Minimal Reasoning About the Minimum Wage

Kevin Drum is happy with the employment situation in Oregon because he feels that he has some serious evidence refuting the notion that higher minimum wages ceteris paribus don’t result in lower unemployment. Talk about over-reach.

In 2002, voters here raised the state’s minimum wage — and mandated automatic annual increases to keep up with inflation. Oregon’s 100,000 or so minimum-wage workers are paid at least $7.50 an hour, a rate that will increase to $7.80 in January, well above the federal $5.15 minimum.


Four years later, though it is impossible to say what would have happened had the minimum not been raised, Oregon’s experience suggests the most strident doomsayers were wrong. Private, nonfarm payrolls are up 8% over the past four years, nearly twice the national increase. Wages are up, too. Job growth is strong in industries employing many minimum-wage workers, such as restaurants and hotels. Oregon’s estimated 5.4% unemployment rate for 2006, though higher than the national average, is down from 7.6% in 2002, when the state was emerging from a recession.

What is wrong? None of these statistics actually measure just minimum wage jobs. For example, that the unemployment rate in Oregon is down from its previous peak means very little. Unemployment generally drops during an economic expansion and wages rise. So this doesn’t mean that wages and unemployment dropped because the minimum wage increased. Further, the argument isn’t: if you raise the minimum wage the economy wont expand. The actual argument is: ceteris paribus, raising the minimum wage (or instituting any price floor) will lead to excess demand. This isn’t really as controversial as people like Kevin Drum like to make it sound. Draw supply and demand curves and set the price floor above the market clearing wage. Supply is larger than demand. Granted this is theoretical, but there is not just some evidence that supply (demand) curves have a positive (negative) slope with respect to price; there is a veritable mountain of evidence. This is why we have the term law of supply and demand. If we were really going to use this evidence we’d have to look just at minimum wage jobs and then control for the fact that Oregon’s economy went through an expansion. Failure to do this means that the data isn’t telling us all that much.

That wages are rising also doesn’t help either. After all, as Brian Doss notes, people who benefit from this kind of policy weren’t going to get higher than minimum wage jobs anyways so noting that wages have gone up for those in jobs that pay above minimum wage is irrelevant. And even the fact that there is job growth in industries that have minimum wage workers, while a bit better, still doesn’t tell us all that much. After all, the economy is expanding so it is reasonable to think that part of the expansion is due to the expanding economy. That is, we’d have seen an expansion in this sector of the job market as well. Again, we’d want to control for the effect of the economic expansion.

Basically, what we’d like to know is, how many minimum wage jobs would we have at say $5.15/hour and at $7.50/hour, and take out the effects of the expanding economy? None of the above statistics tell us this answer, not even close.

So when Kevin writes,

Read the whole thing. Deborah Solomon provides both sides of the story, but it’s worth noting that virtually all the evidence on the anti-minimum wage side is either anecdotal or theoretical. The evidence on the pro-minimum wage side is concrete and statistical. You can decide for yourself which kind of evidence to believe.

It becomes painfully obvious how ignorant he is of this issue and what actually does constitute evidence. For as Alex Tabarrok and Tyler Cown have noted most of the people who work at minimum wage jobs are below the age of 25. Completing their education and developing a work history will ensure that many of these people will see a substantial rise in their hourly wage rate. Thus the idea that the minimum wage is a tool to fight poverty is highly suspect.

And surely Kevin Drum isn’t intending to impugn the reputation of David Neumark. Prof. Neumark has, you know, actually done the kind of statistical analysis that Kevin thinks he has done. Prof. Neumark’s conclusion in regards to raising the minimum wage,

The evidence from a large body of existing research suggests that minimum wage increases do more harm than good. Minimum wages reduce employment of young and less-skilled workers. Minimum wages deliver no net benefits to poor or low-income families, and if anything make them worse off, increasing poverty. Finally, there is some evidence that minimum wages have longer-run adverse effects, lowering the acquisition of skills and therefore lowering wages and earnings even beyond the age when individuals are most directly affected by a higher minimum.

