Preface by the author, January 2nd 2019:
I wrote this post (which appears below in its original form) nearly five years ago, while a faculty member at Duke University. I had not yet received a job offer at the University of Washington, let alone signed up to join a multi-disciplinary mixed-methods evaluation of Seattle's groundbreaking minimum wage ordinance. The ordinance hadn't even been passed.
Having worked on this evaluation for more than four years, I should emphasize that there are certain arguments made here that I now consider to be wrong, or off-point.
Firstly, I argued that the minimum wage was paid for largely by a tax on food. I'm now a co-author of a peer-reviewed analysis showing no impact of Seattle's minimum wage on grocery prices. Our team's analysis of restaurant prices has found little to no evidence of increases. While good news for consumers, the absence of price impacts may reflect the ease with which businesses have found other methods of compensating for higher labor costs, including staffing cutbacks.
Second, I argued -- as many have before -- that the minimum wage poorly targets assistance, helping not just adults supporting themselves but teenagers who do not. Our research team has since uncovered evidence that in fact the income gains associated with Seattle's minimum wage have been concentrated among workers logging more hours and who have spent more time in the labor force. Put differently, gains appear to accrue to the adult workers and families whom advocates aim to help by raising the wage.
Our evidence also suggests that teenagers and others with little to no labor market experience at best break even and at worst find it considerably harder to land a job. This, in turn, confirms arguments made by minimum wage opponents. It is not contradictory for our research to support the arguments of both advocates and opponents as they largely confine their arguments to separate segments of the low-wage workforce.
Third, although it wasn't necessary to conduct our research to say so, a policy intervention that provides more income to families in poverty without actually boosting them above the poverty line, or that increases the income of families just above that line, does good.
In all, despite the fact that the work of our UW research team has been held up as supporting an anti-minimum wage agenda, I come away from this work more inclined to support reasonable minimum wage increases. There are some who share this inclination while arguing that the minimum wage causes no job loss, that employers just can't think up ways to cut back on labor as it becomes more expensive to them, or that the minimum wage fights collusion or market power among low-wage employers. I encourage readers to contemplate this proposition the next time they are busing their own dishes at a quick-service restaurant, queueing for service at a retail establishment, or providing that service themselves.
My read of the evidence: the minimum wage does make it harder for some to find work, and results in lost hours for at least some of those who continue to be employed, but these losses are concentrated among a subset of the low-wage workforce who will be fine in the long run: young workers on an upwardly mobile trajectory. They are working not to support themselves, but just to gain experience. As a consequence, their "reservation wage" is nearly zero. Indeed, many young workers engage in volunteer service revealing this directly. Workers aiming to support themselves must compete against them in the labor market. The minimum wage is a regulation that effectively declares this competition unfair.
Arguments that a higher minimum wage poses a severe threat to business, according to research by members of our team, are exaggerated. As many have pointed out, business continues to expand in Seattle. But our research team has also found that new businesses tend to rely on less labor than outgoing businesses. Business is doing fine because they've found ways to offset higher wage costs, and one offsetting mechanism is a reduced reliance on labor.
Arguments that a higher minimum wage has caused an economic boom in Seattle are not directly addressed by our research. But as many critics stated in response to our work, there's another more obvious source of Seattle's boom, namely Amazon's recent hiring of tens of thousands of highly paid employees at their headquarters.
This hiring binge, coupled with the city's failure to expand its housing stock, has made Seattle one of the nation's most expensive cities. Those who would attempt to support themselves on the basis of low-wage work in this city are finding that $15/hour isn't enough to make ends meet. Some of them are winding up homeless, others are moving away. The high cost of living in Seattle means that with or without the city's minimum wage ordinance, local businesses would have faced a reality of scarce labor in lower-paid jobs.
As other jurisdictions look to Seattle's experience, this is the greatest caution: economists of all persuasions would agree that a minimum wage set below the market-clearing level is of no consequence whatsoever. Having recently seen an iconic Seattle burger chain advertise starting wages of $18/hour in a city where the highest minimum is now $16, it is reasonable to conclude that as high as Seattle's minimum is, there are actually very few jobs where it is binding.
In the course of conducting this research I've had many opportunities to speak with labor advocates who readily admit that a higher minimum wage isn't a one-size fits all solution to working poverty. Among other things, 90% of the low-wage work force falls short of working sustained full-time hours. There are copious corporate profits in the modern American economy, but they are disproportionately accruing to businesses that employ no low-wage workers; in some cases these businesses are actively developing technologies that will threaten or eliminate a number of low-wage job categories.
So advocate as you see fit, but realize that we'll need more tools than just a minimum wage increase to achieve a society where every person can support themselves through work.
My original February 2014 post follows in unedited form.
Suppose I suggested that the government adopt an anti-poverty program with the following characteristics:
- It would spend $6 to deliver $1 to a family in poverty.
