/* trackback code -- i added this */

Saturday, August 20, 2005

The Incomparable John Roberts

There's been a lot of discussion in the MSM lately regarding Roberts and issues related to gender discrimination. First, the easy thing: like a lot of people, I thought yesterday's frontpage WaPo article totally misread this 1985 statement:
"Some might question whether encouraging homemakers to become lawyers contributes to the common good, but I suppose that is for the judges to decide."
But a lot of other quotes have come up, too. I find it hard to know what to think about some of the more controversial-sounding ones without knowing more than the articles I've read have provided about the context in which Roberts was writing, and the written context of those words of his that have been quoted.

For example, consider this discussion, from Linda Greenhouse's August 16 NYT piece:

In the documents, there was nothing "squishy" about Judge Roberts's response to an employment discrimination theory known as "comparable worth" then being put forward by women's rights groups.

The theory started from the premise that antidiscrimination policies that addressed solely "equal pay for equal work" failed to take into account the fact that heavily female occupations generally paid less than male-dominated job categories.

In 1984, a federal district judge in Tacoma, Wash., accepted the theory and ordered new pay scales for female state employees in job categories where 70 percent or more of the workers were women.

Three members of Congress, all Republican women, wrote to the White House not to join the state's appeal.

"I honestly find it troubling that three Republican representatives are so quick to embrace such a radical redistributive concept," Mr. Roberts wrote in a memorandum to Mr. Fielding. "Their slogan may as well be 'From each according to his ability, to each according to her gender.' " The federal appeals court in San Francisco overturned the ruling the next year in an opinion by Judge Anthony M. Kennedy, whom Mr. Reagan named to the Supreme Court a year later.

And Todd Purdum's August 19 NYT piece has this quote of Roberts's:
"staggeringly pernicious law codifying the anti-capitalist notion of 'comparable worth' (as opposed to market value) pay scales."
I leave open the possibility that there is more to the context, but for the sake of argument let me take Roberts's words at face value. Let me also state my understanding of what "comparable worth" means broadly, i.e., not necessarily specifically in the context Roberts was discussing. Comparable worth compensation systems basically seek to develop objective (i.e., describable) measures of the type of work done in a job and the skills needed to do it. They then rates different jobs on some index that is meant to allow comparisons across different jobs in terms of the assessed features. Thus two jobs that involve the same basic skill set (say, a highschool degree and the ability to type X words a minute) and have very similar other features (most work is done in a desk during business hours) would be assigned the same salary in comparable worth systems.

Let's consider the issue of whether markets always lead to what might be considered neutral (if not otherwise "fair") wages and salaries across genders. Under perfect competition with no market failures, economists would expect unregulated markets to lead to equilibria (outcomes) in which wages equal what's called the value of a worker's marginal product (VMP). In other words, we'd expect people to be paid the amount they contribute to a firm's revenues, holding constant the firm's physical capital, number and quality of other workers, and so on. A firm that discriminates in the sense of paying less than a worker's VMP is vulnerable to losing workers to competitors who do not so discriminate. The result would be that equally qualified women face no discrimination in equilibrium: either they get paid the same as men, just at different employers, or the discriminating firms go out of business because they wind up paying a higher wage to equally productive workers (i.e., men).

This argument is the one often deployed by economists to assert that competition undermines discrimination, and I imagine it is the argument Roberts had in mind when implicitly appealing to "market value" in normative rather than positive terms (I will make this assumption, rather than assuming that he simply doesn't care whether there is a normative justification for unequal pay).

I think there are (at least) two serious arguments against what I understand to be Roberts's position here.

