Top critical review
24 people found this helpful
Faulty Assumptions and Data
on April 27, 2006
Evelyn Murphy uses personal stories and anecdotes to build a plausible case of discrimination around her core "fact". I could have matched every story she has collected with a similar story by a male. Personal stories are for human interest, not for serious arguments. Murphy needs to employ a reputable economist before she writes a book whose entire premise rests on any comparisons of "full-time workers". She claims this gives a comparison of "apples and apples". But it does not. The fragility of her cornerstone is most true for simple comparisons of pay averages for men and women. What are we comparing when we compare pay averages of full-time workers?
The definition of "full-time" includes all those who regularly work thirty-five hours a week or more for a simple majority of weeks during the measured year. "Full-time" work lumps together those who usually work thirty-five hours per week with those who regularly work 50-60 hour weeks and some who might be working 100 hour weeks. It also classifies those who worked 27 weeks with those who worked 52 weeks. That is, the measure is very muddy. Hours worked per year affect pay directly for those payed on a hourly wage and for overtime. Hours worked can also affect pay indirectly in a number of ways, including changing the likelihood of pay increases, bonuses and promotions. Women who work 35-39 hours a week earn 109% more than males with similar hours but only 64% of males who work 40-44 hours per week, but they are both full -time workers. Is the full-time comparison fair?
Males work 128% of female hours and are paid 130% of the median income. Females work 78% of male hours and are paid 74% of the median income. This statistic, naked as it is, indicates a scant 1.96% pay difference (women earn about 98% of men's earnings). Why not use this as the "real" pay gap?
Further, when comparing full time workers as a group without rigorously controlling for continuous years in the workforce, tenure in the current job, employment sector, etc. your comparisons will be like comparing apples with buffalo. Murphy pays no heed to the remaining considerable differences in human capital and hours worked per year between male and female full time workers.
According to PSID data, as examples of relevant male/female differences, women: are 8% less likely to be in the workforce; are13% less likely to have "labor force participation" (workforce activity is relevant to collective human capital and, hence, to average pay), work 78% of male hours, are 37% less likely to work full time, have 190% greater absenteeism, have 4.3 years less work experience and 15.2 fewer months in the same job, accrue 308% more time out of the labor force (in weeks) and have similar educational levels.
Are there male and female differences? Obviously. Signs of discrimination, no. Some very strong evidence against discrimination shows up in a GAO report (GAO study "Women's Earnings: Work Patterns Partially Explain Difference between Men's and Women's Earnings", 2003). Women get about a 31% advantage in pay over men for each hour worked and they get a nearly 30% greater return for each year of education (GAO Report, Table 4, pp. 46-47). These objective advantages point to a workplace recognition of the value of women's work and of their potential as workers. Despite Murphy's contrary protestations, and according to the GAO study, men still have a 42.5% advantage in individual human capital.
Murphy prefers victimhood over facts. The explanation of discrimination over male and female differences in expectation and preference. She does not mention so many things that are relevant. One is that marriage is, for the most part, an individual agreement to reproduce the gender pay gap and the presence of children seals the deal. Polls show that both men and women see this arrangement as appropriate.
If we call marriage a "culprit" in the gender pay gap, what about comparisons of the never-married? The Bureau of the Census shows that for the never-married, as far back as the 1950s, there was an overall two-percent pay gap favoring men but a six-percent pay gap favoring women - among white workers forty-five to fifty-four years of age. Back in the old days of the 1960s - before anti-discrimination laws - among unmarried women and men who had continuous labor force experience, women earned more. The pay gap was slightly in their favor! For both those full-time workers who never had children and for those currently with "no children present", never married women average from 15.5% - 17.5% more in earnings than never married men. For unmarried women and men who work fifty or more hours a week the average earnings are identical.
And then there is this startling fact: between 1983 and the present (beginning in 1999) measures of women's average job prestige, job status and job desirability has come to surpass those same measures for male workers.
In the U.S., we find that northern pay is higher than southern pay (77%). We do not blame discrimination for the regional pay gap. We discuss differences in economic structure and the history of unionization. The gender pay gap varies from city to city (sometimes favoring women!) and state by state (with as much as 23 cents-on-the-dollar variation). The proffered explanation: economic structure. When we discover that Asian men in the U.S. have higher average incomes than their white counterparts, we do not say that white males face discrimination. If Mexican American women earn more than males of the same heritage, what kind of discrimination is that? How about the fact that divorced women average higher earnings than divorced men? Do women discriminate against lower earning men, and so divorce them? (Yes.)
This discrimination creates a large pool of low earning, never-married males and the phenomenon of hypergamy (the strong tendency of women to marry "up"). Male income is very much related to the probability of both marriage and divorce. This is not the case for females. Nearly all U.S. CEOs have wives who have never worked for pay.
Low earning males are "losers", low earning females are called "mommy" and "victim".
Then there is differential mortality. Poor males die at four times the rate of poor females and five times the rate of well-to-do males. With poor males dying at such high rates it raises the average pay of males and leaves the average pay of females untouched - except that their pay becomes relatively lower. Dead men drop out of comparative data, but the effects of their deaths linger on.
My choices: how many hundreds of thousands of dollars have I lost during my life because I joined the Marine Corps for four years and then became a university professor? Choice. My choice.
I truly wish that Evelyn Murphy would get herself out from under the old paradigm so she can see that a gender revolution has taken place. The choices for women are now their own, at least to the same extent that men have choices. A woman's choices are no longer narrowed by the artificial constraints sharply imposed by society. And, as it is for men, the decisions and defaults of a person's life show up in measurable ways in aggregate statistics. Or, is good of equality news not anything the purveyors of victimhood are willing to hear?