Feb 02, 2022
On January 26, The Capitol Forum hosted a conference call with Evan Starr, Assistant Professor of Management and Organization at the Robert H. Smith School of Business at the University of Maryland, to discuss his work on noncompete and no poach agreements in employment contracts. The full transcript, which has been modified slightly for accuracy, can be found below.
MR. TEDDY DOWNEY: Good morning, everyone. And thanks for joining The Capitol Forum’s conference call on Noncompete and No Poach Agreements in Employment Contracts. I’m Teddy Downey, Executive Editor here at The Capitol Forum. And I’m joined today by Evan Starr, Assistant Professor of Management and Organization at the Robert H. Smith School of Business at the University of Maryland.
A quick note before we get underway. For the first 20 minutes or so, we’ll interview Evan and then we’ll move into a Q&A format where we will entertain questions from the audience. If you have questions for us, please email them to firstname.lastname@example.org. That’s email@example.com.
Evan, thanks so much for doing this today.
MR. EVAN STARR: My pleasure. Thanks, Teddy, for having me.
MR. TEDDY DOWNEY: So you’ve done a lot of research on noncompete and no poach agreements. And I think it’s safe to assume that at some point soon, perhaps shortly after Alvaro Bedoya is confirmed at the FTC, the FTC will at least start looking into, if not start, a rulemaking process to prohibit noncompete. And I want to ask you, what does your research show about whether or not this is a good idea?
MR. EVAN STARR: Well, I think if we back up and talk about the history just a moment, noncompete policy in the U.S., since roughly the late 1800s, has been stagnant. In the 1800s, there were three states that banned noncompetes. Altogether, those were California, Oklahoma and North Dakota. And by all together, I’m referring to the employment context here, not the sale of the business. And then basically, since then the laws in every other state have been largely stable. And what we’ve seen in the last five years is this huge raft of new policies proposed. And this year, or I guess in 2021 I should say, there were over 65 new bills proposed to limit noncompete agreements.
So we’re in the middle of this kind of huge reckoning. And I think that’s been driven both by new research on how prevalent noncompetes are. In some of my findings, we find that anywhere
between 18 percent and 20 percent of workers are signing noncompete agreements. They’re more common at the upper end of the income distribution, but they’re also super common in low wage jobs, even in nonprofits. Even for volunteer workers, you’ll find noncompete agreements. They sort of pervade every corner of the labor market. And so, combined with some research on how common these things are, there’s been some research that’s come out of mine looking at what happens to workers and firms when states decide to either ban noncompete agreements or enforce them even more vigorously. And those, I think, broadly tend to show harm both to workers and to businesses and entrepreneurship.
And so I think if you combine that very, very recent research, really in the last seven years or so, combined with new anecdotes coming out all the time of new companies that are using noncompete agreements in abusive ways to stifle competition, then you have kind of a recipe for change, which is what we’ve seen and the FTC is certainly moving in that direction.
MR. TEDDY DOWNEY: And you mentioned in your research that there is little real distinction between a noncompete and no poach agreement when it comes to the practical effect. Can you walk us through that? You know, how you came to that conclusion?
MR. EVAN STARR: Yeah. So we should talk about what those are just briefly. So a noncompete agreement is a clause in an individual’s employment contract. It’s something that you, the worker, would sign, either before you take your job or opt on the first day of the job, or sometimes even with a promotion or a raise. And so it’s a term that’s baked into the worker’s contract.
A no poach agreement is a different arrangement. It’s an arrangement that is typically invisible to the worker because it was agreed to by the heads of the organizations themselves. So if you think about the Silicon Valley no poach case, it was Schmidt and Jobs at Google and Apple who are agreeing not to do this hiring in conjunction with some of the HR matters. But for most workers, they’re totally unaware that they’re not allowed to be recruited by other tech firms in Silicon Valley. And so both of these contracts operate to restrain trade and labor markets. But the noncompete agreements are agreed to by workers, and no poaching agreements are agreed to by firms. And so you might think that workers who are willing to agree to restrictions in noncompete, you know, that they won’t go, depart their firm and join a competitor, that workers wouldn’t sign that unless they were compensated in some way, shape or form to make them better off. But what we’ve seen is that’s actually just not true.
