Minimum Wage and Magical Thinking

By Ronald Bailey

Sunday, April 05, 2015

If all other factors remain equal, the higher the price of a good, the fewer people will demand it. That's the law of demand, a fundamental idea in economics. And yet there is no shortage of politicians, pundits, policy wonks, and members of the public who insist that raising the price of labor will not have the effect of reducing the demand for workers. In his 2014 State of the Union address, for example, President Barack Obama called on Congress to raise the national minimum wage from $7.25 to $10.10 an hour. He argued that increasing the minimum wage would "grow the economy for everyone" by giving "businesses customers with more spending money."

A January 2015 working paper by two economists, Robert Pollin and Jeannette Wicks-Lim of the Political Economy Research Institute at the University of Massachusetts Amherst, claims that raising the minimum wage of fast food workers to $15 per hour over a four-year transition period would not necessarily result in "shedding jobs." The two acknowledge that "raising the price of anything will reduce demand for that thing, all else equal." But they believe they've found a way to "relax" the all-else-being-equal part, at least as far as the wages of fast food workers go. Pollin and Wicks-Lim argue that "the fast-food industry could fully absorb these wage bill increases through a combination of turnover reductions; trend increases in sales growth; and modest annual price increases over the four-year period." They further claim that a $15-per-hour minimum wage would not result in lower profits or the reallocation of funds away from other operations, such as marketing. Amazing.

Pollin and Wicks-Lim calculate that doubling the minimum wage for 2.5 million fast food workers would cost the industry an additional $33 billion annually. They further calculate that reduced turnover will lower costs by $5.2 billion annually and that three years of sales growth at 2.5 percent per year combined with price hikes at 3 percent per year will yield $30 billion in extra revenues.

Let's consider turnover first. Pollin and Wicks-Lim claim that an increased minimum wage will make employees less likely to leave their jobs, saving fast food companies money that can now go to pay higher wages. New York Times columnist Paul Krugman convincingly refuted this argument in his review of Pollin's 1998 book The Living Wage. Krugman wrote: "The obvious economist's reply is, if paying higher wages is such a good idea, why aren't companies doing it voluntarily?" (That question goes unaddressed in the current study.) Krugman continues, "But in any case there is a fundamental flaw in the argument: Surely the benefits of low turnover and high morale in your work force come not from paying a high wage, but from paying a high wage 'compared with other companies'-and that is precisely what mandating an increase in the minimum wage for all companies cannot accomplish." So scratch that $5.2 billion.

What about Pollin and Wicks-Lim's sales growth projections? Well, sales don't always grow. McDonald's reported a sales decrease of 1 percent in 2014. Some analysts think fast food sales may have already peaked in the United States.

But there's a deeper problem. In the absence of the higher minimum wage, employers would generally hire more workers to meet an increased demand for fast food. Boosting the minimum wage means that the revenues that would have otherwise been used to hire new workers is not available. The end result: fewer jobs created and more folks unemployed.

Pollin and Wicks-Lim acknowledge that raising the price means that people will eat fewer hamburgers and fries. They calculate that a 3 percent per year price increase results in a 1.5 percent per year decline in what sales would have been, which means that revenues would increase by 1.5 percent. Then they assume that the price increases won't affect the underlying 2.5 percent annual sales growth rate. With this statistical sleight of hand, Pollin and Wicks-Lim roughly generate enough revenues to cover the higher wages by calculating that a three-year increase in prices and three years of sales growth will net $10.6 billion and $19.8 billion, respectively. Adding these to the postulated turnover savings of $5.2 billion yields $35.6 billion, which handily covers the extra wage costs of $33 billion.

Meanwhile, two new studies by economists that are much better grounded in actual wage and employment data have just been published. Both find that in the real world, the law of demand still applies to labor.

In the first paper, published in the December 2014 issue of the Journal of Labor Research, Andrew Hanson of Marquette University and Zack Hawley of Texas Christian University analyzed how low-wage employment would be affected in each state by the imposition of the national $10.10 per hour minimum wage supported by President Obama. The Hanson/Hawley study takes into account how wages relate to the varying cost-of-living levels among the states. First they report the number of workers in a state who earn less than $10.10 per hour. Next they apply the widely agreed upon formula that for every 10 percent increase in wages there is a corresponding 1 to 2 percent decrease in demand for labor. They then straightforwardly estimate that boosting the federal minimum wage from $7.25 per hour to $10.10 per hour would result in the loss of between 550,000 and 1.5 million jobs.

The second study, authored by Jeffrey Clemens and Michael Wither of the University of California, San Diego and published by the National Bureau of Economic Research in December, parses how the actual increase of the federal minimum wage from $5.15 to $7.25 per hour between July 2007 and July 2009 affected employment rates. Using U.S. Census employment data, they focus specifically on how low-skilled workers fared when the minimum wage rose as the Great Recession proceeded. The authors compare what happened to the employment rates of such employees in states where they started out generally earning below the new minimum wage versus in states where low-skilled wages were already higher than $7.25. They refer to the first set of 27 states as being "bound" by the increase and the second set as being "unbound" by it.

The minimum wage, Clemens and Wither show, exacerbated unemployment. Their analysis starts in December 2006, when the employment-to-population ratio-defined as the portion of working-age Americans (ages to 16 to 64) who are in the labor market-stood at 63.4 percent. It ends in December 2012, when that ratio had dropped to 58.6 percent. They estimate that by the second year following the implementation of the higher minimum wage, the employment rates of low-skilled workers "had fallen by 6 percentage points more in bound than in unbound states."

In other words, job losses were considerably higher in states where unskilled workers had been earning less than the new minimum and employers were now forced to pay more. Overall, the authors estimate that the minimum wage increase "reduced the employment-to-population ratio of working age adults by 0.7 percentage points." Stated otherwise, not raising the minimum wage would have boosted the 2012 employment-to-population ratio from 58.6 to 59.3, which implies that we actually had 1.4 million fewer jobs than we otherwise would have had.

The conclusion is clear. Defying the law of demand will end up harming lots of the people minimum wage proponents aim to help.