Perhaps Prof. Neumark is simply a Republican stooge. And regarding the actual question we want answered here, Prof. Neumark notes that for every 10% increase in the minimum wage employment for teenagers falls by 1% to 2% (lets call it 1.5%). This translates into just under 8% fewer jobs for teenagers than if the minimum wage hadn’t been raised. And here is an interesting finding Prof. Neumark mentions, “A higher minimum wage increases the probability that a teenager will leave school to look for a job.” Not exactly the best way to go about reducing poverty and peolple living on low incomes.

It is probably true that small changes in the minimum wage have an effect that is too small to notice. However, changing the minimum wage from $5.15 to $7.80 isn’t a small change, that is a 50% increase in the minimum wage. And on top of this there are better ways of fighting proverty such as with the Earned Income Tax Credit. Under this policy it not only pays to work (i.e., reduces dependency on welfare and other government transfers) it also pays to take better paying jobs. And finally, Kevin’s dopey comment about theory vs. statistics is just that, dopey. What does Kevin think economists use to develop their statistical models if not theory?

FILED UNDER: Blogosphere, Economics and Business, US Politics, , , , , , , , , , ,
Steve Verdon
About Steve Verdon
Steve has a B.A. in Economics from the University of California, Los Angeles and attended graduate school at The George Washington University, leaving school shortly before staring work on his dissertation when his first child was born. He works in the energy industry and prior to that worked at the Bureau of Labor Statistics in the Division of Price Index and Number Research. He joined the staff at OTB in November 2004.


  1. I would be much more impressed if he took the same data (unemployment rates, wage growth, etc) and then added what happened to the 100K minimum wage jobs (increased, decreased, etc) Plus if you could get the data the jobs paying just above minimum wage.

    Here in my part of Texas, the standard starting wage for flipping burgers (the quintessential McJob) is about $7.50. But that is due to market forces, not government programs. So even better would be to compare the growth in minimum wage jobs to different states.

    Of course if you are relying on Drum for your economic analysis, you get what you deserve.

  2. M1EK says:

    “And on top of this there are better ways of fighting proverty such as with the Earned Income Tax Credit.”

    Every time I see this argument, I look for places where the author has actually advocated raising the EITC.

    Just as I thought. A diversionary tactic.

  3. Steve Verdon says:


    Keep putting words in my mouth and I’m going to start changing your name here to DickHead complete with a link back to your website. Are we clear on this?

    I’d be all in favor of raising the EITC if it was revenue neutral–i.e. it was matched with cuts for transfer programs.


  4. Steve Verdon says:

    Oh and M1EK, next time you decide to speak for me, you arrogant nitwit, why don’t check first so you don’t make a moronic mistake like missing this post of mine. Or, google is your friend.

    Further, the minimum wage is basically a tax on the employer and the consumer. Often the consumer are the very same people that the minimum wage is trying to help. Not exactly good policy when you stick it to those you are trying to help. A better way to do this would be to expand the earned income tax credit.

    Tell me M1EK, does it hurt to be this stupid or does the lack of a central nervous system simply mean you don’t feel any pain?

  5. Richard Gardner says:

    Both Oregon and Washington State have minimum wages well above the national level. Besides restaurant employees, who also get tips (in WA restaurant employees minimum wage is NOT offset by tips, unlike most states – think it is 36-14 on the states), I agree with Steve that most of the minimum wage earners are not heads of households. In Oregon, I expect that the largest number are pumping gas (like NJ, Oregon has a Keep Our High School Dropouts Employed Act, mandating a ban on self-serve gas).

    Despite cheery words to the contrary, WA and OR have some of the highest unemployment levels in the country, but in ways that are different from the rest of the country with most of the unemployment in the rural areas.

  6. M1EK says:


    What I’m looking for is an instance where you lobbied for an increase in the EITC. Not “instead of this, it would be smarter to do that”, but just “it would be a good idea to increase the EITC”.

  7. spencer says:

    I have not seen Neumark’s study. But if you look at BLS data on teenage employment you find that both teenage employment and the teenage participation rate have a positive correlation with changes in the minimum wage. The national data strongly implies that raising the minimum wage draws additional teenagers into the labor market and that they find more employment as a result.

    So it is very difficult for me to reconcile his claims with what what I see in the national data.

  8. As someone with an economics and investments background, I can tell you that changing the minimum wage, even by 50%, is not going to have a huge impact on the economy. Most employers are well over $7.80/hr, so it would only affect a small number of firms.

