What Does the Minimum Wage Do? An Interview with author/economist Paul Wolfson

July 24th, 2014 at 7:39 am

To celebrate the fifth anniversary of the last increase in the federal minimum wage and to call attention to the fact that the federal wage floor has not risen in five years, the US Department of Labor has declared July 24th to be a “Day of Action.”

Coincidentally, a new book surveying the scholarly literature on the effects of the minimum wage, What Does the Minimum Wage Do? came out earlier this month, written by Dale Belman and Paul Wolfson.  Below, I interview Mr. Wolfson (who’s an old friend, btw).

Jared Bernstein: What does your work suggest about the Fair Minimum Wage Act (FMWA): the proposal to raise the federal minimum wage from its current value of $7.25 to $10.10 in three annual steps and then index it to inflation?

Paul Wolfson: First, our work: in our book, we surveyed more than 70 analyses of the effect of the minimum wage on employment.  By and large, the strongest studies in terms of statistical rigor reported an effect on employment that ranged between negligible and none.  In addition, we performed our own meta-analysis, a procedure that combines the results of different studies in a statistically rigorous way, and this confirmed the result of “negligible to none.”

How does this relate to the FMWA? The proposal would increase the federal minimum in three $0.95 steps of 13%, 12% and 10% each.  In the last 35 years, increases in the Federal Minimum Wage have ranged between 7% and 14%, with an average increase of 11%-12%.  Thus, the proposed increases are quite typical of historical experience, and this suggests that if there is any effect on employment it will be too small to be detectable.

At least as interesting as the question of the minimum wage and employment is “What is the effect of the minimum wage on low income families?”  What can you say about this?

PW: We discovered in the course of writing the book that there’s actually little useful work that addresses this issue.  Nearly all of the studies in this area instead ask whether the minimum wage has reduced the percentage of families whose income places them below the poverty line.  For several reasons, this turns out not to be an interesting question.  First, the poverty line was developed about a half century ago and is widely regarded as an out-of-date measure of economic well-being.  Second, it leaves out a variety of government programs that effectively increase family income.

A more interesting question asks whether the minimum wage has a noticeable effect on the incomes of low income families, some of whom are officially poor, some of whom are not.

Economist Arin Dube of the University of Massachussetts-Amherst recently completed a statistically sophisticated analysis of the effect of the minimum wage on different parts of the income distribution, from the incomes of very poor families all the way up to those we might call the lower middle class or solidly middle class.  He reports that “the evidence clearly points to moderate income gains for low income families as a result of minimum wage increases.”

Your book is titled: What Does the Minimum Wage Do? How about giving us the elevator ride answer to that question.

PW: Limiting my answer to the historical experience, it does not have much or any effect on either the level of employment or on the unemployment rate.  It appears to reduce churn in the labor market, both the amount of hiring and the amount of quits and firings.  In fact, this may be one way in which the wage increase is absorbed: by lowering costs to low-wage employers of turnovers, vacancies, and training new workers.

The minimum wage increases wages for the lowest paid 10% of all employees and perhaps as much as the lowest paid 30% of women.  This is pretty much all that we can be fairly confident about, and it is these effects that led my co-author to say that the American experience of the minimum wage is largely one of intended consequences.  Despite the hundreds of studies in just the last 25 years, much remains unknown or known with insufficient certainty: the effect of the minimum wage on low incomes, on prices and output, on the different employment and unemployment experience of men and women, on the people’s choices to remain in or leave school, to name just a few.  Work has been done in each of these areas, but either too little to be yet confident of the results or too much of what exists turns out to be plagued by statistical problems of one sort or another.

So, if it’s not through cutting jobs, how do low-wage firms absorb the increase in labor costs?

PW: Again, we don’t know for sure, but one possibility mentioned above is the reduction in turnover.  This has two effects.  One is a more experienced labor force (since employees hang around longer).  The second is lower hiring costs, perhaps low enough to encourage firms to expand slightly since they expect to have a longer period for amortizing these expenses.  In addition to the reduction in turnover, there is mixed evidence that restaurants (and other firms) respond by increasing prices, somewhat more evidence in favor than against.  Similarly, studies that look for the response in lower levels of firm profits also report mixed evidence, generally considerably weaker than that for prices.  Finally, there are hints that the higher minimum wage draws people into the labor force who had been sitting it out, leading to an increase in the average quality of prospective employees and thus more efficient firms.  All of these things may be in play, each individually too small to be easy to detect in the available data with current techniques.

