Raw and adjusted gender wage gaps

Advertisements

Why is the Swedish gender wage gap so stubbornly stable (and high)?

The Swedes are supposed to be in a left-wing utopia. Welfare state, ample childcare and long maternity leave but their gender wage gap is almost as bad as in 1980. They must be a misogynist throwback.

swedish gender wage gap by percentile

Maybe Megan McArdle can explain:

There are countries where more women work than they do here, because of all the mandated leave policies and subsidized childcare — but the U.S. puts more women into management than a place like Sweden, where women work mostly for the government, while the private sector is majority-male.

A Scandinavian acquaintance describes the Nordic policy as paying women to leave the home so they can take care of other peoples’ aged parents and children. This description is not entirely fair, but it’s not entirely unfair, either; a lot of the government jobs involve coordinating social services that women used to provide as homemakers.

The Swedes pay women not to pursue careers. The subsidies from government from mixing motherhood  and work are high. Albrecht et al., (2003) hypothesized that the generous parental leave a major in the glass ceiling in Sweden based on statistical discrimination:

Employers understand that the Swedish parental leave system gives women a strong incentive to participate in the labour force but also encourages them to take long periods of parental leave and to be less flexible with respect to hours once they return to work. Extended absence and lack of flexibility are particularly costly for employers when employees hold top jobs. Employers therefore place relatively few women in fast-track career positions.

Women, even those who would otherwise be strongly career-oriented, understand that their promotion possibilities are limited by employer beliefs and respond rationally by opting for more family-friendly career paths and by fully utilizing their parental leave benefits. The equilibrium is thus one of self-confirming beliefs.

Women may “choose” family-friendly jobs, but choice reflects both preferences and constraints. Our argument is that what is different about Sweden (and the other Scandinavian countries) is the constraints that women face and that these constraints – in the form of employer expectations – are driven in part by the generosity of the parental leave system

Most countries have less generous family subsidies so Claudia Goldin’s usual explanation applies to their falling gender wage gaps

Quite simply the gap exists because hours of work in many occupations are worth more when given at particular moments and when the hours are more continuous. That is, in many occupations earnings have a nonlinear relationship with respect to hours. A flexible schedule comes at a high price, particularly in the corporate, finance and legal worlds.

Gender pay gap shown to be a myth by @paulabennettmp @women_nz

The Minister for Women Paula Bennett and the Ministry of Women published excellent research in February showing there cannot be a gender wage gap driven by unconscious bias. The Minister has blamed a large part of the remaining gender wage gap on unconscious bias.

… up to 84 per cent of the reason for the Pay Gap, that’s right, 84 per cent, is described as ‘unexplained factors.’ That means its bias against women, both conscious and unconscious.

It’s about the attitudes and assumptions of women in the workplace, it’s about employing people who we think will fit in – and when you have a workforce of men, particularly in senior roles then it seems likely you’re going to stick with the status quo – whether they do that intentionally or just because “like attracts like”.

It’s because there is still a belief that women will accept less pay than men – they don’t know their worth and aren’t as good at negotiating.

The reason why this February 2017 research on the motherhood penalty contradicts earlier Ministry of Women research on unconscious bias and the gender wage gap is simple.

There is a large difference in the gender wage gap from mothers and for other women. As the adjacent graphic from Ministry of Women research shows, the gender wage gap for mothers is 17% but it is only 5% from other women.

Source: Effect of motherhood on pay – summary of results Statistics New Zealand and Ministry of Women February 2017.

We men, us dirty dogs all, have no way of knowing whether a female applicant is a mother. Remember we are dealing with unconscious bias, the raised eyebrow, the prolonged pause, the lingering glance, not a conspiracy or a prejudice of which we are self-aware and take overt steps to implement. Unconscious bias is unconscious by definition.

Because the bias against women is implicit and unconscious, we men, dirty dogs all, do not know we are biased, so we do not know we have to make further enquiries to check if the female applicant is a mother so we can discriminate against her more than we do for other women.That is before we consider other drivers of the gender wage gap such as whether there are relatively large spaces between the births of her children. 

Large spaces between the birthdays of children greatly increases the gender wage gap because women spend much more time out of the workforce and part-time if they spread births. This reduces their accumulation of on-the-job human capital and encourages women who plan large families to choose occupations and educational majors that do not depreciate rapidly during career interruptions.

I have no idea how an unconsciously biased employer can discover  if a woman has children with spaced out ages and therefore discriminate against an even more, unconsciously, of course. We men, dirty dogs all, do not know that in order to discriminate against them, especially in shortlisting for initial hiring when we have no information beyond the application about them.

Do women have more unconscious bias against women than men? If not, there should be differences in the gender pay gap in firms with more women managers or owners.

Perhaps there is more unconscious biased in promotions because managers may have accidentally learnt are the ages of the children of  female applicants and unconsciously taken a note to remember that when unconsciously discriminating against them in promotion. This unconscious bias involves a lot of very conscious data collection and retention.

All in all, the unconscious bias hypothesis simply cannot explain such a large difference between the gender wage gaps of parents and non-parents. There is too much evidence whose existence that is strictly forbidden by the hypothesis of unconscious bias against women in the workplace.

A gendered division of labour and household effort

A major factor driving the gendered division of labour and household effort is technology. Tiny differences in comparative advantage such as in child rearing immediately after birth can lead to large differences in specialisation in the market work and in market-related human capital and home production related work and household human capital (Becker 1985, 1993).

These specialisations are reinforced by learning by doing where large differences in market and household human capital emerge despite tiny differences at the outset (Becker 1985, 1993). This gendered division of labour and household effort is hard to change because large payments must be made to influence choices about care giving by highly specialised people with large but different accumulations of market and household human capital.

division of household leisure and shores

From a luck egalitarian perspective, many of the differences in earnings and occupations flow accidents of birth in deciding gender and who parents might be. Social inequalities that flow from brute bad luck call for interventions to put them right, if they work.

Many laws already make up for brute bad luck such as job protections while on maternity leave, and government funded parental leave pay and child care subsidies. Employers can do little to redress these accidents of birth nor do they have sufficient resources to put them right. For this reason, for example, parental leave pay is usually taxpayer funded rather than employer funded.