Governments can't close the gender wage gap

By Mark Wooden

The Australian,
7 March 2008, p. 12

Any raw deal women get in the workplace is connected to an unequal sharing of responsibilities within the home, notes Mark Wooden

EARLIER this week Deputy Prime Minister Julia Gillard, spurred by some trade union report and perhaps the arrival of International Women's Day tomorrow, indicated that the Rudd Government would address the problem of the gender pay gap.

But how serious is this problem and is the Government really in a position to do much about it?

First some facts. Yes, women, on average, earn much less than men. According to the latest Australian Bureau of Statistics data on average weekly earnings, the gap between the sexes is considerable: the average female employee in late 2007 earned about $690 a week compared with $1060 for the average man.

In other words, women, on average, earned just 65 per cent of what men earned, meaning a pay gap of 35 percentage points.

Such comparisons are largely meaningless, however. Most of us would surely expect people with more experience and who put in more hours to be paid more, and in general both of these factors favour men. Most obviously, men tend to work much longer hours than do women. Weekly hours of work among male employees in late 2007 averaged 38.5. By comparison, the average female employee worked just 29 hours. Using hourly earnings rather than weekly earnings reduces the pay gap to 13 percentage points.

Then there is the issue of experience. When economists measure the pay gap, they attempt to take into account those worker and job characteristics that influence pay and hence the pay gap. The portion that is left unexplained is then often assumed to reflect that part of the pay gap that is due to discrimination.

The problem with this approach is that it is impossible to observe and measure all of the factors that influence pay. One factor that invariably is measured poorly is experience.

In particular, most studies are unable to adequately take into account the effect of career interruptions and the number of hours worked in the past. Thus another reason the average woman will earn less than the average man is simply that the average woman has less experience as a result of the tendency for many women to put their careers on hold while their children are growing up and the apparent preference of women for jobs offering fewer hours.

The effect of this is reflected in empirical research that I published in the late 1990s showing that among young people (aged under 30), where employment histories of men and women tend to be similar, the pay gap is zero.

It is among older cohorts, when the careers of men begin to take off and women's career interruptions begin to affect pay structures, that large pay gaps emerge. It also explains why the gaps are more prevalent among skilled, educated workers; experience matters more for skilled work than it does for unskilled work.

The bottom line is that some of the 13 percentage point pay gap will be the result of differences in average experience levels, which are rewarded in the labour market through higher pay.

But isn't there evidence of women's work being undervalued? Yes, past research has demonstrated that part of the reason women earn less than men is that while male and female employees doing similar work are required to be paid the same, work of comparable value in different occupations is not equally remunerated.

That is, female-dominated occupations (say, hairdressers) seem to be less well rewarded than male-dominated occupations (say, motor mechanics). The size of this effect, however, is not large: perhaps about four percentage points.

But just because some occupations seem to be better rewarded than others provides no case for intervening in the market.

Relatively low earnings for hairdressers, for example, may simply reflect competition and the amount consumers are prepared to pay. Attempts to raise the earnings of hairdressers thus may only reduce demand for hairdressing services and hence the employment of hairdressers. In any case, in a world where enterprise bargaining is the main vehicle for setting wages, there is no obvious mechanism for adjusting wages at the occupation level.

So is there then any room for public policy intervention?

The approach traditionally favoured by the union movement is enhancement in the value of the minimum wage. This, however, will have absolutely no effect. As numerous studies have demonstrated, the gender pay gap is much smaller among low-paid workers than among high-paid workers. Indeed, in the most recent research, the gap is negligible among the lowest paid.

Instead, the biggest source of gender pay differentials is among the highest paid private sector employees, what is widely referred to as the glass-ceiling effect. This end of the labour market is not particularly amenable to any sort of regulation or intervention, with pay mostly negotiated on an individual basis.

Ultimately, then, there is probably little government can do to directly influence gender pay relativities, at least in the private sector. Reforms such as the introduction of the right to request part-time work, while improving the access of women to employment, will do nothing to narrow the wage gap; indeed, if anything, it will widen it.

In any case, a focus on employers as the source of pay inequality may be misplaced. If women get a raw deal in the workplace, it is almost certainly because of the unequal sharing of responsibilities within the home.

While women continue to take on the bulk of responsibility for child care and other household tasks they will continue to be disadvantaged in the workplace relative to men.

Mark Wooden is professorial research fellow and deputy director of the Melbourne Institute of Applied Economic and Social Research at the University of Melbourne.