Aaron Yelowitz joined host Dan Loney and University of Pennsylvania sociologist Jerry A. Jacobs to discuss paid-sick-leave requirements for federal contractors, workplace externalities, employee absenteeism, differences across industries and countries, and proposed collection of gender-pay information.
Participants
- Dan Loney: Host
- Aaron Yelowitz: Associate Professor of Economics, University of Kentucky
- Jerry A. Jacobs: Professor of Sociology, University of Pennsylvania; affiliate of the Wharton Center for Human Resources; executive director of the Work and Family Researchers Network
Related research
Transcript
00:00:03 — Knowledge@Wharton announcer: This podcast is brought to you by Knowledge@Wharton. For more information, please visit knowledge.wharton.upenn.edu.
00:00:19 — Dan Loney: It’s an important move by the government, and one that shows them using their influence for good. Companies that have federal contracts will now be required to provide sick leave to their workers, who will be able to accrue up to seven sick days a year. In addition, the government is asking for information from these contractors to determine the pay levels of both men and women, as a way to perhaps try to eliminate the issue of gender pay in the United States.
To discuss this move, we’re joined here in the studio by Jerry Jacobs, who is a professor of sociology here at the University of Pennsylvania. He’s also an affiliate at the Wharton Center for Human Resources and executive director of the Work and Family Researchers Network, which is an international and interdisciplinary society of researchers and scholars that takes a look at the intersection of work experiences and family life.
Also joining us by phone is Aaron Yelowitz, who is an assistant professor of economics at the University of Kentucky. Jerry, nice to meet you. Thanks for coming in.
Jerry A. Jacobs: Pleasure to be here. Thank you.
Dan Loney: Aaron, great to have you on the phone with us.
00:01:23 — Aaron Yelowitz: Nice to be here. I’m an associate professor here at UK.
00:01:26 — Dan Loney: Exactly. Great to have you. I guess the first thing to jump on, Aaron, is the designation by the government actually making this move. As I said at the top, this is seemingly a situation of the government using its power through government contractors to show a benefit for employees who may not have been receiving benefits over the last couple of decades.
00:01:54 — Aaron Yelowitz: A little background on this, Dan: the president, back in September 2015, issued Executive Order 13706, which basically extended this paid sick leave to federal contractors and their workers. The estimates of how many workers are actually affected have varied a bit. When the president signed this executive order back in September 2015, he said something like 300,000 workers would be affected. That number was ratcheted up to 437,000 back in February 2016, then about 600,000 new workers, and altogether the final ruling that I think was just put out today suggested something like 1.15 million workers would be affected.
That means some workers—about half a million or so—had some paid sick leave, but they didn’t have seven days per year. So, basically, this is giving perhaps 600,000 workers paid sick leave when they didn’t have it before and increasing the number of days for about half a million workers or so.
00:03:06 — Dan Loney: The importance of this, Jerry, in your mind, is what?
00:03:10 — Jerry A. Jacobs: This is a step that the administration is taking because it can’t get paid sick leave passed as legislation on a national basis. We’re talking about a relatively small number of workers here. I don’t know if a million is a small number or a large number—it depends on your point of view. From the point of view of the American economy, it’s a relatively small number, but that’s what President Obama is able to do within his purview in terms of executive orders and decisions about the kinds of contracts that the federal government will undertake.
About a third of American workers are not covered by paid sick leave. Whenever I talk with people about this issue, the first thing I like to bring up is: when you go to a restaurant, do you want to be served by somebody who has to come to work sick? The reason I use that is that it not only highlights the “ick factor” in terms of sickness, but it raises all the relevant issues. Workers who come to work sick get their coworkers sick, so it’s not good for the business. They get customers sick, so it’s not good for society. They might be leaving a sick child at home, and it’s not good for the family. There are externalities, in economic terms, on all fronts. That’s why I think it’s appropriate for a government to set a floor in terms of what the paid-sick-leave policy ought to be.
00:04:40 — Dan Loney: And also the impact on the food itself that they would be serving in said restaurant, just as one example.
00:04:46 — Jerry A. Jacobs: Yes. Now, it turns out that I don’t think most of the employees covered by Obama’s new order—by the Department of Labor’s new policy—are in the food-services industry.
00:04:59 — Dan Loney: Right.
Jerry A. Jacobs: But nationally, it turns out that’s a big point. That’s a big industry affected by this whole issue.
00:05:07 — Aaron Yelowitz: In economics, the kind of question we’d ask about paid sick leave more broadly—which Jerry was alluding to with the substantial number of low-wage workers throughout the U.S. economy who don’t have paid sick leave—is this question of externalities. What externalities mean is: do you do something that causes harm—in this context, getting someone else sick? Sometimes we talk about positive externalities, which aren’t really relevant here.
