David R. Henderson  

80 Years Before Card and Krueger

PRINT
How much idealism is ideal?... Two Good Economists' Letters...

A labor economist friend who studies the minimum wage writes:

I found a 103 year old BLS report on a minimum wage increase in Oregon that had a stronger grasp on credible research design than Card & Krueger. Also, one of the authors is named "Bertha von der Nienburg" and she was 24 when the report was written. If I was the sort of person who did that sort of thing, I'd blog about this.

The study is "Effect of Minimum-wage Determinations in Oregon," Bureau of Labor Statistics, July 1915.

The authors are Marie L. Obenauer and Bertha von der Nienburg.

I got a little bogged down working my way through it and the authors don't come to any strong conclusions. But the care they take and the granularity of their data are amazing.

One of the things they did, besides getting data from many firms affected by the newly implemented minimum wage law, was interview some of the women affected. Here's an interesting segment from their page 68:

The belief was very prevalent among store women that the minimum wage had wrought only harm to them as a whole. The experienced women contended that formerly they had gotten through the day without any hurry or strain. If it was necessary to work a few minutes overtime, they did so willingly. Now, they said, they are under constant pressure from their supervisors to work harder; they are told the sales of their departments must increase to make up for the extra amount the firm must pay in wages. With business declining, this was hardly a possibility. The result was that the women were very worried and the worry was intensified in November, 1914--the month they were visited--because of the fear that large numbers would be dismissed after the Christmas rush was over and the dull days of January confronted the employers. These women did not ask themselves to what extent the same conditions would have prevailed in a poor business year had there been no wage regulation. They knew there were wage and hour regulations and that some women had been benefited, but if they had not personally benefited and had only experienced the pressure from above and the fear of the future, their anxiety found vent in heated denunciation of the minimum wage as the visible cause of their jeopardy.


Comments and Sharing


CATEGORIES: Labor Market




COMMENTS (12 to date)
Shane L writes:

Loosely related, but perhaps of interest to readers here, I recently read Niccolò Machiavelli's The Prince (1513) which makes this comment about onerous taxation and government seizure of private property hindering investment in the private sector:

"...he should encourage his citizens to practise their callings peaceably, both in commerce and agriculture, and in every other following, so that the one should not be deterred from improving his possessions for fear lest they be taken away from him or another from opening up trade for fear of taxes".

I was surprised to see this quite modern understanding of incentives in such an old document. I am not sure if this was a unique insight of Machiavelli's, or if it was common knowledge among administrators of the day. Wonder what he would make of minimum wages.

Daniel Kuehn writes:

Can you give some sense of what your friend meant by “stronger grasp of credible research design”? That’s intriguing but seems pretty implausible given the timeline alone, much less Card and Krueger’s own considerable contributions there.

Scott Sumner writes:

That final quote is interesting. In a recent post I concluded with this comment:

"I would add that the question of whether higher minimum wages are desirable is very different from the question of whether they affect employment levels. There are other important issues to consider, such as the impact of minimum wage laws on working conditions."

Rojellio writes:

My concern with many of the recent studies on minimum wage is that they are often defaulting to a secondary measure of the expected effect. Economics suggests that raising the minimum wage per hour should reduce demand for hours worked. It will only reduce the demand for employees if the employer chooses to keep the hours per employee the same. There is little or no reason to expect this.

An increase in minimum wage would be expected to not just reduce demand for low skilled workers, but to increase the supply of those looking for work. In addition it increases the incentive to ensure that the best lower skilled workers are used and each is optimally managed to earn their pay.

In other words, an increase in minimum wage will shift the power balance toward the employer. A big increase will do so a lot. Since most minimum wage jobs are hourly, one effective strategy for the employer is to use the improved power balance to enlarge the available labor pool of workers. Most will get fewer hours, as predicted by theory, with the better workers perhaps gaining a bit more and the majority losing a lot. A larger work pool gives the employer scheduling freedom, at the expense of more reliable hours for the low skilled workers. It is a predictable expected response to a minimum wage hike.

When an economist who is unfamiliar with how labor markets actually work conducts a study, they then "prove" minimum wage increases (or has no effect) on employment. The part time worker pool may have actually increased. But the real effect is indeed lower demand and lower numbers of hours worked.*

I am not an economist, obviously, but I did manage large numbers of employees for thirty years. My guess is most managers of part time employees would agree. Not sure why economists never seem to ask us.

* This also implies that studies which DO prove reductions in employment are also underestimating the negative effect. If they looked at hours worked instead of the secondary measure they would see greater negative side effects.

David R Henderson writes:

@Scott Sumner,
Good point. There are many margins on which to adjust. See my WSJ piece.
@Daniel Kuehn,
More to follow, probably tomorrow.