So, for example, when Oregon banned noncompete agreements for low wage workers, we saw job mobility rise about 17 percent. We also saw wage growth rise up to about five percent over the next five years. If you look at an experiment, a national experiment, where Hawaii banned noncompete agreements only for tech workers in 2015, we find that mobility among tech workers
in Hawaii also rose about 11 percent, and new hire wages also rose about four percent. And so if you believe that the workers who are signing these noncompete agreements are better off, then you wouldn’t expect that when states ban noncompetes that workers benefit. But that’s precisely what we’re finding.
MR. TEDDY DOWNEY: But in terms of just the idea that workers would be better off in that they would be paid more because they were getting a restriction on their future mobility and earnings potential effectively by switching jobs. They should get compensated upfront for that effectively.
MR. EVAN STARR: Yeah. Well, there are different ways of thinking about it, but the kind of neoclassical economists would frame it this way. They would say you can never get a worker to voluntarily agree to something that makes them worse off. And so that would be their position. And so they would think that a worker would ask for such a high—they’d be able to value the harm that this noncompete is going to do to them in the future. And then they’re going to discount that to the present day. And then they’re going to ask for that sum upfront. And the firm may pay some of that if they’re willing to or the firm may say, okay, well, how about we’ll give you some training? We’ll give you these trade secrets that will also make you more valuable. So there are other ways that you can compensate workers. But that’s the general idea about why noncompetes might actually be good for workers, and that’s what the evidence just hasn’t borne out.
MR. TEDDY DOWNEY: So this neoclassical economic theory was just totally, absolutely completely wrong based on the research that you’ve done.
MR. EVAN STARR: Well, for the average worker, that appears to be true. When you look at executives, the relationship seems to reverse. When executives sign noncompete agreements, they tend to have higher wages. There are several studies that have now documented this fact, including some of my own. And so there is sort of a gap here between the super high executives who have legal teams that can help negotiate their contracts, and the average worker who may be unaware that these terms exist may be asked to sign them on day one, who maybe doesn’t have the access to the legal system that an executive would.
MR. TEDDY DOWNEY: And so this idea has been effectively proven wrong except for these very high-level executives. The research is showing that there are real costs in terms of wages and mobility, which is amazing. Those are big numbers. I mean, 17 percent mobility increase, five percent wage growth. I mean, these are really—if you extrapolate that over the entire economy, that’s a really big deal.
MR. EVAN STARR: Yeah. if you think about over the last—notwithstanding the last year or so, which has seen a market rise in wage growth, especially for work at the low end of the wage
distribution—over the last 40 or so years, the median worker hardly saw a bump in real wages. I think it was like a growth rate of about one percent over about 40 years. All the other growth we’ve had in earnings have come at the very top of the wage distribution. The median worker has roughly been about the same as well off as they were 40 years ago.
MR. TEDDY DOWNEY: Yeah, and despite wage growth, real wages, because of inflation, actually didn’t go up last year, I think on average.
MR. EVAN STARR: That’s actually right, yeah.
MR. TEDDY DOWNEY: So this is a big deal. The FTC, like I like I mentioned, there’s a big expectation that this is early on their list of potential rulemaking. And I just want to make sure, you think that’s a good idea to ban this because of all the benefits to getting rid of it and the lack of benefit from what the neoclassical economists’ theories suggested that would occur.
MR. EVAN STARR: Well, I prefer to be guided by the evidence in these things. I do think at this point that we have accumulated a substantial amount of evidence on the harms of noncompete agreements, and that includes even for executives. And if I could just say one linkage between noncompete and no poach agreements that I think is the strongest justification for a ban on noncompete agreements. I’ll use no poach agreements as an example.
No poach agreements, one of the reasons that they—at least naked no poach agreements where there’s no ancillary contract or transaction, where it’s just an agreement between two firms not to poach each other’s workers. In that situation, those agreements are really just designed to hurt a third-party—the workers. They’re designed to keep worker mobility low. They’re designed to keep wages low. And so there’s a third-party harm that is a direct outcome of those arrangements. And if you look at the Silicon Valley no poach case, there’s a recent paper that finds that those workers who work at a firm who had no poach agreement had 2.55 percent lower salaries for every no poach agreement that the firm had agreed to. So there’s a third-party harm issue.
The same thing is actually true in the noncompete case. There are third parties that are harmed by noncompete agreements. And in particular, the firms that might hire a worker who is bound by a noncompete agreement are harmed by that worker’s noncompete. Because the firm would like to hire workers who are really good fits. And so there’s one study by Leon Shi that looks at this, and it finds that, even for executives, the optimal policy is close to a ban, even though executives benefit from noncompetes because of these negative externalities that are on other firms. These firms then can’t hire that worker. So they have to go somewhere else. They have to leave open the vacancy a little bit longer. Noncompetes can clog up the labor market, especially when they’re used en masse.
