Fast food workers struck earlier this week over their low wages, agitating for a hike to $15 an hour. This has led to a lot of nonsense being talked about minimum wages. From the right I heard that minimum wages lead to massive disemployment, and McDonalds can’t raise prices because, if it could, it would have already. From the left, I saw grave accounting errors, and enthusiastic overstatement of the “new consensus” on the minimum wage. So it seems worth reviewing a few things:
- Small minimum wage hikes don’t seem to produce large, immediate declines in employment. While there’s good research showing declines in employment after the minimum wage is raised, and other good research showing no decline, the consensus is that you do not see a huge and immediate decline in employment if you raise the minimum wage by 50 cents or so.
- Companies are not rationally maximizing machines that automatically find the best, perfect price for their products. Arguments that take the form of “If it’s possible to … then why haven’t companies done it already?” are silly. The answer may be “because they are stupid” or more charitably “because they are missing an opportunity.”
- · Increasing wages can lead to greater productivity. Firms that pay higher wages can enjoy less turnover and better work effort from their employees, as well as attracting a higher class of employee -- a concept that economists call “efficiency wages.”
- · Some jobs can’t be automated. McDonalds can bring in a machine to take your order, but for the foreseeable future, a human will be cleaning the toilets.
Does that mean that McDonalds could give in to worker demands with only a paltry impact on profits or sales? Not so fast, friend. There are some other considerations.
- · Size matters. Jeremy Stahl at Slate argues that the costs would be small -- based on some extremely dubious calculations, and the aforementioned non-dubious research on the minimum wage that seems to show that at least in some cases, you can raise the minimum wage without much disemployment effect. We’ll get to the dubious math in a moment. The bigger problem is that the completely reputable studies that Stahl refers to deal with comparatively small increases over fairly short time frames. Doubling the minimum wage is a whole different story. So is raising it by about 50 percent, to $10.50 an hour, which is what the dubious-math crew wants. That’s not to say that the low-wage labor market would completely tank if we did raise the minimum wage that high. It’s to say that we don’t know what would happen, because we haven’t studied it.
- · Timing matters. We’re now in a period when labor markets are weak, inflation is low, and the sort of people who shop at McDonalds are very price-sensitive. This is going to alter the effects of a big wage increase. When times are flush, or inflation is high, it’s easier to get consumers to ignore a big price hike. When consumers are very value-oriented, a price increase is likely to cost you sales. You may think that a $1.00 increase in the price of a Big Mac sounds trivial, but there are families to whom eating out at McDonalds is a bit of a splurge, something they look forward to all week .. and those families are a core market for McDonalds. Affluent, college educated people who work in journalism and public policy work are not. A 25 percent increase in the cost of a meal may not alter your behavior, but it might well alter the behavior of McDonalds’ customers.
- · A big hike means big cost increases for industries that employ a lot of minimum-wage workers. The dubious math referenced by Stahl comes from a letter signed by a bunch of left-wing economists, who say that fast-food restaurants “are likely to see their overall business costs increase by only about 2.7 percent from a rise today to a $10.50 federal minimum wage.” I have no idea where they’re getting that number, but it seems wildly unlikely to me, given that the federal minimum wage is currently $7.25 an hour, so an increase to $10.50 represents a 44 percent increase in the cost of their minimum wage and near-minimum-wage workers. With payroll at 25 percent of revenue, I get a cost increase on the order of 10 percent, not 2.7 percent. (And that labor figure is what you get from reading forums for franchise owners, not just from industry spokesmen, who sometimes exaggerate a bit for poetic effect). A 100 percent increase in the minimum wage would mean a cost hike of around 25 percent for those businesses, which is a very big deal. To be sure, not all workers make minimum wage, but there are ripple effects -- if your line workers are suddenly making what your assistant managers used to make, you have to increase the managers’ salaries too. I’m going to go out on a limb here and guess that approximately 0 percent of the economists who signed that letter tried to work that calculation themselves, because getting to 2.7 percent almost certainly required some fairly heroic (read: dodgy) assumptions.