JK Brown writes:
“And in volume 2, page 33, speaking of New Zealand, " wages in the colony fell generally between 1879 and 1895. In 1889 the minimum amount of wages to be paid in industries was fixed by law. As a result the old and slow workers in the clothing and underclothing trades were all discharged and starved or became paupers," (vol. 2, page 64). ”

The references in the preceding is to: “ State Experiments in Australia and New Zealand,” written by William P. Reeves

The above is from a 1903 debate on socialism by Frederick J. Stimson. Of course, today, we don't call those put out of work paupers or let them starve.

Their blunder in forgetting, at the outset, to make special provision for the old and slow workers in the clothing and underclothing trades, — a blunder which they might have avoided had they watched the methods of the New Zealand Arbitration Court, — had been partly repaired by issuing licenses to old and infirm workers to permit them to work at less than the legal minimum.
Miguel Madeira writes:

"An increase in minimum wage would be expected to not just reduce demand for low skilled workers, but to increase the supply of those looking for work."

Why? If anything, i imagine more the opposite - with higher minimum wage, the workers (at least, for the workers who don't loose their job...) will less need of doing extra hours, or the have second jobs (then, lower labour supply) - I even suspect that this could be one of the reasons because many empirical studies have difficult in finding a strong relation between minimum wage and unemployment.

Alberto writes:

Reading the quote I was reminded of an unjustly neglected 1999 JPE article by Leamer

https://www.journals.uchicago.edu/doi/10.1086/250092

One of the few to explicitly explore the impact of minimum wages on effort. Highly recommended

Fred Thompson writes:

This is an obscure study, yes. But not an unknown one. Neumark cites it, as did Kennan in his 1995 JEL survey and Lechner in his 2011 piece "The Estimation of Causal Effects by Difference-in-Difference Methods Estimation of Spatial Panels."

Rojellio writes:

Miguel,

"Why? If anything, i imagine more the opposite."

Just to clarify, my argument is that minimum wage studies OVERestimate the harmful employment effects on less skilled workers (in case it gets lost in the dynamics).

I assume you agree that it lowers demand. I assume you agree it shifts demand from less skilled lower wage workers to higher skilled (more dependable, conscientious, courteous, productive, etc).

And thus I assume you are only asking why it would increase supply. The reason is that incentives matter and those previously out of labor market would be incentivized to enter the market for higher wages, those working fewer hours would be greater incentivized at the higher wage to work more hours, and those who were laid off due to the higher wages making employment unprofitable (this does still occur) will now join the ranks of those looking for work.

I do agree with you that there could be an effect the other way of minimum wage workers making so much money that they cut back hours. Seems like it would be tiny, but I could be wrong.

My point though is that the proper way to look at the issue is primarily hours worked not percent employed. "Minimimum wages cause unemployment" is not predicted by the theory unless hours per employee stays the same, which itself should not be assumed.

This sloppy shortcut then contributes to the dispute over outcomes. In short, we could and should frame the problem better.

Miguel Madeira writes:

There is a reason to give more importance to the unemployment than to reduction of hours worked is that loosing (or not finding) a job is a clear loss; working less hours is more ambiguous - if the reduction in hours worked is lower than the raise of the hourly minimum wage, you will probably end better.

"I do agree with you that there could be an effect the other way of minimum wage workers making so much money that they cut back hours. Seems like it would be tiny, but I could be wrong."

This is the old question of the supply curve of labour (how, in theory, could be "normal" or backward, because the income-effect and the substitution-effect go in opposite ways - I suspect that, in practice, the two effects largely compensate each other and in the end the curve is almost vertical)

Rojellio writes:

Miguel,

"Why? If anything, i imagine more the opposite."

Just to clarify, my argument is that minimum wage studies OVERestimate the harmful employment effects on less skilled workers (in case it gets lost in the dynamics).

I assume you agree that it lowers demand. I assume you agree it shifts demand from less skilled lower wage workers to higher skilled (more dependable, conscientious, courteous, productive, etc).

And thus I assume you are only asking why it would increase supply. The reason is that incentives matter and those previously out of labor market would be incentivized to enter the market for higher wages, those working fewer hours would be greater incentivized at the higher wage to work more hours, and those who were laid off due to the higher wages making employment unprofitable (this does still occur) will now join the ranks of those looking for work.

I do agree with you that there could be an effect the other way of minimum wage workers making so much money that they cut back hours. Seems like it would be tiny, but I could be wrong.

My point though is that the proper way to look at the issue is primarily hours worked not percent employed. "Minimimum wages cause unemployment" is not predicted by the theory unless hours per employee stays the same, which itself should not be assumed.

This sloppy shortcut then contributes to the dispute over outcomes. In short, we could and should frame the problem better.

POST A COMMENT




Return to top