The gender pay gap that isn’t being discussed?

Stop me if we’ve been through all this before

Yes, its yet another article on the subject of the ‘gender pay gap’ on Comment is Free, this time marking the release a the TUC’s report predictably entitled ‘Closing the Gender Pay Gap‘ which is published on the eve of… guess what?

Yep, its the TUC Women’s Conference and this all comes down to nothing more than the usual ‘by the numbers’ exercise in which the existence of a gender pay and its cause – discrimination, naturally – is put forward as the unquestioned ‘given’ from which all else proceeds even though it blatantly obvious to anyone with the slightest inclination towards sceptical inquiry that there’s rather more to all this than its proponents seem either willing or capable of admitting to.

And so it is that we’re told by the TUC – the article on CiF is so run of the mill that its not even worth commenting on the specifics – that:

Several causes are cited for the gender pay gap, including the concentration of women in low-paid jobs such as childcare and cleaning, the undervaluing of women’s skills and the employment penalty for mothers. This ‘motherhood penalty’ partly explains why the gender pay gap increases so rapidly for women in their 30s.

So – as usual – a large part of this pay gap comes down to nothing more than a matter of women working fewer hours than men and having a shorter working life due to their their valuable ‘earning time’ getting unfairly eaten into by the little matter of motherhood and all things proceeding from it.

And – also as usual – this is all backed up by a table showing ‘mean gender pay gaps for different age groups’ which looks like this:

Age Full-time pay for men Full-time pay for women Part-time pay for women Full-time gender pay gap Part-time gender pay gap
16-17 £4.75 £5.21 £5.14 -9.7% -8.2%
18-21 £7.28 £6.96 £6.96 4.4% 11.54%
22-29 £11.08 £10.72 £8.49 3.3% 23.4%
30-39 £15.64 £13.89 £10.70 11.2% 31.6%
40-49 £17.35 £13.39 £10.21 22.8% 41.2%
50-59 £16.22 £12.88 £9.89 20.6% 39.0%
60+ £13.36 £11.45 £8.90 14.3% 33.4%

At this point I’m struggling to find the right onomatopoeic combination of letters to express the deep, guttural groan tables like this elicit – the TUC may not be capable of adequately interpreting and critiquing statistical data but I am, and so are plenty of other people and the flaws in this table are so stupidly obvious that you wonder just what of kind of contempt the authors of this report have the intelligence of the general public.

This isn’t difficult – yes there’s a difference between the mean full time pay for men and the mean full-time pay for women and this proves women are discriminated against in employment… yes?

No, it doesn’t because we know that general employment patterns amongst men and women in terms of thing like proportions working is different sectors of the economy, different industries, etc. are markedly different and that these different sectors of the economy and different industries (and different employers, of course) offer different rates of pay, pretty much all of which is negotiated via an open labour market.

To adapt a fruit-based analogy, the equation we’re being expected to swallow come in the form:

Apple is different to Orange therefore discrimination. Different. Not equal but also not unequal because we’re being to draw conclusions based on an invalid comparision.

For all the explanatory value of any this you might as well be asking the old nonsense question:

If it takes a week to walk for a fortnight then how bananas are there in a bunch of grapes?

However the real ‘elephant in the room’ here is this statement, which comes from a factsheet (pdf) produced by the government’s own Women and Equality Unit:

The part-time gender pay gap

The part-time gender pay gap is based on the hourly wage of men working full-time and women working part-time, which is defined as being less than 30 hours a week.

Sorry? It’s based on comparing what with what?

Do I really need to explain what’s wrong with particular proposition?

In fact, a little bit of digging around for data on male part time earnings, which would be the valid comparator to use, did turn up this, which appears as a small print side note in this the introduction to this government report on the ‘part-time pay gap

It should be noted, although we do not analyse it, that there is also a large part-time pay penalty for men – the New Earnings Survey suggests that in 2003 part-time men had average hourly earnings that were 32% lower than the average hourly earnings of full-time men

So in 2003, men working part time had average hourly earnings that were 32% lower than men working full time – that’s across men of all age groups so to compare things properly we need to generate the same general average for women working part-time by calculating averaging the age group differentials…

…all of which comes to an average pay gap between male full time hourly earnings and female part-time hourly earnings of…

27.9%

Mmm… Houston, we seem to a have a problem here – we may not be using same year data for our calculations – one would expect the TUC to be using data from 2006-7 so the comparison is not exactly a like for like one, but then what this new report is also saying is that while the full time pay gap is closing over time (down from 17% to 12% over the last ten years) the part time gap, as derived from the government’s own dodgy calculation method, hasn’t really moved that much at all in recent years, which does seem to suggest that there may actually be a gender pay gap in part-time work of about 4-5% in favour of women.

Now get me wrong here, I’ve no real problem with tackling any elements of the gender pay gap that are non-structural, i.e. things like gender discrimination which cannot be reasonably be ascribed to the workins of an open and fair labour market and I’ve no real problem with anyone coming at this from the equality of outcome angle by advancing a case for tackling the structural gap using, say, a Dworkinian insurance scheme – as long as the numbers add and the argument makes sense, I’m happy to consider anything.

But what doesn’t fly with me is the misrepresentation of evidence – seemingly by omission – which appears to be taking place here.

Equality, whether of opportunity or outcome – yes. Dishonesty – no.

Also at Liberal Conspiracy.

  • I’ve been banging on about this for some years now.

    http://eqsq.com/vivreLaDifference/the-gender-pay-gap-in-the-uk.html

    Includes the links to the raw numbers. One of the interesting ones is to look at previous years (not sure they break the figures out this way any more) on the difference btween public and private wages.

    Turns out that public are higher than private by about the same amount that male are higher than female. But when you mention that you get all sorts of, oh, but people are doing different jobs, are more qualified, so that doesn’t count. Strangely, you’re not allowed to use e same arguments on hte gender pay gap.

  • I know you have, Tim.

    It’s in no small measure down to you and Chris Dillow that I started to look into the numbers to set what the actual data was saying.

    The other article I posted at LibCon looked, amongst other things, at how the public/private split might work into this as a factor. I’m aware that, notionally, public sector salaries have outstripped the private sector in recent years but when factoring gender into the equation in the private sector things become more complicated because the male/female balance looks like it splits unevenly towards men when looking at employment sectors with a higher risk/reward element – jobs in which commission and performance bonuses make up a significant component of earnings.

    When you adjust the raw numbers to remove as many of the wild card elements as possible it does work out that public/voluntary sector employment is less advantageous in terms of pay but offers considerably more job security and better terms on flexible working, maternity/leave arrangements, If you’re risk averse and prefer to play safe then the public sector is where its at and its no coincidence, I think, that public sector employment is solidly skewed towards women at all but the highest grades.

    Part of the pay gap here looks very much to be a reflection to women being more likely to trade off better pay and prospects for advancement with less security and poorer non-cash terms for lower pay, more security and better terms, particularly where those terms would be advantageous to a woman who expects, at some point in the future, to have kids.

    It’s a complicated picture and there is still an element of discrimination at work, but risk aversion and how this splits across genders is another significant factor, although one amongst many.

    The key thing for me, in general terms, is going to be teasing out which elements of the gap are structural and which are non-structural because that will necessarily set the limits of what can and cannot reasonably be tackled.

  • Shoz Rahman

    Many studies reporting a pay gap are very unscientific and do not make valid, matched comparisons. Rarely are they subjected to statistical analyses to determine the significance of findings. Worse, when an apparent gap is discovered, blame is immediately apportioned on discrimination. That

  • Pingback: If we are going to discriminate, let’s discriminate for the right reasons « UK Liberty()