SPEAKER 1: As we've seen, the increase in the gender pay ratio was particularly pronounced during the 1980s. Can the tools of economic analysis that we've developed so far help us to understand this development? To answer this question, I'll turn to the results of the study that I did with my Cornell colleague Larry Kahn. I've already shown you some of our results in the previous chart that looked at the pay gap at a point in time.
Now I present our analysis of the sources of the decrease in the gender pay gap over the 1980s. Our data is from a nationally representative data set, the panel study of income dynamics. And it covered the period 1979 to 1988. During that time in our data, the ratio of wages of women compared to the wages of men, that's the gender wage ratio, increased from 62% to 72%. That's an increase of 10 percentage points or 16%. And let me just review that.
If you look over at your right, again, the gender ratio increased by 10 percentage points. 72% minus 62% equals 10 percentage points. This represents a 16% increase in the ratio. 10 divided by 62 two equals 0.16 or 16%. We divided this 16% increase in the gender ratio into the various possible explanatory factors and estimated the contribution each one made. Our findings are shown in the chart at the right.
The first factor we consider is the change in gender differences in characteristics. To see what this means, let's consider some examples. Women, on average, have less experience than men. And this is one of the reasons for the gender pay gap. Now suppose that the experience gap between men and women narrows as women choose to stay in the labor force more continuously.
Well, that will, in turn, reduce the pay gap. Declining differences in the educational attainment of men and women, declining differences in the occupations of men and women all potentially contribute to the convergence in men's and women's wages.
In my research with Khan, we estimated the quantitative effect of each of these changes. What we found, again, is shown in the chart at the right. All the changes and characteristics that we considered taken together would have increased the ratio by about 13%. Most important of the characteristics is labor force experience. This factor accounts for a 6% increase in the gender wage ratio. The gap in full time work experience between men and women declined from 7 and 1/2 years, on average, in 1979 to 4 and 1/2 years in 1988 as women became more firmly attached to the labor force.
Next in importance is occupation. Women moved into managerial and professional jobs where they had not been as well represented before. I should point out that we use very broad occupational categories in this analysis. If we had used more detailed ones, at say the level of elementary school teacher, secretary, et cetera, occupational changes might have explained a bit more of the convergence. But still, the impact of occupational changes is impressive, working to raise the ratio by 5%.
Changes in the industries of men and women were not particularly important in influencing the overall trends. But changes in union membership were, but perhaps not in the way you might have expected. An increase in the ratio of about 2% was due to a decline in the difference between men and women in the probability of being unionized. But in fact, unionization was declining for both men and women. So what this means is that men were losing union jobs at a faster pace than were women.
Taking these results together indicates that changes in the measured characteristics of men and women can go a long way in explaining the increase in the gender ratio, accounting for a 13% increase out of a total increase of 16% But working in the other direction, and here I'm getting at the swimming upstream idea, is the returns to these characteristics. The reward to experience was increasing. And on average, women have less experience. The returns to being in predominately male occupations and industries were increasing, also working against women.
Our research indicates that changes in the returns to characteristics, taken separately, would have actually resulted in an 11% decrease in the gender ratio. So on the one hand, women's improvements and characteristics raise the ratio. But on the other hand, the rewards that the market offers men and women for these characteristics, and the tendency of women to have less experience and to be located in different sectors of the economy, move the ratio in the opposite direction.
One more factor on the positive side of the ledger, that is one more factor benefiting women during this period, was that the unexplained pay gap between men and women decreased. And this factor accounts for a possible 14% increase in the gender ratio. So we have two factors that, taken together, would have increased the ratio by 27%. That is improvements in women's characteristics and a decline in the gender pay gap. But going in the opposite direction was the change in the wage structure that worked against women. Women successfully swam against the current of rising inequality by improving their characteristics and narrowing the unexplained pay gap.
What does the unexplained pay gap really represent? I don't know for certain. That's why I have to call it unexplained. But I can make some reasoned speculations about what may underlie the decline in the unexplained gap that occurred in the 1980s.
One possibility is that discrimination against women decreased. I think that's very plausible. Since I'm looking at changes that have occurred in my own lifetime, my own observations, as well as my knowledge of the literature and my research results, are important in figuring out what's going on. And my own observations support the idea that discrimination declined. Taking a longer period than the 1980s, there's probably little argument that overt, blatant discrimination against women has decreased.
Even over the 1980s, there were probably some additional decreases in discrimination, particularly what economists call statistical discrimination, that is the idea of the statistical discrimination is that employers may discriminate against women based on their perceptions of the average characteristics of the group. So for example, if employers believe that women workers are, on average, more likely to quit their jobs than men, they might be reluctant to hire women or might pay them less than similarly qualified men. As women have become more attached to the labor force, the rationale for this type of discrimination has diminished.
A second thing this decline in the unexplained portion of the pay gap could represent is women's improving their qualifications in ways that we can't readily measure. They might be getting more committed to the labor force or going the extra mile. And they could be doing these things because, increasingly, they see themselves as interested in and wanting careers.
Finally, another factor that professor Kahn and I have speculated as being important relates to those demand shifts that I mentioned before. These demand shifts may have effected men and women differently. Kahn and I call this a gender twist in demand.
The gender twist in demand is not a new dance craze. Basically, the idea is that the demand shifts that have hurt less skilled workers in general have probably hurt men more than women. The reason is that, traditionally, men have been concentrated, or at least have been more concentrated than women, in blue collar manufacturing jobs. And these are the very same jobs that were especially hard hit during the 1980s as we shifted even more towards a service and computer-oriented economy.
Women, for whatever reason, have always tended to be more concentrated in white collar occupations, not the highest level ones but clerical support, secretarial jobs, things like that. Also, interestingly, if you survey computer use, women are more likely to use computers on the job than men. So maybe a factor benefiting women during this period and helping to account for the decrease in the unexplained gap is that women have actually been doing a little better in terms of the shifts in demand that we have seen in the economy and the labor market.
OK. So as we've seen, there was less progress in narrowing the gender pay gap after this period in the 1990s. How do we explain this?
We've received your request
You will be notified by email when the transcript and captions are available. The process may take up to 5 business days. Please contact email@example.com if you have any questions about this request.
Francine Blau, the Frances Perkins Professor of Industrial and Labor Relations and Labor Economics at Cornell ILR School, describes trends in the gender pay gap, considers fundamental explanations for the gender pay gap and uses these explanations to understand the trends.
This video is part 6 of 8 in The Gender Pay Gap series.