Jon Henke
Bruce "McQ" McQuain
Dale Franks
Bryan Pick
Billy Hollis
Lance Paddock


Recent Posts
The Ayers Resurrection Tour
Special Friends Get Special Breaks
One Hour
The Hope and Change Express - stalled in the slow lane
Michael Steele New RNC Chairman
Things that make you go "hmmmm"...
Oh yeah, that "rule of law" thing ...
Putting Dollar Signs in Front Of The AGW Hoax
Moving toward a 60 vote majority?
Do As I Say ....
QandO Newsroom

Newsroom Home Page

US News

US National News

International News

Top World New
Iraq News
Mideast Conflict


Blogpulse Daily Highlights
Daypop Top 40 Links


Regional News


News Publications

Anatomy of a bad idea
Posted by: Bryan Pick on Monday, June 11, 2007

How many times have you heard the statistic that women are paid some percentage of what men make? These days, the most commonly cited statistic is that women working full-time make 81 percent of what their male counterparts do. This statistic (or slight variations, depending on whence the data) is usually accompanied by claims that, even when accounting for other factors, a disparity still exists, and of course, that discrimination is responsible for the gap. This is the justification for the Fair Pay Act of 2007 and for the Paycheck Equality Act, which is mostly based on the idea that women are performing jobs that are of "comparable worth" to jobs performed by men, but are not equally compensated in the form of wages, and would address this by forcing employers to "provide equal pay for jobs that are comparable in skill, effort, responsibility and working conditions." And the acts are specifically designed to force comparisons of completely different jobs in different industries: a typical example is comparing what nurses (who are usually female) make in comparison to truck drivers (usually men).

This is target practice for libertarians and anyone remotely familiar with economics. "Comparable worth," now re-branded as "pay equity," is supported by a pile of fallacies about statistics and economics, and predictably resorts to centralized, third party decision-making as the "solution."


The statistics are less alarming than you'd think from listening to the news. The BLS, for its part, says that by median weekly earnings, female full-time wage and salary workers made 81 percent of what their male counterparts made in 2005, and that in the 20-24 age category, they made 94 percent of what men of the same age made, without adjusting for other factors. Women who had never married made a full 96.7 percent of what men made.
Interestingly, among part-time workers, women's earnings were 108.6 percent of men's... but you won't hear "pay equity" advocates arguing for men to get 8% raises in their part-time jobs; their outrage is oddly selective that way. But I digress.

When trying to explain the gap, first, let's tackle the obvious: everyone knows that women make different choices than men do, and that these choices often sacrifice pay for other considerations. Choosing to have children and usually to be the more involved parent, to work fewer hours, and to pursue educational tracks and professions that are known to result in lower salaries (but may be more rewarding in other ways) are all going to have an effect.

This explains most if not all of the pay gap, even according to those who trumpet the gap as a result of unfair discrimination. leads with the full pay gap figure, but implicitly accepts the invalidity of the number by saying in their FAQ, "Research published by the Council of Economic Advisors has estimated that between one quarter and one third of the wage gap cannot be explained by legitimate factors such as education or experience." Aside from their, uh, creative use of the word "cannot," they're admitting up front that at least two thirds (and as much as three quarters) of the wage gap is legitimate. The American Association of University Women (AAUW) recently published "Behind the Pay Gap" in which Judy Goldberg Dey and Catherine Hill examined the pay gap of college graduates—and remember, more women than men graduate from college. They argued:
A large portion of the gender pay gap is not explained by women’s choices or characteristics.

Discrimination cannot be measured directly. It is illegal, and for the most part, people do not believe that they discriminate against women or other groups. One way to discover discrimination is to eliminate other explanations for the pay gap. To uncover discrimination, regression analysis was conducted to control for the different choices women and men make. An analysis of weekly earnings one year after graduation was examined as a function of full-time employees’ characteristics, including job and workplace, employment experience and continuity, education and training, and demographic and personal characteristics.

If a woman and a man make the same choices, will they receive the same pay? The answer is no. The evidence shows that even when the “explanations” for the pay gap are included in a regression, they cannot fully explain the pay disparity. The regressions for earnings one year after college indicate that when all variables are included, about one-quarter of the pay gap is attributable to gender. That is, after controlling for all the factors known to affect earnings, college-educated women earn about 5 percent less than college-educated men earn. Thus, while discrimination cannot be measured directly, it is reasonable to assume that this pay gap is the product of gender discrimination.
First, before controlling for these factors, Dey and Hill noted that full-time working women made 80 percent of what their male counterparts made one year after graduation. After controlling for well-known legitimate factors, their figure dropped to 5 percent—still a gap, yes, but a much less dramatic one. For the group that had been out of college for a decade, women working full-time made 69 percent of what men made, but after Dey and Hill accounted for other factors, the gap was only 12 percent.

