News outlets regularly report evidence of ‘gender pay gaps’ or figures showing women or people of certain ethnic origins are ‘under-represented’ in certain industries. Earlier this week, it was reported that women in the City would receive much lower bonuses than men, for example. The article implied this resulted in discrimination by employers.

There is a high possibility that in some City workplaces, discrimination exists. But the statistics presented tell us nothing about this. The surveys asked women and men how much they thought their bonuses would be. There was no control for job type (including whether the role was heavily commission-based), company, whether respondents were part or full-time workers, their age, whether they had had time out of the workforce and their working arrangements.*

Not controlling for these factors means we have no idea whether a genuine gap exists due to ‘discrimination’. In fact, some of the accompanying text acknowledges other potential explanations: women are more likely to undertake childcare responsibilities and more likely to not take up trading roles where bonuses are likely to be larger, for example.

In principle then, it is quite possible that these bonus differences largely exist because of purely free choices. Yet as is common with the write-ups of these reports, even differences in career or childcare choices are talked about as if they are evidence of broader ‘problems’ in society. Women are ‘turned off’ certain career choices or are put at ‘a disadvantage’ for caring for children, according to the press release. For many (see for example Polly Toynbee), the starting point seems to be that the existence of any sort of numerical ‘gender gaps’ in employment or pay are either company or societal problems which must be addressed through government policy.

There’s little doubt that social norms and existing policies shape working practices, which can in turn shape outcomes. One can imagine, for example, that certain industries traditionally the preserve of men might have hours which women, on average, might find less attractive. We know too that men tend to be more likely to be employed in riskier jobs. If some of these norms were to change, or policy slants altered – for example, paternal leave being given equal status to materal, then the ‘gaps’ may indeed close somewhat. Choices do not occur in a vacuum, after all. Equally, there may be other policies, such as the discrimination in the tax and benefit system against single-earner households that if reversed might lead to more women making the decision not to work, increasing the employment gap.

But the problem with focusing on the outcomes or gaps as a starting point is that there is no reason why we’d expect, even in a world where there was complete equality of opportunity and no discrimination at all, that all sexes, genders and other groups would be represented in sectors in the same proportions as in the population more widely.

For a start, free choice simply does not lead to outcomes that conform to some set idea of what society should look like. And it is clear, from mountains and mountains of evidence around the world and through history, that certain groups seem to thrive more in certain industries, tasks or pursuits. As Thomas Sowell has made a career of highlighting: ‘Even in activities where individual performances are what determine outcomes, and those performances are easily measured objectively, there is seldom anything resembling equal representation.’

We accept this in some areas, such as sports, without moaning about underrepresentation. Kenyans seem to be overrepresented as top marathon runners relative to athletes from many other countries. Black players tend to be overrepresented in professional basketball. White swimmers tend to dominate Olympic events. Children of Indian descent seem to always win American spelling bees. Yes, there may be some social factors which influences who decides to pursue these activities, but few would argue that these outcomes are the function solely or even largely of any inherent bias or discrimination against others. Opportunities largely exist for all groups and these outcomes are regarded as meritocratic.

Yet it seems taboo to suggest that, particularly in the case of gender, there may be some systematic differences in the sorts of careers that men and women might decide to pursue which might result in different employment or remuneration. No, in many sectors (usually the very elite levels) it is simply asserted that any employment or wage gaps that arise as outcomes must be reflective of either unequal opportunities which adversely shape choices or else of overt discrimination, rather than being a manifestation of free choice.

The problem with this ‘underrepresentation’ framework then is that it creates an unattainable goal (absolute equality or equal representation) which legitimises a high degree of social engineering through government policy. Rather than just focusing on equality under the law or even opportunities, the examination of ‘gaps’ legitimises the pursuit of equality of outcome, irrespective of why outcomes arise. Whilst improving opportunities for women, for example, in one field may happen to close some employment or wage gap, taking the gap itself as the problem implies too that even free choices should be curtailed if they do not result in outcomes regarded as ‘right’ according to some preconceived idea of just outcomes.

This really is why the strand of feminism concerned with these outcomes is largely the preserve of the left. Like broader egalitarianism it observes the world and assumes something is fundamentally wrong with it if outcomes do not conform to absolute equality.

But in recent years non-egalitarian voices too have given these voices a free-pass, perhaps because highlighting your concern about ‘gender pay gaps’ or underrepresentation is increasingly a good way to ‘virtue signal’, and male attempts to contribute to the gender debate in particular are often shouted down (indeed, I have twice been described to be ‘mansplaining’ when discussing this issue. See also the ‘check your privilege’ meme). Women who resist the egalitarian call to arms are treated as traitors to their gender (see many feminists’ reaction to the success of Margaret Thatcher). Even non-egalitarian voices – not least David Cameron – now (wrongly) use outcomes as proxies for opportunities.

The results of this are more significant than simply inconsistencies not being exposed (see, for example: ‘we should have quotas of women on boards because women bring different qualities’, and ‘no pay gaps should exist because there’s nothing inherently different between the sexes’). No, the result of this equality of outcomes worldview going unchallenged is that the political window of what it is acceptable to say in this area is narrowing all the time. Evidence that shows the current generation of women (who, for the first time have had similar higher educational opportunities as men) actually earn more than men, on average, up to the age of 40, is hidden by the use of ‘average stats’ with the conclusion ‘more must be done’, for example. And most news stories contain little in the way of dissenting voices against the idea that gender pay or employment gaps are huge problems, even though there is a broad economic literature that has explanations for pay and employment gaps other than discrimination.

It’s high time for the media to be more analytical about these issues, and for more evidence and logical thinking to be injected into the debate. At the moment we are heading towards a world where inequality as ‘difference’ is regarded as a problem. The implications for a free society of this ideological worldview are very worrying indeed.

Ryan Bourne is the IEA’s Head of Public Policy. Recommended reading: ‘Should We Mind The Gap?’ by Len Shackleton.