gaps can be skewed by a few high-earners, as shown by the fact that nine of the ten organisations with the greatest differences between their median and mean pay-gaps are football clubs.

All this leaves plenty of room for spin, misinterpretation and counterproductive responses. Two narratives have emerged. The first is that the gaps prove how sexist and discriminatory the workplace still is. The second is that they are adequately explained by men’s greater share of senior jobs, and have nothing to do with discrimination. Neither is quite right.

Equal pay, unequal work

One misperception is that a gender pay gap reveals pay discrimination. It doesn’t. Pay discrimination refers to people who do the same job, or jobs of similar value, being paid differently. It has been illegal in Britain since 1975 and underpins several big complaints, such as an ongoing £4bn ($5.6bn) claim against Tesco by female shop staff claiming that they should have been paid the same as mainly male warehouse workers (Tesco denies wrongdoing).

The new reporting exercise is not about such discrimination. Airlines have reported some of the widest pay gaps, yet are unlikely to have unequal pay, because collective bargaining tends to determine salaries in the industry. Of easyJet’s pilots, 94% are men (average salary: £92,400) whereas 69% of cabin crew are women (earning an average of £24,800). The airline industry does not have a pay-discrimination problem so much as a recruitment problem.

Other industries fail to promote enough women—and Britain is not alone in that. McKinsey, a consultancy, found that, whereas half of graduate entrants in American law firms were women, only one in five equity partners was. Such “leaking pipelines”, where the share of women drops at every level of promotion, are also clear in financial services, which has the second-widest pay gap of any industry in Britain. A study by SKEMA Business School found that, although women made up 52% of banking employees globally, only 38% of middle managers and 16% of executive committee members were women.

The pay-reporting exercise comes with limited enforcement. Some firms large enough to report have not done so, and several published what looked like implausible pay gaps (a cluster around 0% suggests that some are fibbing or failing to understand the exercise).

But others have accompanied their data with explanations and plans to fix things. RBS, a bank, vowed to achieve gender balance across all ranks by 2030. The Telegraph newspaper promised to close its 23.4% gap by 2025. Because this will be a yearly exercise, many firms’ priority will be to ensure the figures for 2019 show improvement.

They should beware quick fixes. Companies could outsource low-paid administrative work, which would improve their figures overnight. They could also stop hiring junior women, who exacerbate gaps in the short term. Voluntary pay-cuts by the best-paid men—as taken recently by some BBC stars—may be good PR, but they can create backlashes, as in Iceland a few years ago when some men’s salaries were cut after equal-pay claims by women.

Many of the more sensible solutions take time and start with diagnosing what lies behind the numbers. “I’m concerned that this British reporting is ostensibly focused on pay, but in reality it’s just about representation,” says Brian Levine of Mercer, a consultancy. “What companies really need to find out is whether they are hiring equitably, paying equitably and are offering equitable opportunities to advance.”

Men’s and women’s salaries start diverging from the childbearing years. This “motherhood penalty” is often followed by the “good-daughter penalty”, when elderly parents require care and daughters prove more conscientious than sons. The median pay gap is only 2% among full-time workers in their 30s, yet jumps to 14% in their 40s and 16% in their 50s, according to the Office for National Statistics.

Tempting as it may be to blame women’s lack of progression on their wombs, this is only part of the explanation. The other is structural discrimination. One in nine new mothers is dismissed, made redundant or treated so poorly that she leaves, according to Britain’s equality watchdog. Subtler biases favour men in hiring, performance reviews, pay and promotions. A study in 2016 by Warwick University found that, among workers who asked for pay rises, men were 25% more likely than women to get the nod.

Unlike Britain, many European countries tackling pay gaps have focused on discrimination between people with similar jobs, rather than gaps across whole companies. Nevertheless, Britain’s blunter exercise is having an impact. The data have got board members, shareholders, customers and employees talking about pay. Now that the numbers are out, executives are keen to “win”, says one consultant. Iris Bohnet, an economist at Harvard University, has noted a similar effect caused by awarding hygiene ratings to restaurants and putting energy-efficiency labels on white goods.

If women’s positions do improve, it will be hard to say whether it was caused by mandatory reporting or broader winds of change. In America, where President Donald Trump has rejected a similar policy, companies nevertheless increasingly publish data on pay equality, often under pressure from shareholders. In Britain several accounting and law firms disclosed partners’ salaries, though they were excluded from mandatory reporting.

And within companies the case for diversity is increasingly made not in terms of PR but profit, as evidence linking diversity to performance mounts. McKinsey found that companies in the bottom quartile for gender and ethnic diversity in leadership were 29% less likely to achieve above-average profitability.

Even if every company became scrupulously fair, the pay gap would endure as long as more women than men worked part-time and in industries that pay poorly. This prompts a question that is often overlooked: what size should the gap be? Theresa May this week said she wanted her government to end the “burning injustice” of the gender pay gap. The injustice to aim at is not the lack of equality in outcomes, but rather in opportunities.

* A full breakdown of our figures can be found at