If you narrow the industries where we think the spread of contagious diseases is most important, Jerry appropriately brings up food services—the “ick factor.” Companies like Chipotle have voluntarily decided to offer paid sick leave, in part to reduce the spread of diseases such as norovirus, which has been traced to restaurants. Other industries that we might think are affected include health care, because workers there interact with so many unhealthy people that the spread of disease seems paramount, and childcare or education, where there can be a large spread of disease. If you have kids, you probably would agree with that.
Jerry brings up this really interesting point that the spread of disease is a very important idea. It’s probably concentrated in some industries and perhaps not as much in other industries. The computer programmer in a cubicle is perhaps less likely to spread disease than the restaurant worker who’s coughing on a hamburger.
00:06:48 — Dan Loney: 844-942-7866 is the number to give us a call. You’re more than welcome to join us. We’re talking about the move by the government to have its contractors provide paid sick leave for employees. If you can’t get to your phone, send us a comment via Twitter, either @BizRadio111 or my account, @DanLoney21.
Now, it seems like, Aaron, that the majority of the people who would benefit from this would be people contracted to do computer work, or it could be construction crews doing projects for the U.S. government. That’s probably the majority of the people who would be linked into this, correct?
00:07:32 — Aaron Yelowitz: Right. I think there are two things to break out here. There is this executive order, which, for lack of a better word, I’d say is a little bit symbolic. It basically takes a stand on where the current administration thinks labor policy should be. The idea that we’re going to spread much less disease is probably less likely in this federal-government-contracting case.
The bigger policy question is how this perhaps symbolic move relates to what’s going on on the ground in various cities and states. For example, California passed and implemented a paid-sick-leave mandate in 2015, which in a sense would give more paid sick leave to a wide swath of employees. There’s a bridge between this executive order and how it maps into the broader labor market.
00:08:35 — Dan Loney: Jerry?
Jerry A. Jacobs: Yes. We currently have a patchwork of legislation nationally. There are currently five states that have paid-sick-leave legislation and something like 30 or 36 cities and counties across the country. I think you can make an economic argument that we’d be better off with a single policy nationally, because lots of companies operate in more than one jurisdiction and have to pay attention to the variation in local laws. Given inaction on the federal level, lots of advocates in this area—whether in Philadelphia, San Francisco, Washington, D.C., or New York City—have been pushing forward because they figure that if you can’t get it nationally, let’s at least try to do it locally.
00:09:17 — Dan Loney: I would think there have to be many companies based here in the United States that would be considered multinational and are doing work in Europe, Asia, or wherever it might be. They almost have to run two systems because, as we’ve talked about on this show, paid sick leave in other parts of the world is much more advanced than it is here in the United States. Take your pick of companies: they may not have the same type of policy here in the United States as they would, say, in Europe. Correct?
00:09:54 — Aaron Yelowitz: Dan, you bring up a good point. I’ve looked at a fair bit of cross-country variation as well. One interesting comparison is the United States and Germany. Germany has six weeks of paid sick leave at 100 percent pay, and then for a very long time after that, you can get about 80 percent of salary.
One thing we ought to talk about at the same time that we talk about the spread of disease—the negative externality we just mentioned—is the very real concern about absenteeism. If you look at the number of days absent in the U.S. for all workers, it’s about four days or so. In the next-lowest country in some of these comparisons with European countries, the next-lowest number is something like 16 days, if I recall correctly. There are dramatic variations in the number of days that people are off from work.
What we have to ask is: are they off from work because, in a sense, we’re subsidizing them not to be at work, or is it because people are sick 16 days each year and they’re not there because of that? I thought those numbers were quite striking. Clearly the U.S. is far less generous, and that leads to two sides of the same coin: perhaps more spread of disease, but at the same time perhaps less absenteeism.
00:11:23 — Jerry A. Jacobs: Sure. The issue of abuse of paid sick leave by employees comes up all the time. It turns out that studies conducted after the introduction of paid sick leave in places such as San Francisco have found that employees hoard these sick days. They’re precious to them. These are low-wage workers, many of them working part time. They want to have a couple of paid sick days in the bank because they know they might need them.
A phrase often used in this area is “predictably unpredictable.” You know you’re going to get sick, but you don’t know when. We’re talking here about accumulating, based on how much time you’ve put in, up to seven days a year. On average, people use two or three of them. Germany is much more generous in this respect than in many other respects, but the experience in the United States so far has been that workers don’t use the full seven days. They use only two or three on average and bank them to have that insurance protection for themselves.
00:12:55 — Aaron Yelowitz: Let me jump in for a second. One of my fine collaborators here at the University of Kentucky, Tom Ahn, and I have a paper where we look at the U.S. As Jerry mentioned, there is a paucity of evidence in the U.S., in part because these city-level mandates are often so new. Only a handful of cities implemented paid-sick-leave mandates before, say, 2014. This is really quite new, and the data one would use to answer these questions are only starting to come online now.
