In the paper, the authors use marginal productivity theory to identify instances of wage discrimination. (Women do not constitute 50% of the work force in the sample, BTW, nor do women’s wages differ from MP in all instances.) The authors find that the marginal product of women in the sample is lower than the marginal product of men, but that the wage paid to some women is lower by more than can be accounted for by their lower marginal product. In others cases, marginal productivity does a good job of explaining wages.

In economics, the marginal productivity theory is a powerful way of understanding wages and wage differentials, as Steve said. I think even Steve would agree that MP theory does not require that everywhere and everytime wages must equal MP, but we can think of no more powerful theory for understanding wages than MP.

Either the cost-of-life is an input to his formula (and he omitted it), or it is the output of the formula. In the latter case, it would be useful to compare this across different industries. If some economist is toiling away at this, well then the well-read Steven Landsburg should direct his readers to him or her. Or if no one is doing this, it represents an opportunity for research…

Jon

Neil – I can only see the abstract of the paper you mention, but the one example they pick out there is women, which is quite a big sector. “Divergence” of 50% of the population seems to blow a bit of a hole in the theory. That women are paid less than men for similar jobs is well documented. The “marginal product” theory can only explain this if women have lower productivity, or are paid in employment “amenities”. I find these explanations somewhat unsatisfying. I am sure this must have been extensively studied.

@ G: I take objection to your claim that this article is distateful. People make risk-reward decisions regarding potentially dangerous fields of employment (coal miners, fighter pilots, etc.) all the time. Sometimes the outcomes of these decisions end up on the bad side of the coin- that’s just reality. Isn’t it better to study and try to understand those choices and tradeoffs in order to learn how make the best decisions possible rather than just pretend that the tradeoffs don’t exist?

Fine print: The statement ‘In a competitive labor market, workers are paid their marginal product’ is tautological — being paid marginal product is essentially the DEFINITION of a competitive market. The author’s assertion that there is ‘ample evidence’ of the statement is therefore ignorant.

At the micro level, see a paper by Hellerstein, Neumark and Troske in the Journal of Labor Economics, July 1999. They examine differentials between wages and marginal productivities using plant level data and find only one example where there is significant divergence.

Divergence does not disprove the theory because there may be other factors that cannot be fully accounted for, such as when workers are paid partly in job amenities.

A few Google searches will point you in the direction of David Von Drehle, who wrote a 2004 book “Triangle: The Fire That Changed America.” Enough of the book is available via Google Books preview, in case you don’t have ready access to it at the Rochester library. It explains that it was never fully known whether the 2nd door was in fact locked. This is important because the judge in the criminal trial instructed the jury that the owners could only be convicted if they knew that the door was locked at that time. And, in fact, conflicting testimony was given at the trial. The law at the time had called for all doors to be unlocked.

Your whole analysis falls to pieces when you assign agency to the workers here. You completely forget that the factory owners, if they were rational economic actors, would have calculated the calamitous effects of a criminal conviction for negligence, or civil recompense to the families of the victims.

As it happened, the jury had enough of a reasonable doubt that the owners had cognizance of the doors being locked at the time of the fire, and they were acquitted. But they were open to civil suits, and ultimately had to pay $75 per victim. I don’t know the amounts of comparative suits in those days, but 10 weeks salary is slim by today’s standards. Let’s face it: any rational actor in society with a fair system of civil and criminal law must make decisions to reduce their exposure to punishment.

Your conclusion is thus under-informed and patently “I’m guessing that no 1911 garment worker would have wanted to work in a factory with unlocked exit doors.” You should have researched New York State law in 1911: “All doors leading in or to any such factory shall be constructed as to open outwardly, where practicable, and shall not be locked, bolted, or fastened during working hours.” A Google search will bring you references to this law from the literature.

And to *thus* conclude that “the market worked” after such half-assed research on your part is painfully heartless. (I appear to have done more research while watching the NCAA basketball today).