I mean, if you think about an industry like health care where more than roughly 50 percent of doctors are bound by noncompetes or high tech where they’re super common or executives where 70 to 80 percent of executives are bound by noncompete agreements, in those areas who’s going to start those new firms? And then those new firms, who are they going to hire? So you can think about what happens when we have this whole mass use of noncompete agreements clogging up the labor market. And the net effect, and one of my favorites, we find is that wages are lower for everybody in the whole market. Mobility is lower in the whole market. And it’s just a less dynamic economy.
MR. TEDDY DOWNEY: And you found also, I mean, I thought this was really interesting, in states that don’t ban noncompete, you even see spillover effects into other states, nearby states.
MR. EVAN STARR: Yeah, there’s a great paper by Kurt Loverde, Matt Johnson and Mike Lipsett that shows that when states increase their enforceability of noncompete agreements, not only do wages fall in that state, but it also affects states that have labor markets that share a border with that state. And if you are in Florida, if Florida increases their enforcement of noncompetes, then you could see declines in wages in Georgia, for example.
MR. TEDDY DOWNEY: I mean, that’s really interesting. And so you’ve got all this evidence suggesting what these things basically should be banned. Even when you say it’s a benefit to the executives, it’s really not a benefit. It’s just kind of a wash. Because you’re getting paid a little bit more, but you’re restricted. So it’s just like, I don’t know if you could really call that a benefit.
MR. EVAN STARR: Well, the individuals might benefit, but others are getting hurt, other firms.
MR. TEDDY DOWNEY: Yeah, I guess, I mean, but are they really benefiting? I mean, they’re just getting compensated for the restrictions. They’re just kind of equaling it out.
MR. EVAN STARR: Well, presumably they wouldn’t agree unless they were weakly better off, I mean, theoretically.
MR. TEDDY DOWNEY: Yeah, yeah, yeah.
MR. EVAN STARR: We have spillovers in entrepreneurship and innovation.
MR. TEDDY DOWNEY: So total harm is clear, right? The overall harm. I mean, five percent wage suppression across the economy in states that do this and nearby states, that seems like pretty good evidence that there’s the potential negative effect. So the FTC is going to do this rulemaking process. You mentioned a couple of different options in terms of doing it under unfair methods or
deceptive practices. Do you have a view on how the FTC should do their rulemaking process? It seems like from a competition standpoint, this new rulemaking authority under unfair methods of competition would be more streamlined and a little bit more applicable. I want to get your thoughts on that if you’ve thought about that.
MR. EVAN STARR: Well, so I’m not a lawyer, and I know there’s a lot of complications with the legal process of rulemaking and then challenges to those rules. But I will highlight what I think is the key challenge the FTC is going to face. Which is if you think about the dichotomy I discussed earlier about executives versus the low wage workers, and you might even throw out high tech workers in there, the argument for noncompete agreements is effectively that without noncompete, businesses are subject to workers who can just walk out the firm, take the firm’s information and then compete against their employer at an advantage because they didn’t have to pay for the development of all the information that they just took. This is sort of the classic argument for why we need noncompete agreements. Because otherwise then the firms wouldn’t invest that information and we could just end up in kind of a low productivity environment.
And so that’s kind of the classic argument that is going to come up here. So there’s this dichotomy of, okay. Do you try to carve out low wage workers where you think these things don’t make much sense? Or do you just do an all-out ban like California or North Dakota or Oklahoma that covers everybody? And I think that is the key challenge that the FTC is going to have to face. And what we’ve seen over the last five years are states taking very different approaches to that. You have, for example, Washington, D.C., which took a very broad approach. They banned them for nearly all workers. The State of Washington took a wage threshold approach of $100,000 a year. And then you have at the low-end a stat, like Illinois. They passed their law in 2017 and they invalidated not competes for those making under $13 an hour. And so the question is how broad is the FTC going to go? Are they going to go all out? Are they going to try to carve out just really low-income people? Is it going to be moderate income, high? How are they going to draw the line, if they draw a line?