First, the premises of the competition-undermines-discrimination argument may not hold. It is well understood that when consumers in a perfectly competitive market have discriminatory preferences, the free-market equilibrium not only will not penalize discrimination, it will reward it. Thus firms that hire women to work in traditionally male occupations will lose customers to the extent that customers are sexist and are aware of the firm's nondiscriminatory hiring. This sort of thing would be one explanation for the existence of systematically "male" and systematically "female" occupations; to the extent that women are kept from high-wage occupations, this dynamic would help cause and sustain a gender gap in wages even in the presence of healthy competition. Some might argue on consumer sovereignty grounds that there is nothing wrong with such an outcome. While I would certainly disagree with that view, it is beside the point in any case: the issue is whether or not "market value" should always necessarily be privileged when considering the fairness of different compensation systems.

Moreover, perfect competition is only a model; no one reasonable really thinks any actually existing market is perfectly competitive. That fact certainly does not mean that "very" -- or even "moderately" -- competitive markets won't chase out discrimination. But it could take a long a time to happen, especially if firms have deep pockets and/or indifferent or imperfectly-observing shareholders (kind of like governments). There is some decent econometric evidence that deregulation of the banking industry reduced compensation of women working in banking considerably less than it reduced men's compensation--exactly as competitive theory predicts (see "The Division of Spoils: Rent-Sharing and Discrimination in a Regulated Industry", by Sandra E. Black and Philip E. Strahan, AER, September 2001; link for AER subscribers). Such evidence is good news in the sense that it means that more competition can be good, but bad news in the sense that it suggests that women working in industries with less competition may well face serious discrimination. (For more evidence on this issue, see Hellerstein, J. K., Neumark, D., & Troske, K. R. (1999) "Wages, Productivity, and Worker Characteristics: Evidence from Plant-Level Production Function and Wage Equations," Journal of Labor Economics , 17, 3, 409-446 [Link (via JStor)].)

Second, even ignoring the points just made, I don't really think there is a very good per se explanation for Roberts's opposition to comparable worth as departing from market value. The jobs in question were government jobs, many---if not most---of whose compensation levels I imagine are set via civil service rules. If I'm right, then comparable worth is just one particular way of setting salaries. I don't know if it's the most reasonable, partly because I don't know the details of the particular comparable worth policies proposed, and partly because I don't know the details of existing civil service compensation rules. But the idea that existing public sector compensation policies somehow perfectly reflects the exalted equilibria of "market value" strikes me as about halfway between nonsequitur and silly. In this sense, comparable worth's "redistributive" nature would be "radical" only insofar as it led to large changes in amounts; it would hardly be radical in departing from "market value". Note that none of this means I think current civil service pay scales are necessarily better or worse than other alternatives, e.g., merit pay, though I will say that I am skeptical that the public sector can simply be run like private firms.

One of my biggest concerns about Roberts is that he seems to have relatively little experience living life the way most people do. He is the son of wealthy people, he went to an exclusive boarding school, he went to college and law school at Harvard, and his entire professional experience was spent either relatively far up the ladder in the executive branch, as a lawyer in what I understand to be a high-dollar Washington law firm, or as a clerk or Judge for appellate courts. Now, most of these things are prima faciae evidence of legal qualification in the sense of knowing or understanding the legal concepts Roberts will need to consider if he is confirmed. And in that sense they are certainly good things.

But what is glaringly missing from Roberts's bio is any evidence that he knows what most people's lives are like. It's unclear that he believes there is or has been (since...when? perhaps the Civil Rights Act? earlier?) any economically sustainable discrimination based on gender (and, from other comments I've seen, race). I do not think these issues are ones that can or should be considered only in terms of abstract theories---legal or economic---by either lawyers or economists.

Say what you will about O'Connor, she had a politician's attunement to the practical, real-world implications of the Court's decisions.

The more I read about Roberts's comments and papers, the less I think he understands---or, at least, understood---about the world inhabited by most people.

Lastly, let me note that the NYT website has posted three long files (one, two, three) concerning Roberts's views on comparable worth. I doubt I'll have time to read them, but it could be illuminating.

Note: For some related and interesting discussion, see this post at Body and Soul (which I noticed at Kevin Drum's WM blog) and this one by Dahlia Lithwick.

0 Comments:

Post a Comment

<< Home