Second, Dey and Hill have employed what economist and author Thomas Sowell calls the residual fallacy: the idea that one can control for the well-known independent variables in the statistics, and assume that whatever discrepancy is left over must be due to some other single factor X; unsurprisingly, as Sowell points out, it's also been used to "prove" genetic inferiority. In this case, Dey and Hill casually attribute the last 5 percent not to their own ignorance about factors affecting income, but to a definite cause: gender discrimination. And they're not alone: the same or greater sloppiness, parading as researched facts, shows up in most arguments that women are systematically being treated unfairly.

Further, this study (like all the others I've seen) neglected to disaggregate important groups. Many studies, for example, claim to control for education and other characteristics. What you don't hear in the news is that the researchers often group people together when significant differences exist within groups. Some studies simply group people by the numbers of years they were in college, or by basic degree level. Dey and Hill, to their credit, did separate graduates into a few levels of selectivity of their colleges and by their majors, but they didn't separate people by other important factors in their education. For example, grouping schools together by selectivity doesn't control for some schools having particularly respected programs. For another, not all graduates within a major are alike: while two students may get economics degrees from the same university, the guy with a Bachelor in the Sciences who took econometrics and several extra math courses (far more common among males than among females) is going to have an advantage on the job market over the lady who took less rigorous courses and earned a Bachelor in the Arts. Different types of engineering pay differently, as do different fields of law.

In short, the supposed "pay gap" is entirely, or very nearly entirely, explained by perfectly legitimate factors, separating workers by self-selection and qualifications rather than by sex. It really is just business.

Consider the alternative explanation: that women who are equally qualified and equally productive are taking the same jobs as men but accepting four-fifths of the wages. Any businessperson looking to gain a substantial edge over the competition would not discriminate against women but in favor of them, obtaining equal output with lower costs. Raise your hand if you really believe that all those greedy capitalists in all those lucrative industries have conspired to keep women out even at considerable expense to themselves.


The Solution

If the justification for "comparable worth" is nonsense, the prescribed method of addressing it is much worse. The idea of "equal pay for comparable work" is, as already stated, that employers be forced to "provide equal pay for jobs that are comparable in skill, effort, responsibility and working conditions." The problems with this are legion.

For one, what do skill, effort, responsibility and working conditions have in common? None of these things determines the value of the good or service provided to the consumer by the worker. The supporters of this bill, therefore, wish to divorce wages from what the consumer is willing to pay, thumbing their noses at the economic consequences that will follow as surely as night follows day. For the same reasons that modern "progressives" refuse to believe in the unintended consequences of the minimum wage, they cannot see that raising the cost of employment in fields dominated by women will have the exact opposite of the intended effect. If the cost of a thing rises, people will generally buy less of it. This is as true for labor as it is for consumer goods. If you're running a hospital, and the cost of each nurse rises 20 percent, do you:
  • (a.) raise the price of the health care you're providing?
  • (b.) cut corners in other parts of the hospital to cut costs?
  • (c.) cut non-wage benefits for the nurses, cut their hours, perhaps fire some of them?
  • (d.) force the nurses to work harder so that they produce that extra value?
  • (e.) some combination of the above?
  • (f.) take it on the chin and not meet your budget?
None of the above choices is good for either nurses or the patients they serve.

Next, who gets to decide what comparable levels of skill are, across industries and jobs? Why, a regulatory agency, that's who. Skill, effort and responsibility aren't objective criteria; just try measuring them in some way besides what people voluntarily trade for the output. I don't know when or if all people will finally discard the idea that effort determines value, but for anyone who's still a believer: would you rather pay your week's salary to a person who gives himself blisters while repeatedly digging a hole and filling it back in, or to a person who saves your child's life without breaking a sweat?

This agency also determines the relative value of working conditions. What is currently the subjective valuation of the people who actually work in these industries and in those particular locations will be overridden by distant bureaucrats.

Other provisions are no better. Supporters of the Fair Pay Act also say it would "prohibit companies from reducing other employees' wages to achieve pay equity." Where does this leave firms with employees who are already making more than the state-sanctioned wage for their skill, effort, etc.? To make payroll, heads may roll. And as a sign of the sponsors' magnanimity, the act would "[a]llow payment of different wages under a seniority system, merit system, or system that measures earnings by quantity or quality of production." But then, what is merit if not "skill, effort, and responsibility," which are all to be defined by regulation? What determines the quality of production if not the consumer's preference?