- · It can’t all come out of profits. Very few fast-food franchises have profit margins of 25 percent. Ten percent is considered pretty decent. Low-wage industries are generally characterized by fierce price competition and value-conscious consumers who won’t pay extra for service. To be sure, they’re not going broke, either; franchises don’t have the razor-thin margins of say, a grocery store. But they’re not big enough to absorb a huge hike in their labor cost -- especially since the franchisees usually have to live off those profits.
- · Companies aren’t all-knowing value maximizers, but they probably know more about their pricing power than a random person on the Internet. Currently, McDonalds is struggling to get customers to move up to a higher value-point; it would like to move people off the value menu, but keeps getting forced back to an ultra-value strategy. That tells you something about its ability to simply pass on price increases. And McDonalds is more profitable than most of its competition.
- · Time frame matters. All the studies of the minimum wage are relatively short-term. They have to be; over long periods, too many other factors could swamp your results. Restaurants will go out of business, or cut back, because tastes have changed, neighborhood demographics have shifted, or the owner developed a killer cocaine habit that drove the business into the ground. So the studies tend to look at restaurants right before and right after the change. We don’t know what happens five years down the road. And this matters quite a lot. A restaurant represents a large fixed investment in equipment and interior décor that can’t be changed on a dime. Over longer time frames, however, higher labor costs may change the math of opening a new restaurant … or installing a machine that takes orders, and another that cooks the burgers. Those things will not show up in studies of the minimum wage. But they will suppress minimum-wage employment, just the same.
- · Most of the efficiency wage arguments you read fundamentally misunderstand how efficiency wages work. For example, you periodically see people advocating for a national minimum wage because “Costco shows it’s possible.” Or they drag out the old saw about Henry Ford’s $5 a day wage. Raise the wage and people will work so much harder and better that you’ll still be awesomely profitable. One of my favorite sociologists has aptly characterized this argument thusly: “Why, Dorothy, you’ve had the power to solve inequality all along, just click your heels three times and say 'Efficiency wage, efficiency wage, efficiency wage.' ” Sadly, this is not how efficiency wages work. Paying a high wage can reduce turnover, get you better quality workers, and induce those workers to work harder, smarter, better. But there’s a catch: This only works if your wage is higher than the wage for other, similar jobs. If your job is unusually well paid for the type of work, people are going to be keenly interested in keeping that job. They’ll be extra nice to the customers, extra nice to their co-workers, and extra diligent about making sure the bathrooms are clean. If everyone raises the minimum wage, however, there’s no incentive to work extra-hard to keep your job. Efficiency wage models are not about the level but on the gap between you and your labor market competition; they rely on a sharp differential between your firm and other firms, or maybe your city and other neighboring places. Get rid of the differential, and you get rid of the benefits for the employer of paying a higher wage. Oh, perhaps there’s some sort of social solidarity that keeps workers working harder because they feel more respected. But we don’t have either a model, or data, to back that theory up.
- · It isn’t true that paying workers at the bottom a lot somehow magically makes the economy grow. I quote Scott Winship of the Brookings Institution: “Compensation outpaced productivity growth during the mid-20th Century (in the peak years of unionism), and recent decades have seen a correction in which productivity levels have had to catch up to pay. On balance, the numbers still favor workers: If President William McKinley had been told in 1900 how much higher productivity would be in 2013 and had used that information to guess how much higher hourly compensation would be, his prediction would have been too low.” Yet President McKinley was not presiding over some sort of stagnant dystopia; the economy grew pretty fast in the late 19th Century (though to be fair, it also had some nasty recessions). It may be true that paying workers at the bottom a lot somehow makes the economy grow faster, but this is not an Established Scientific Fact; it’s a supposition, often based on the observer eyeballing the minimum wage in 1968 and wildly overextrapolating. This sort of effect is extremely hard to establish definitively, and so far, it hasn’t been done.
I’ve been doing this too long to think that any of this is going to change anyone’s opinion of the minimum wage. But at least we don’t have to say silly things while we’re debating it.