The median gender pay gap is not a very useful measurement
Bangladesh has a low gender pay gap -- an interesting and useful thing to know. But it’s not the same, not the same at all, like the statement, that there is no discrimination about wages with respect to gender.
For the opposite is also not true -- places which have higher gender pay gaps also contain distinct and direct pay discrimination on the grounds of gender. This is a point that isn’t generally appreciated yet; it’s one that is absolutely vital to grasp to be able to make sense of matters.
Take, for example, Britain -- the median gender pay gap for hourly pay is around 20.6%. This number is virtually useless as a guide to whether there is discrimination in the setting of pay.
Quite apart from anything else, it is illegal, to vary pay for the same job purely on the grounds of gender. It’s also entirely true that people doing the same job in the same firms do not show a gender pay gap.
One reason for there appearing to be a gap is that we’re not -- by using the median of the total population -- comparing apples to apples. In Britain, those working part-time make less per hour than those working full time.
That’s just how the labour market works. There are many more women working part-time than men. So, if we add up all the hourly wages we’ll find that we cooked into our numbers a gender component to the pay gap that really just doesn’t exist. There are other factors like this -- education levels for example -- which make the total population number not all that useful.
Once we adjust for those other factors, we get to the factor adjusted pay gap. This is where Bangladesh leads the world, having one which is uniquely in favour of women. Sadly, this is also not all that useful a measurement.
The reason why Bangladesh does so well is because of the industrial sector, which is driving the economy. We all know very well that the RMG sector is the major exporter from the country.
It’s also true that export sectors tend to pay higher wages than non-export parts of the economy. Finally, we all also know that the RMG industry is largely staffed by women. Put those three together and we have our higher paying export sector largely employing women.
This is such a large part of the economy that it drives up the median wage for women in the country. Just as the low portion of men employed in the same industry drags down male wages. And yes, this effect really is large enough to explain our total gender pay gap figure.
We would hope that an international report on the subject would explain this but sadly it doesn’t. We have to walk through the implications of this ourselves.
Is there anyone at all willing to claim that there is no discrimination in pay rates in Bangladesh? Good, no, we didn’t think so, did we? And yet we do have this figure which, by the normal international measures, tells us that the very mild discrimination which does exist is in favour of women. This isn’t what we see around us in daily life now, is it?
Yes, we should indeed make that next logical leap. If we know, very well, that the Bangladeshi absence of a gender pay gap is a result not of the absence of discrimination, but just a quirk of the female employment ratio in the dominant export industry, then what do the recorded gender pay gaps of other countries tell us?
There are many countries, like Britain, which have substantial gender pay gaps by the usual measures. But which also have strict, and obeyed laws against gender-based variations in pay for the same job.
The answer is that the existence of these gaps isn’t proving discrimination, just as the absence of such isn’t proving the absence of the discrimination. For the gaps that we can see by this usual measure are coming from much the same cause. There’s a certain amount of gender segregation in the occupational sector.
Men and women aren’t paid different amounts to be, say, nurses. But there are many more female nurses than male. Just as there are many more male miners than female -- or used to be before all the mines closed. And if mining pays more than nursing, which it used to, then our countrywide figures will show a gender pay gap without there being discrimination or even, really, a gender gap at all.
Therefore, there’s a problem at the heart of these numbers. It is taken as proof that if, on average, men and women get paid different amounts, then this must be because of discrimination by employers or society. But that’s not necessarily the case. If men and women make different choices -- which job to do, part or full time, and so on -- then they’ll earn different amounts.
The median gender pay gap, or even its absence, isn’t telling us what near all assume it does. Thus, it’s not a good guide to public policy.
Tim Worstall is a Senior Fellow at the Adam Smith Institute in London.