The Fair Pay Act and the issue of "pay equity" all smack of a mystifying faith in the competence of government. Whence springs this trust that political elites will make more prudent and efficient decisions than hundreds of millions or billions of people who are looking out for their own? Who can effectively gather and process the volume of information that is currently diffused throughout the entire population, and who considers himself trustworthy to determine, with the force of law behind him, what everyone else deserves? What else can you call it but hubris, through and through?

(For another good roundup of the issues surrounding comparable worth, see this article at the Library of Economics and Liberty. Cait Murphy at Fortune has more, including information on the current Fair Pay and Paycheck Fairness Acts.)
Return to Main Blog Page

Previous Comments to this Post 

It has to open the door for government dictating wages for everyone eventually after a decade or so.

And as for pay equity. In my admittedly limited range of work experience, women get paid less because they are less likely to quit over their pay. In fact, they seem to choose career over pay because they believe complaining about pay appears disloyal.
Written By: jpm100
URL: http://
Gee, it was 59 cents back in the 1980s. What’s probably happening is that as society becomes less sexist (women get into better jobs, seek to become attorneys and doctors, etc.) the pay rate goes up. There is nothing government can and should do here; rather, parents should work against sexist stereotypes that demand women stay home and watch the kids while men go out and earn a living. That not only harms both the man and woman, but damages the children who start believing that such a division is natural. That doesn’t mean that one parent can’t choose to stay home; however, it shouldn’t automatically be the wife. The only way you’ll get pay equity is if you have a cultural equity. That’s going to happen someday (we’re making progress), but it’s not going to be because of anything government does. (Though government shouldn’t descriminate in policies by, say, privileging the role of the mother vis-a-vis father in issues involving children).
Written By: Scott Erb
Excellent article. The governments involvement in this issue is unwarranted.

The difficulty I have with issues such as equal pay compressions is that it always drives the conversation to cold, hard, numbers. These numbers are merely “average” facts and do not speak to the “up-close” human element.

It is easy to see the potential inequality from a 1000 mile view, but inequality (as with many work related issues) really can only be gauged at a much closer view.

I have been fortunate enough to have worked in several fields, including the military, police, and now in the private software arena. In the military and police I saw equal pay for equal work, but based on merit (job performance and testing). Yet in the military and police work women candidates physical fitness standards were half of what was expected of a man.

In my current profession I have not witnessed any “perceived” inequality that could not be explained when you take a closer look at the individuals. This is not to say there aren’t people experiencing inequality, but utilizing average data only tells a quarter of the story.

Written By: Jeff
Excellent analysis, Bryan.
Written By: McQ

I also think Bryan did an excellent job finding the gapping flaws in the study on "pay equity." That is an example, IMO, of agenda-driven academic research and bad social science.

In my current profession I have not witnessed any “perceived” inequality that could not be explained when you take a closer look at the individuals. This is not to say there aren’t people experiencing inequality, but utilizing average data only tells a quarter of the story.
I can give an example in a microcosm the implications of trying to "fix" this with government policy.

When we hire people who teach in disciplines where their knowledge has considerable demand outside of academia we have to pay higher salaries than we do for people who teach such things such as political science and history. For some that is seen as unfair, since the teaching load and criteria for promotion are the same. But if we tried to pay the ’usual’ salary for such people, we’d either not fill the positions or get somebody who wasn’t very good. That’s comparing the exact same kind of job responsibilities and work load at the same place, and finding sometimes large differences.

So what are our options? 1. Pay the higher rate for those whose knowledge is in demand, and thus offer better quality programs; 2. Raise the salaries of everyone to equal those with high demand knowledge. This would break the budget and mean that there would have to be major cuts in programs and staff, and it would diminish the quality of the institution; or 3. Avoid the problem by not hiring people with that kind of knowledge, which would mean we’d be lacking in programs/courses that obviously have high demand and to which students are attracted. If we did that we’d hurt the institution and offer an inferior product. Obviously "1" is the best option.

Written By: Scott Erb
Fact of the matter is, when apples are compared to apples, there is no pay gap. When you factor out the amount of women who work part time; who give up seniority (choose to stay home for a few years after having children) and other things women voluntarily do to bring the "pay gap" on themselves, it comes down to a penny or two.