- For every adult worker in poverty assisted, it would assist two teenagers.
- Some indeterminate number of vulnerable workers would be at risk of losing their jobs.
- It would be paid for largely with a tax on food.
- It would not guarantee that a person working full-time year-round would earn income sufficient to place them above the poverty line.
This is a basic description of how the minimum wage operates. This isn't the right-wing spin on it either, points 1-3 were derived directly from a New York Times editorial this morning. There's plenty of controversy regarding the CBO's highly publicized but very rough estimates of how many jobs would disappear following a minimum wage increase, but one could glibly assume no negative effect on employment and still come to the conclusion that the minimum wage is a highly symbolic but dreadfully ineffective way to lift families out of poverty.
The minimum wage is too blunt an instrument to effectively fight poverty
The positive way to think about the minimum wage is that it attempts to ensure that any person trying to raise a family by being a full-time worker will earn enough to get by. By the CBO estimates, some 16 million workers would be affected by an increase of the minimum wage to $10.10 per hour, and some 900,000 would be lifted from poverty. Thus if the goal of the minimum wage increase were to lift families from poverty, estimates suggest it will have a 6% success rate. Flip that around, and you've got a 94% failure rate. And this is assuming no adverse impact on employment whatsoever.
In the 94% of cases where workers receive wage increases but do not experience a lift from poverty, there are two basic things going on. In most cases, the families benefitting from the wage increase aren't lifted from poverty because they weren't in poverty in the first place. Minimum wage workers are often secondary or tertiary workers in their household -- including teenagers. In some cases, however, the increase in the minimum wage would not be sufficient to escape poverty. For a full-time worker raising a family of 4, the income earned from working 2,000 hours per year at $10.10 per hour would not be sufficient to rise above the federal poverty line. Many minimum wage workers are part-time workers; even if their job continues to exist after the increase there is no guarantee they'll be able to work the same number of hours.
The minimum wage is largely a tax on food
Minimum wage workers are found across a wide range of industries, but the highest concentrations happen to be in industries tied to the production of food, based on data from the 2010 American Community Survey. More than one-tenth of minimum wage workers work in the restaurant or food service industry. Also represented in the list of top 5 industries employing minimum wage workers are grocery stores, discount and department stores, and K-12 education -- it isn't the teachers making minimum wage, but the cafeteria workers and related staff. Crop production sits just outside the top five.
A truly progressive antipoverty policy would transfer resources from the wealthy to the deserving poor. Presumably the goal of a minimum wage increase would be to transfer resources from the highly paid executives and wealthy shareholders of major corporations to their low-paid workers. To understand why this just can't work out in practice, consider the case of Wal-Mart.
Wal-Mart's most recent annual report shows that the company paid shareholders about $5.4 billion in dividends, and paid their top executives somewhere around $60 million. Suppose we zeroed out those numbers -- forced the executives to forfeit 100% of their pay, and shareholders 100% of their dividends -- and transfered the money to the company's 2.1 million domestic employees. We'd have enough to give each of them about $2,583 per year. For a full time worker, this would amount to a raise of roughly $1.29 per hour.
Wiping out the shareholders and top executives, in other words, would be sufficient to fund less than half of the proposed $2.85 increase in the federal minimum wage. Granted, the average Wal-Mart employee already earns between $12 and $13 per hour, but the basic message is clear. Minimum wage work occurs largely in low-margin industries, where there just aren't a whole lot of profits to be plundered. The real fat cats of the economy, working in knowledge industries or on Wall Street, don't employ a whole lot of minimum-wage workers.
To pay for the wage increase, then, costs would have to be passed along to consumers in the form of higher prices. Given the heavy reliance on minimum wage work at nearly all stages of the food supply chain, the higher prices would be most apparent in a family's food budget. Ask yourself, what type of family spends the highest share of its income on food? Not the wealthy.
There is a better way
The intent of the minimum wage is to raise the payoff from work for society's most vulnerable people. It effectively asks business owners to cover the costs, and requires them to spend a large amount of extra money on raising the payoff from work for a larger group of citizens who are not society's most vulnerable people. These business owners, who by the nature of their business are already serving many vulnerable people as their clientele, cover the costs in part by passing them on to their customers. They rob Paul to pay Paul.
If our goal as a society is to ensure that no person who devotes themselves to full-time work should find themselves living below the poverty line, there is an alternative strategy that is simultaneously more efficient and more progressive. It's the Earned Income Tax Credit.
I am an economist by training, so perhaps it is no surprise that I think the EITC beats the minimum wage. But the selling points of the EITC bear repeating, particularly since the drumbeat for expanding the EITC as an alternative to a minimum wage increase is not exactly deafening.
The EITC effectively multiplies earnings for families that work but don't make much money by doing so. It delivers significant amounts of cash to single parents raising a family on the basis of low-wage work, but not a penny to the teenage child of a high-income family delivering pizza for a bit of spending money.