MR. TEDDY DOWNEY: I actually think that’s probably not as big of an issue, just because it would seem to be straightforward. If all this research says they’re not really indistinguishable from no poach agreements and they have all these negative effects, it would just be simpler from a legal standpoint to ban or prohibit them, which is just kind of where the petitions are. But you bring up another, I mean, let’s get into the other problem, which is there’ll be workarounds. These companies are sophisticated. You write about how the lawyers are already—I want to get back to lawyers in a second—but lawyers are already recommending that their clients explore alternatives to noncompetes to put other kind of restrictions on employee mobility. What are some of these workarounds? What are the lawyers scheming up next? And to the extent that the FTC can be aware
of that, what are the things that the FTC could do to sort of head off those alternatives when they’re doing this rulemaking?
MR. EVAN STARR: Yeah, you’re right on that lawyers are already advocating that firms consider alternatives. And if you look to what they’re typically advocating for, it’s to review—the firms are using other agreements that can substantively protect those same interests. So they’re saying, okay, go see if you’re using non-disclosure agreements to protect confidential information. Go see if you’re using non-solicitation agreements for your coworkers, for your clients. Go see if you’re using intellectual property assignment agreements and check on those. And all of those restrictions are in some ways less broad than a noncompete agreement because a noncompete says that the worker cannot join a competitor or start a competitor after they leave. So it’s a restriction on mobility. Those are restrictions not on whether a worker can leave, but what they can take with them or how they can behave after they leave.
And so those have been typically seen as less restrictive, but it’s possible that they can be just as restrictive as noncompete agreements. And we saw this recently. There was a case in California where a Google non-disclosure agreement was thrown out because it was overbroad and so broad as to be construed as a noncompete.
And there’s other alternatives as well, which are a little bit more pernicious. These are, for example, training repayment contracts where a worker is required to repay training expenses if they leave before a predetermined period. And the concern there is that firms could simply inflate the training costs. So there’s a great article by Jonathan Harris that reviews a lot of these cases where companies have dramatically inflated the training costs they provided. And if they inflate the training costs and a worker wants to leave, then that actually has a broader effect than the noncompete because that training cost is going to apply wherever the worker goes, not just to competitors.
And sometimes firms also just used liquidated damages clauses where they literally just require the worker to pay a fine for leaving. And obviously, that’s broader than a noncompete. Because again, it applies to every potential destination for that worker.
There aren’t there are many there are many ways that firms could potentially respond on these dimensions. I will say that when you look at the natural experiments we’ve seen, the Hawaii natural experiment for tech workers where they banned noncompetes to protect workers in 2015, the Oregon experiment where they banned noncompetes for low-wage workers in 2008, those studies suggest that however the firms responded, we still saw wage gains for workers. We still saw mobility gains. And so even if firms substituted towards these alternative mechanisms in those cases, they weren’t enough to constrain workers’ wage growth and mobility.
MR. TEDDY DOWNEY: So are there any that typically have similar results as—I mean, they’re not quite as effective but when the FTC thinks about this and thinks about closing off loopholes, one, they could just respond what you just said, which is, look, even if we just ban noncompetes, we’re going to get a big result. But two, were there any in particular that you say, look, they would get a little bit more bang for their buck if they banned or sort of constrained or restricted companies’ ability to do training, pay for training or liquidated damages or something, as effectively resulting in the same scenario as a noncompete or similar effect?
MR. EVAN STARR: Yeah. Well, what I’ll say is this. If you look at the all the research that exists, what we have is a growing body of evidence on noncompete agreements. What we don’t have is a lot of evidence on the effects of these other restrictions. It’s just not there. And so it’s hard to say from an empirical perspective which of these the FTC should focus on. Because we honestly just don’t know which of those is going to have the biggest bang for the buck.
That said, I think one approach the FTC could take is the same one adopted by the Uniform Law Commission. And full disclosure, I was an observer on the Uniform Law Commissions Committee when they developed the Uniform Restrictive Employment Agreement Act. And there, if you don’t know the Uniform Law Commission, they develop uniform laws like the Uniform Trade Secrets Law, which states can then adopt right off the shelf. They don’t need to go through their own process. They can say, wow! This was a very carefully written law. This is the one we’re going to adopt. And that creates some uniformity across states. And the Uniform Law Commission took a broad approach. They didn’t just regulate noncompete agreements like most of the laws we’ve seen in the last few years do. They recognized that firms have an array of restrictions that they can use. And they actually wrote out for each possible restriction what the limits were. And so I can see a broader approach that directly addresses these alternatives as being a potential way forward.