The methodology for this - is this the AAUW study? - is seriously flawed - one example is that that particular study lumped all medical professionals in one group - therapists, nurses, doctors, specialists. It doesn’t account that women do not tend to choose careers which are unpleasant of dangerous. It doesn’t account that women will decline overtime far more than men.

And it doesn’t even answered the most glaringly begged question - if women will work that much more cheaply than men, why are men being hired at all?

Target practice for us? More like fishing in a barrel. And grenade fishing at that.
Written By: The Gonzman
URL: http://
If this were really true, that a company can get the identical result at 81% of the cost, then why would any profit maximizing company ever hire a man?
Written By: Paul
URL: http://
One factor in determining pay rates that the cited study appears to have neglected is what the employee asks for before they get the job.

Let’s say that I’m looking to fill a position. I first look at pay studies that the HR department subscribes too. This tells me the pay range and percentages for the job title I’m looking for. (For example, let’s say that the median is $60/hour, with 80% between $50/hour and $80/hour.) I then adjust for factors such as the cost of living where the job is, the number of people there that have the skills and are out of work and so on; finally I decide what level of pay will attract the minimum skill level of employee I want. Let’s say that the minimum I can pay to get the people with the skills, experience and work quality I want is $55/hour. That number forms the bottom of my range.

I then do a completely different analysis: how much is the job worth to me. That tells me the total burdened cost that I can accept. Let’s say that I derive $105/hour in benefits from the work of this position, if done to my standards. Then, in order for the position to produce profit for me, I need to have no more than, say, $90 in total costs. Subtracting out benefits, taxes and the like, let’s say that leaves me with $75 that I can pay, maximum. Let’s further stipulate that I trust my recruiters, so I don’t have to play any games with that rate to make sure that they don’t let the offered rate creep up when I’m not looking.

That leaves me a range of $55 to $75 that I would pay for the position, and obviously, the better the candidate, the higher within that range I would be willing to pay. (That, in fact, is what it means when you see rates quoted as "depends on experience.")

Now, let’s say that I have 5 candidates that I interview who I decide would be capable of doing the job to my standards. Those 5 will have been told that we are looking to pay $55 per hour. They will then ask for something, and let’s say that they ask for, respectively, $55, $55, $60, $70 and $80. I would immediately eliminate the person asking for $80, because the job would then not profit me as much as I need, unless I could talk him down. (There are exceptions, where I can shift money between positions, but let’s keep it relatively simple.) Of the other candidates, I have to ask myself which would be the best person for the position, and hire that one.

Now here’s the factor that Dey and Hill apparently left out of their study: note that the people asked for different amounts of pay, given what they knew I was looking to pay and their own circumstances. It has been my experience that men split down the middle, with some being willing to take whatever I offer, on the grounds that they need a job, now, and something is better than nothing; while others will ask for quite a bit more than I’m wanting to pay, figuring that if they don’t ask above the maximum and get talked down, they’re leaving money on the table. Women, in my experience, will almost always accept the rate I quote as what I’m looking to pay. I assume that means that women who want to make higher rates go to jobs that list that as their minimum, but I have nothing but intuition to guide me on that.

Now, if I have two positions, I might hire one in at $55 and one in at $65, because those were the two best candidates within my range. Is it discrimination that I’m paying the one $55 and the other $65? Why? Does your answer change if they are both men, both women, or if the woman is making more than the man? If your answer does change, why does it change?
Written By: Jeff Medcalf
If this were really true, that a company can get the identical result at 81% of the cost, then why would any profit maximizing company ever hire a man?
Great point, but one that if followed, eliminates the whole debate ;-)

HELP - someone please HELP. I think I agreed with everything the good Dr. from Maine wrote!!!
Written By: meagain
URL: http://
Paul and the Gonzman - Exactly. As I argued, if a business owner can get the same output from a woman for four-fifths of the cost, he has all the incentive in the world to discriminate in favor of women in hiring decisions, rather than against them.
Written By: Bryan Pick
It would be more pay equity if men could put aside their egos and fairly compete.
From one woman who has a traditional male job.
Written By: VRB

Add Your Comment
  NOTICE: While we don't wish to censor your thoughts, we do blacklist certain terms of profanity or obscenity. This is not to muzzle you, but to ensure that the blog remains work-safe for our readers. If you wish to use profanity, simply insert asterisks (*) where the vowels usually go. Your meaning will still be clear, but our readers will be able to view the blog without worrying that content monitoring will get them in trouble when reading it.
Comments for this entry are closed.
HTML Tools:
Bold Italic Blockquote Hyperlink
Vicious Capitalism


Buy Dale's Book!
Slackernomics by Dale Franks