The EITC doesn't ask business owners to bear the cost of society's goal alone, but rather spreads the burden through the full government system of taxation. If you want the wolves of Wall Street to pay for our social investment in the lives of the vulnerable, the EITC will do it but the minimum wage won't.
And from a pragmatic perspective, the will in Congress to pass an EITC expansion this year might actually exist, given the bipartisan focus on inequality this year. Do you really think the House is going to pass a minimum wage increase anytime soon?
There are criticisms of the EITC. Some employers might use the EITC as an excuse to cut wages, but coupled with the existing state and federal minimum wages it isn't really possible to do that for the workers we're ostensibly trying to help. Moreover, even if we accept estimates that only 73 cents of every dollar spent on the EITC lands in the hands of a deserving family, that surely beats the ratio we'd get with a higher minimum wage.
The minimum wage, in short, makes for good symbolism but bad policy.
[Note: the original Wal-Mart numbers I had in this post were taken from some poorly annotated figures I jotted down a few weeks ago. Walmart reports paying $1.59 in dividends per share in fiscal 2013, with a total of about 3.374 billion shares outstanding. So that's returing quite a bit less than the $100 billion I originally cited, which would have made for a much greater windfall per worker if redistributed.]
Did you read the source you're citing?
Here's an excerpt from page 3 of Wal-Mart's most recent annual report:
"Last year, Walmart delivered a really good financial performance.
Our earnings per share increased 10.6 percent to $5.02. With the addition of $22 billion in net sales, we are now a $466 billion company. Our operating income was up
4.7 percent to $27.8 billion. We also grew free cash flow 18.1 percent to $12.7 billion. All of this enabled our company to return $13 billion to shareholders in dividends and share repurchases. In fact, Walmart shareholders enjoyed the best overall return in stock performance and dividends
for our company this year than in more than a decade."
Moreover, on page 2 in a large graphic Wal-Mart states more than $60 billion were returned to shareholders through dividends and share repurchases during the past five years (2009-2013).
The economy is recovering and expected to grow at a higher rate during the next five years. I'm not defending the minimum wage vs. EITC as the better policy, but just pointing out the mistakes in your analysis. I think if you're going to cite a source, you should at least read it.
If you used the correct figures, your analysis would show that shareholders can easily afford to cover the $2.85 proposed increase in the federal minimum wage.
Posted by: Alex | 02/23/2014 at 12:54 PM
Another mistake I found:
In 2013 Wal-Mart made four quarterly dividend payments of $0.47/share, for a total of $1.88. You used $1.59, which was the total from 2012.
Posted by: Alex | 02/23/2014 at 01:55 PM
Thanks for the comment, Alex, but I stand by my post. I said you could wipe out executive compensation and dividends and not have enough to cover even half the cost of raising wages $2.85, and there's nothing in your comment to contradict that statement.
Now, it is true that Walmart's profits exceed the amount returned to shareholders as dividends. Walmart, like many corporations, returns some money to shareholders by buying back shares (the lion's share of the $13 billion/60 billion over 5 years statistics you cite). While both dividends and repurchases put money in shareholders' hands, a repurchase requires the shareholder to give up something in return - their stock. Remaining shareholders may benefit to the extent that the reduced number of shares outstanding gives them a larger stake in future earnings, but that's conditional on there being future earnings in the first place.
Even if we count repurchases the same as dividends, given WMT's current market cap stockholders have been treated to a return between 5 and 6% over the past year. That's not bad, but it's nothing like the deals venture capitalists get when a million dollar stake turns into a billion dollar payout within a couple of years.
It is also true that Walmart's profits are not distributed 100% to shareholders. Some of those profits are used to do things like build more stores (hence creating more employment). Walmart hoards some cash, but less than one-tenth the amount a company like Microsoft has on hand. Further, Walmart holds nearly $8 in debt for every $1 in cash on hand.
So, depending on how you wave your hands about the accounting, you could potentially pay for a $2.85 wage increase if you went beyond wiping out the dividends and took away the money the company had been intending to use to expand their business and/or pay down their debt. Were Walmart to do that they'd have a hard time raising more cash in either the equity or debt markets. This leaves them one place to raise cash -- by raising their prices.
Posted by: Jake Vigdor | 02/24/2014 at 07:28 AM
And Alex, regarding the second "mistake," my apologies but I used the FISCAL 2013 annual report, which is the most recent available.
Posted by: Jake Vigdor | 02/24/2014 at 07:30 AM
Kudos for courteously handling a hostile criticism... and I think I agree with all of your points, upon reflection.
As for your discussion of the EICT versus minimum wage... I note that your argument that the EITC won't incentivize cutting wages /relies/ upon an existing minimum wage.
Thanks for the points to ponder.
Posted by: Joe Marfice | 02/26/2014 at 02:08 PM