Exploring The Drivers of Income Inequality

Rising income inequality is an increasingly serious problem in many countries—and MIT Sloan Professor Nathan Wilmers wants to understand its causes. In his research, Wilmers, who is an Assistant Professor of Work and Organization Studies at the MIT Sloan School of Management, has identified some surprising factors that contribute to income inequality—from the power of big corporate buyers to how tasks get allocated among colleagues. Wilmers, a member of the faculty steering committee for the Good Companies, Good Jobs Initiative at MIT Sloan, recently sat down with Martha Mangelsdorf, the Initiative’s Director of Strategic Communications, to explain some of his recent findings and their implications. What follows is an edited and condensed version of their conversation.

Good Companies, Good Jobs Initiative:  You did an interesting research project about how corporate buyers influence wage inequality in the U.S. Tell us a little bit about that study.

Wilmers: I was thinking about how pay has basically stagnated for U.S. workers without a college degree since the 1970s. A lot of the jobs that used to be relatively well-paying positions for workers without a college degree are concentrated in manufacturing- and logistics-related industries. And I read some anecdotal evidence that there had been some consolidation among the large buyers that these manufacturing and logistics companies sell to.

I was interested in testing whether that consolidation affected working conditions at suppliers and could be part of the reason why middle-skill workers had seen very little wage growth since the 1970s. The difficulty with testing this is it's hard to figure out what these networks are between buyers and suppliers; that is typically proprietary information. But for a subset of supplier companies that are publicly traded, there was an accounting regulation that went into force in the late 1970s that required companies to disclose to investors any big buyers that accounted for 10% or more of their revenue.

Using that data, I found that, over time, as a company becomes more reliant on sales to big buyers, its wages and labor costs tend to decline. I also looked at what happens when big buyers merged together. I found that when big buyers consolidate, wages at their manufacturing and logistics suppliers go down. And this phenomenon—the increasing power of large buyers—explains around 10% of wage stagnation since the late seventies.

Ten percent is a fairly significant chunk. Is that wage stagnation in the U.S. economy in general, or wage stagnation for lower-wage workers?

Wilmers: It's overall wage stagnation in all nonfinancial firms. If you narrowed in to manufacturing and logistics industries that are most affected by this, it would probably be a bit higher. And I found in another part of the analysis that the negative wage effects at suppliers are concentrated on wages for workers without a college degree.

I certainly remember hearing anecdotally about big, price-oriented companies like Walmart trying to drive down prices at suppliers.

Wilmers: That’s right. Part of the story is the rise of big retailers like Walmart or, more recently, Amazon, that sit atop huge supply chains connecting suppliers to consumers—and are able to put a lot of pressure on suppliers for lower prices. The other set of buyers that increased a lot in prominence in my data during this period are tech and information companies. There the story is different. It's more about outsourcing. A lot of large tech companies employ a core of highly skilled workers, like managers and computer programmers, and then outsource things like janitorial and food service work to other organizations. That’s a buyer-supplier relationship, but it involves outsourcing labor.

As we're talking, I think about the Coalition of Immokalee Workers, a group that’s had some success in improving working conditions for workers on Florida tomato farms. They negotiate with buyers like Whole Foods or Taco Bell for better tomato prices that enable better wages for farm workers—rather than just focusing on negotiating with farmers. It sounds like your research suggests that's probably a smart strategy.

Wilmers: Yes, I think the Coalition of Immokalee Workers is a great example. In one industry on a small scale, they essentially realized, "Wait, it's not that these farmers we're working directly for have big profit margins and can pay us more. We need to go to the big buyers themselves to get some bargaining leverage.”

Another example globally is the anti-sweatshop movement, which doesn't focus on bad employers at the level of, say, a particular factory owner in Honduras. It focuses on big brands that contract with these firms.  There's been an attempt to build up supply chain codes of conduct, but it has largely been focused internationally.  I think these findings suggest there could also be a role in domestic sourcing for some sort of good jobs supply chain code of conduct for responsible buyers.            