    Also, the 50% figure is misleading in that it is a one-time hit. Spread out over a number of years, it is closer to the level of inflation.

    What a higher minimum does do is protect workers from being abused by employers.

  9. spencer says:

    From 1955 to 1978 there were 7 increases in the minimum wage and the real minimum wage rose from
    $2.80 to $4.23. Over this period teen employment more then doubled from under 4 million to some 8 million. Part of this was the baby boomers going though this age bracket. But teen employment actually rose more then the increase in the teen population as the teen participation rate rose from 44.5% to 57.9%. Since 1978 the real minimum wage has fallen
    from $4.23 to $2.70. Over this period teenage employment fell from some 8 million to under 6 million and the teen participation rate fell back to 44.4%.

    Now can you explain how your assertion that a higher minimum wage leads to teenage unemployment can be reconciled with this historic data?

  10. Steve Verdon says:


    I know you are aware of controlling for things like economic expansions, yet you make no such effort to control for this. So on one hand we have a respected economist who has spent a good chunk of his life studying this topic and we have you taking 15 minutes to pull some data from the BLS servers. Who do you think I’m going to believe?


    What I’m looking for is an instance where you lobbied for an increase in the EITC. Not “instead of this, it would be smarter to do that”, but just “it would be a good idea to increase the EITC”.

    Goal post moving duly noted jerk.

    Political Critic,

    As someone with an economics and investments background, I can tell you that changing the minimum wage, even by 50%, is not going to have a huge impact on the economy.

    And I expressly noted that this was not the basis for opposition to the minimum wage. In fact, this is a complete strawman argument.

    Also, the 50% figure is misleading in that it is a one-time hit. Spread out over a number of years, it is closer to the level of inflation.

    Actually, I think it was a large initial increase followed by inflation adjustments. That one time increase is where you’l likely see the biggest effect. Tying it to the CPI would likely mean that whatever effecs there are (both positive and negative) would be too small to observe.

    What a higher minimum does do is protect workers from being abused by employers.

    Considering that the transactions are voluntary this is errant nonsense.

  11. spencer says:

    On Yes Steve — never confuse you with the facts.

    I will not accuse Neumark of torturing the data
    until it confesses to a crime it did not committee.

    All I am asking you to do is explain why the data does not support your position.

    Your reply does not address my question.

    Advanced largely by those who mistakenly think their static, purely competitive model explains real world dynamic .

    My position is consistent with the general consensus among mainstream economists.

    Your position is a very, very minority position.

  12. Steve Verdon says:


    My position is not inconsistent with rising teen employment. If over a given time period the general trend is positive, then raising the minimum wage would likely result in a downward shift in teen unemployment at the date of the increase. Why should we expect teen employment to be rising? Because the economy usually expands. That is, expansions tend to be longer than contractions. Add on the increasing population of the U.S. as well as the demographic effects of the baby-boomers and it seems to me that the minimum wage could have a negative impact, but still other factors result in rising teen unemployment.

    Now, if the above is true, then over a given time horizon it is possible that teen employment increases while there are periodic negative effects due to increases in the minimum wage. I know you are smart enough to figure this all out spencer.

    I know it is all fun and giggles to take the other person’s argument and stretch it to include something that it doesn’t. I haven’t written that raising the minimum wage will devastate our economy, nor have I written that teen employment both in absolute numbers and in participation rates can’t change. You, better than most commenters, should realize that there is more at work in a given time series than simply the price.

    My position is consistent with the general consensus among mainstream economists.

    Frankly I don’t know what to call this other than a lie. The number of citations at the end of Neumark’s article are extremely long.

    Bottomline is you really have about as much evidence as Drum does, which is pretty much nothing.

  13. spencer says:

    We all know that there are conflicting studies on the impact of the minimum wage, but I will stick to my position that the recent studies have found essentially no evidence that it has the impact classical theory calls for and that has become the consensus of main stream economics.

    Even Donald Boudreaux has conceded that there is no evidence that the minimum wage causes the damage he claims it does but still says in will stick with his theory.

    I suggest that you actually look at the data I referenced rather then insulting me to demonstrate your point.

    I am a PhD. economist with ten years experience in government, ten years as Chief Economist for a major multinational financial firm and 20 years running my own consulting firm not somebody who spent 15 minutes taking some data from the BLS.