What my colleague and I do is compare employees who work in different industries. It turns out that there is dramatically different variation in the offering of paid sick leave across industries. As you might imagine, industries that tend to have more of a public focus have very high rates of paid sick leave. Industries such as construction or food services tend to have quite a bit less.
We look at administrative workers—the kind of person who very well could be in one industry or another and has largely the same skill set. Because of the norms in that industry, they tend either to be offered or not offered paid sick leave, as well as a host of other benefits. We find that paid sick leave causes about an additional one or 1.2 days of absenteeism. There is some effect on absenteeism. I’m not going to claim that this is economically devastating or anything like that, but we do see people taking a little bit more time off when they have paid sick leave compared with when they don’t.
We also see in the surveys that people are able to report whether those days were moderate days or severe days. You’d imagine that moderate days are more discretionary, and that is where all the action is. If you’re so sick that you can’t get out of bed, we don’t actually see that paid sick leave matters much, because whether you had it or not, you’re basically going to call in and either take an unpaid day or, if you have paid sick leave, take a paid day.
One statistic I think is neat is that absenteeism seems to be highly concentrated. My coauthor and I find that 20 percent of workers take about 80 percent of days. The way I would interpret that is that there might be a segment that might abuse it, and then there is probably a fair amount that might not abuse it—or at least part of that segment is also probably just less healthy.
00:15:34 — Dan Loney: We’re joined on the phone by Aaron Yelowitz of the University of Kentucky and here in the studio by Jerry Jacobs of the University of Pennsylvania. Your comments are welcome at 844-WHARTON, 844-942-7866.
The other piece of this that obviously is going to draw a lot of attention deals with gender pay and the fact that the government would like data from these companies to say: how much are you paying male employees and how much are you paying female employees? It’s another topic that has been at the forefront of a lot of conversations and has been brought up in the presidential race as well. The data, I guess, become very important to move forward in the discussion on a greater scope for the country as a whole. Jerry?
00:16:18 — Jerry A. Jacobs: Again, the Obama administration has tried to make pay issues more transparent. They want workers to be able to know what their coworkers are getting paid. This is controversial. There’s some concern that this creates issues of competition and dissatisfaction in workplaces. There’s also a compliance-cost issue for companies. But as we move to a world where they’re increasingly relying on complex data systems to administer payrolls, it seems to me that for larger employers, compliance issues are really not terribly significant. If you have a payroll system that generates paychecks for employees, adding a line that passes that information along is not a big deal. The issue is always for smaller employers. I think it’s clear that there’s increasing reliance on computer systems for payroll administration across the board. Aaron?
00:17:25 — Aaron Yelowitz: I would think that the data-gathering part of the gender gap seems like, in the world we live in, it can be produced at hopefully not too much cost. The natural question is: imagine that there is a gap, which surely there is to some degree. Where is that arising from? If we collect information only on male and female, are we also collecting information on job titles, experience, and those sorts of things?
Without thinking about it too much, my one concern would be: are we collecting enough information to make apples-to-apples comparisons—comparing people with the same experience or job titles and then seeing whether there are gender differences based on that—or are we looking overall at just males versus females, where many other factors might also come into play?
00:18:25 — Dan Loney: It almost seems like, in some realms, it’s a taboo subject to want to gain that data, which is a little disappointing. It continues to be a topic brought up seemingly on a weekly basis in this country. Aaron, as you alluded to, the desire to collect the data seemingly in some cases is not there.
00:18:51 — Aaron Yelowitz: Let me take on a skeptic’s role for a second. We know there are class-action lawsuits all the time about things such as wage discrimination. One might be worried that if I publish something, I’m going to get sued by someone for something. That could be one reason why. It would seem to me, without knowing all that much about payroll systems, that it can’t be that hard to produce this stuff. The one question is: do people use the numbers out of context? That would be my biggest concern about all of it.
00:19:25 — Dan Loney: How would it be out of context?
00:19:27 — Aaron Yelowitz: For example, imagine Dan earns X dollars, Jerry earns Y dollars, and Aaron earns Z dollars, and we don’t account for differences in occupation, experience, and that kind of thing. With the three of us, obviously, we couldn’t learn about gender discrimination.
00:19:47 — Dan Loney: Right. But when all is said and done, are we controlling for those other factors? That’s what I’m talking about. Great to have you both on the show today. Thank you very much. Unfortunately, I’d love to continue this, but we’re at the top of the hour and have to get our break in. Aaron, great to have you on the phone with us.
00:20:02 — Aaron Yelowitz: Thank you very much. I appreciate it.
00:20:03 — Dan Loney: Thank you, Jerry. Great to have you in the studio.
Jerry A. Jacobs: It’s been a pleasure. Thank you very much.
Dan Loney: Jerry Jacobs, sociology professor here at the University of Pennsylvania; Aaron Yelowitz, who is an associate professor of economics at the University of Kentucky.
00:20:17 — Knowledge@Wharton announcer: For more business news and analysis from Knowledge@Wharton, please visit knowledge.wharton.upenn.edu.