Print Friendly, PDF & Email

9 comments in reply to "What Does the Minimum Wage Do? An Interview with author/economist Paul Wolfson"

  1. Nylund says:

    Most minimum wage studies I’ve seen use variations in state-level wage rates. This is fine for assessing the direct impact of the wage rate. My curiosity revolves around the fact that there are numerous union contracts that contain clauses regarding ties to the federal minimum wage. If the federal minimum wage were to change it should, in theory, trigger these clauses and change wages not only for minimum wage earners, but also those whose contracts specify wage rates at some specific level (or ratio) above the federal rate. This could have effects that you wouldn’t pick up using a methodology based on changes in state wage rates.

    If there’s research that includes this contract-triggering effect, I’d be very interested in seeing it.

  2. Oliver Roosevelt says:

    Things like this may not be discoverable, but over time, say the first five years beginning with implementation of a higher minimum wage in an area, I’d like to know the effect on government programs, eg,
    Does the amount spent on assistance programs go down any appreciable amount? Less money on WIC? on SNAP? etc, etc, etc.

    I would be willing to pay $.15 more for a hamburger at the local fast food joint, particularly if I knew that the increase was largely going to the workers. There have already been some increases. I would guess to increased cost of ingredients. I would be less than enamored if the delta was $.30 much of which was going to raise the CEO’s annual pay from $2 million to $4 million, or from $8 million to $16 million. Likewise, I would hope that the increased prices at the local Walmart would go largely to the workers, not to 6 individuals whose net worth already exceeds that of the poorest 40 % of the US population.

  3. Lloyd Levy says:

    What about the assertion that a higher minimum wage causes unemployment among teenagers? What is its standing now as doctrine?

  4. jonny bakho says:

    Even at low wage businesses, wages may not be a large percentage of the business cost. A 10% increase in min wage can be absorbed through much less than 10% price increase.

    In poor countries with no minimum wage, business hire workers to do low productivity work. There is little incentive to invest in training or tools to make workers more productive. Establishing a floor on wages forces employers to consider how to make their workers productive enough to pay for the wage. Minimum wage promotes investment in worker training and tools to improve worker efficiency.

    Raising Minimum wage also increases SS revenue, something EITC does not.
    -jonny bakho

  5. Robert Buttons says:

    From a scientific standpoint, we know nothing about the minimum wage.

    All of the studies suffer from low statistical power. Many suffer from confirmation bias and a few from simple idiocy. A good example of the latter 2 is: “Minimum wage Channels of adjustment” Hirsch, et al. In the last year of their study, the studied businesses suffered an unexpected 20% decline in profits. Did they extend the study to see if this a long term effect of higher MW? Nope. They just blamed it on macroeconomic factors and concluded higher MW has no ill effects.

    We know, from the “experiment” in Samoa, a massive increase in MW destroys economic activity. I would like to ask an econometrician: (1) What is the inflection point? Ie.At what wage, exactly, does MW go from helping workers to hurting the economy (if there isn’t an inflection point, we could raise wages to infinity) (2) How was $10.10 MW arrived at? What statistical methods were used to arrive at this proposed wage and how has a single wage for our (very diverse) whole country been validated?

  6. urban legend says:

    Discussion of the impact of minimum wage seldom seems to mention the fact that competitors for most companies will face the same challenges. Employers generally will not face a competitive disadvantage. Empathy for the employer facing higher costs is strengthened, and perhaps exaggerated inappropriately, along with failure to recognize that price competition (if there is no collusion) will make it harder to pass through the cost increases. The net is that over time there are many ways to absorb moderate increases, including the turnover/productivity improvements, minor price increases with minimal if any decrease in sales volume, minor reductions or delays in profit-taking, and increased sales from people whose incomes have been improved.

    Mr. Buttons, there almost certainly is an inflection point because, we have to assume from lifetimes of observing reality in many contexts, curves always have inflection points. But we aren’t talking about huge increases, and when we haven’t had an increase in over five years, and had gone 10 years without one from 1997 to 2007, and in real terms even the new rate will be below where it was a few decades ago when prosperity was shared more equally in general, and wages in general have been relatively stagnant for almost a generation, we can be pretty confident we are not even close to such an inflection point.