Monday 13 November 2017

Claudia Goldin on the gender earnings gap

Claudia Goldin (Professor of Economics at Harvard University) writes, in the New York Times, on How to Win the Battle of the Sexes Over Pay (Hint: It Isn’t Simple.)

The executive summary
In sum, the gap is mainly the upshot of two separate but related forces: workplaces that pay more per hour to those who work longer and more uncertain hours, and households in which women have assumed disproportionately large responsibilities.
And now a bit more detail.
Yet it is also true that the time demands of many jobs can explain much of the pay difference, a finding that has sobering implications. Eliminating the gender earnings gap will require changes in millions of households and thousands of individual workplaces.
and
The gap is larger among more educated people, for example, and varies according to occupation, often in big ways. Among college graduates, it is far larger in business, finance and legal careers than in science and technology jobs. In health care, it is larger when self-employment is high (think dentists) and much lower when professionals are mainly employees (think pharmacists).

What’s more, the gap is a statistic that changes during the life of a worker. Typically, it’s small when formal education ends and employment begins, and it increases with age. More to the point, it increases when women marry and when they begin bearing children.
and
Similar patterns appear using data for women and men who have earned master’s degrees in business administration. Immediately after graduation, women earn 92 cents for each male dollar. A decade later they earn only 57 cents.

Correcting for time off and hours of work reduces the difference in the earnings between men and women but doesn’t eliminate it.

On the face of it, that looks like proof of disparate treatment. It may seem understandable that when a man works more hours than a woman, he earns more. But why should his compensation per hour be greater, given the same qualifications? But once again, the problem isn’t simple.
and
The data shows that women disproportionately seek jobs — including full-time jobs — that are more likely to mesh with family responsibilities, which, for the most part, are still greater for women than for men. So, the research shows, women tend to prefer jobs that offer flexibility: the ability to shift hours of work and rearrange shifts to accommodate emergencies at home.

Such jobs tend to be more predictable, with fewer on-call hours and less exposure to weekend and evening obligations. These advantages have a negative consequence: lower earnings per hour, even when the number of hours worked is the same.

Is that unfair? Maybe. But it isn’t always an open-and-shut case. Companies point out that flexibility is often expensive — more so in some jobs than others.

Certain job characteristics have a big impact on the gender earnings gap. I have looked closely at these issues, including the extent to which workers are:
  • Subject to strict deadlines and time pressure
  • Expected to be in direct contact with other workers or clients
  • Instructed to develop cooperative working relationships
  • Assigned to work on highly specific projects
  • Unable to independently determine their tasks and goals
Occupations with a lower level of these characteristics (like jobs in science and technology) show smaller gaps, corrected for hours of work. Occupations with a higher level (like those in finance and law) have greater gaps. Men’s earnings tend to surge when there are fewer substitutes for a given worker, when the job must be done in teams and when clients demand specific lawyers, accountants, consultants and financial advisers. Such differences can account for about half the gender earnings gap.
Ask yourself, do you really care who your pharmacist is versus do you care who your doctor or lawyer is? A particular pharmacist not having to be there to deal with customers mean greater flexibility in hours worked but this comes with lower pay while the fact that people want a particular doctor or lawyer to deal with them means long hours with little flexibility but with higher pay to compensate.
These findings provide more nuance in explaining why the gap widens with age and why it is greater for women with children. Whatever changes have already taken place in American society, the duty of caring for children — and for other family members — still weighs more heavily on women. And if you thought that moving to a more family-friendly nation would eliminate the gap, think again. In several nations, including Sweden and Denmark, a “motherhood penalty” in earnings exists, even though these nations have generous family policies, including paid family leave and subsidized child care.

Such considerations bring us to a very sensitive area: domestic arrangements at home, especially among couples with children. These are personal questions. In theory, gender earnings equality is possible when both parents take off the same amount of time and enjoy the same flexibility at work.
So domestic arrangement with a more equal distribution of childcare may reduce the wage gap but it may also make the family poorer.
From a classic economic standpoint, if one spouse or partner can earn more by working less flexible hours, as a family, the couple would earn more money by having that parent in that job, while the other partner accepts the more flexible one. A man can certainly be the more flexible member of this household — though he typically is not. Such decisions need to be made couple by couple.
So the answer to the pay gap may be in the choices made within the home. And that makes it difficult for public policy to deal with.

4 comments:

Tim Worstall said...

We can approach this from the other end. And I'd just love to see a proper paper on it too.

Fathers earn more than non-fathers after adjusting for other characteristics like age, education etc. Some say 6-8%.

Mothers earn less than non-mothers, 9 % (in one estimation) for first child and smaller greater amounts for subsequent.

Mix and match against proportions of each in the workforce (more mothers than fathers) and we've pretty much explained all of the observable gender pay gap.

Thus, what we've actually got is a primary child carer gap, something that as Goldin says is dependent upon a family by family decision about how to organise domestic life.

Paul Walker said...

Tim

The father non-father thing could just be that wanting to be a father correlates with some other (difficult to measure, maybe social) high wage earning factors.

Also lesbians earn more than men (approx 9%) so another answer to the gap is for all women to become lesbians!!!

Jim Rose said...

Tim, key part of the Goldin research is even tiny absences from high powered jobs can lead to large pay gaps.

Many jobs cannot be restructured such as in 24/7 industries or the lawyer or doctor or cop you want now, not five minutes from now.

Tim Worstall said...

"Also lesbians earn more than men (approx 9%) so another answer to the gap is for all women to become lesbians!!!"

Possibly lesbians are less likely to be mothers. Gay men earn less than hetero too.......