Another implication of these findings relates to antitrust enforcement and competition policy, which in general is directed towards seller concentration rather than buyer concentration and is focused on consumers rather than workers. Regulators should consider outcomes for workers when assessing, say, a merger between two big retailers or two big buyers.

Interesting. Now let's switch gears a bit. You are working on another study that has intriguing findings about a different dynamic, one that affects income inequality on a much more micro level—within a particular workplace. Tell us a little bit about this research.

Wilmers: I started this project because sociologists who study organizations talk a lot about organizational structure and management practices without talking much about work itself and what tasks people do day to day at work. Insofar as we have studied tasks and inequality, it's mostly been about big-picture compositional changes in the kinds of work that people do over time, such as a decline in routine tasks and less manufacturing. I was interested in whether you can better understand wage inequality within organizations, among coworkers, by thinking about how tasks are allocated among them. Assuming that there's a set of tasks that a workplace needs to get done, does arranging those tasks differently across the workers that you're employing affect inequality?

Specifically I tested two things. One is the wage effect of having more or less multitasking. If you've got two tasks that need to get done—let's say computer programming and administrative work—and you employed two workers, you could have both of them doing both tasks. Or you could have one worker focusing on administrative work and the other doing programming. There has been some previous research showing that multitasking or job rotation tends to increase pay for workers because they're more versatile and have to have a larger number of skills. So that's the first thing I looked at.

But I also looked at whether you're the only person in your workplace doing what you do. I call it job turf when you're the sole person in your workplace doing a certain task. My prediction was that if you have job turf, that's going to increase your bargaining power vis-à-vis an employer, because you're the only person in the workplace who knows how to do that task.

What I found in my research is that as multitasking increases, wages tend to increase. And as job turf increases, pay also increases for workers. Interestingly, there's a relationship between these two aspects of task assignment: If your coworkers’ tasks get more specialized and singular, you tend to get more job turf.

Let me see if I’m understanding how that works: Suppose initially, you and a coworker are doing a variety of bookkeeping tasks that include payroll-related tasks. But the company grows, and suddenly your coworker is so busy doing just bookkeeping tasks related to payroll that you end up doing a broader range of bookkeeping tasks.

Wilmers: That’s right. And that tends to get you more job turf, because the person who's just doing payroll used to also do other things that now you're the only one doing. The interesting upshot of this is that it means that as multitasking is decreasing among some workers, which is going to lower their wages, job turf is increasing among other workers, which is going to raise their wages. So that's why you end up with this within-workplace inequality effect among coworkers.

This finding has career management implications for people at all levels in an organization.  How could an employee apply this finding to his or her own work situation?

Wilmers: It suggests that, on the one hand, it makes sense to try to do lots of different things so that you're more versatile and building more skills. But on the other hand, you want to be doing things that other people in your workplace aren't already doing. You need to make yourself distinctive from your coworkers.

Right. And I assume that if you develop skills in an area that over time becomes less valuable to your employer, that probably won’t increase your wages. So, as an employee, I imagine you want to figure out: Are there tasks that I can put myself forward for that are valuable to the organization but nobody else is doing?

Wilmers: That's right. If you can find some kind of niche distinctive thing to be doing, that job turf could increase your bargaining power.

It strikes me that there's another implication of this research—which is that you really don't want to specialize in a task that other people in your workplace also do. I'm guessing that if you find yourself stuck just doing one task that other people in the company also can do, you should seek to get either something more distinctive to do or a greater variety of tasks.

Wilmers: Yes, sure.  

What's next in your research?

Wilmers: My main ongoing project involves trying to figure out how much of inequality is due to rising pay differences between occupations vs. differences between workplaces.  So far, we're finding that there's been an increase in workplaces that pay a lot and employ a lot of high-paying occupations and also an increase in workplaces that don't pay much and employ relatively lower-skilled occupations. But it’s too early to say much about that study; we can talk about that it at a later time.

We've covered a lot in a relatively short interview. Are there important points that I've missed?

Wilmers: A fair question for you to ask would be, across these different kinds of things that affect inequality, which are the most important and explain most of the trend. Unfortunately, I can't answer that question yet. Going forward, hopefully I'll be able to do more to sort out which factors contributing to income inequality are most important.