MR. TEDDY DOWNEY: And you mentioned that no one has adopted this new uniform law agreement, but the Uniform Law Commission has been effective on other things. Are there good prospects for this to get momentum? How do states typically react? And what’s kind of the timeframe around the effectiveness of what the Uniform Law Commission puts out?
MR. EVAN STARR: Well, I think that everyone’s kind of in a holding pattern, as several, as I said, 65 plus laws relating to no competes were proposed last year. There’s also federal proposals and FTC rulemaking. So right now, there’s kind of too many cooks in the kitchen and everyone’s trying to figure out where the chips are going to fall. So I’m not sure exactly which is going to take precedence here. But one thing I like about both the federal approach or the Uniform Law Commission is that interstate commerce with noncompete agreements is really difficult. If you are a worker who’s working in the D.C. area, for example, where they just banned noncompete agreements—it hasn’t gone into effect yet, but it should, I believe, in March. And now suppose
you joined a company in D.C., but you now want to take a job in Minnesota. Whose law should apply? And sometimes that law is written into your contract as D.C. law. Sometimes it will be Florida law. And sometimes maybe you want to try to apply Minnesota law because that’s where you’re going.
And so when every state adopts its own unique laws, it can be really difficult for workers to transition across states. And this, I think, is going to be even more pronounced as we move towards a period of work from anywhere which the pandemic has obviously exacerbated. And so my sense is that there is a lot of value in uniformity once we have kind of converged on what we think the best policy is, and whether that’s a federal law or whether that is a uniform law.
I mean, this has been a huge issue. Just to give one example California, I told you, banned noncompete agreements in 1872. But there was a workaround until recently where firms could require other states’ laws to apply in a given worker’s employment contract. And in 2017, California passed Section 925 which said If you’re a California worker, we are applying California law, regardless of what your contract says. And so you can imagine if every state is doing different things with their policies, it can just get very messy both for the workers and for firms who are trying to hire.
MR. TEDDY DOWNEY: Yeah, and it’ll be interesting to see how the FTC thinks about these other laws and the Uniform Law Commission. One thing I wanted to make sure we touched on is in your article you mentioned lawyers carve themselves out from noncompete enforcement. That was just an amazing little anecdote. Can you tell us a little bit more about that? They’re okay for everyone, except for lawyers. I mean, it’s kind of a remarkable detail.
MR. EVAN STARR: It is a remarkable detail. Lawyers have, since the invalidated noncompete agreements among themselves as part of their ethics rules, and it’s been adopted in every state. And the rationale for the four invalidated noncompetes among lawyers is exactly the same rationale we talked about for invalidated noncompetes everywhere. It’s that there’s third-party harms. If you are a client and your attorney has to sit out of the market or leave and go far away because of a noncompete agreement, then that noncompete has just infringed upon that client’s freedom to choose an attorney. And so that is the basis for this for this carve out for lawyers, that noncompete agreements have third-party effects that negatively impact a client’s choice of attorney.
But the same argument is also true for physicians. There are several horrifying stories of physicians who have to leave the area and their patients can’t get care, and they don’t know where their physician is. The same thing happens in any personal services context that you can possibly think of. And so it is really interesting. You may ask how do lawyers cope with this right? They’re very sophisticated. They obviously know the law. And they’ve figured out how to deal with this for the
last 60 years. So they’re not trying to advocate that noncompetes should be enforced among lawyers anymore. But many of them do and will continue to argue that noncompetes should be enforceable for other workers.
MR. TEDDY DOWNEY: It’s shocking. I mean, it’s just a shocking carve out or workaround that they’ve created for themselves. And it’ll be interesting to see if that comes up in the rulemaking process and lawmaking at the state level. And you mentioned, I think one thing that’s kind of interesting about all of this is the activity around trying to figure out the pressure for the FTC to act. You mentioned all the state laws. You mentioned how much of an increase there has been in the use of noncompetes. In your in your article, you also mentioned there was an FTC workshop and you mentioned the president’s Executive Order on competition asking for something on the noncompetes from the FTC. What were your takeaways from the FTC workshop? And I’m also curious if there’s something bigger going on about why people are interested in this topic all of a sudden. Is it that there is a broader fascination and interest in worker pay and things like that going on?
MR. EVAN STARR: Yes, as that’s a great question. Since the Obama administration, effectively in 2014, is when this stuff started gaining traction due to the Jimmy John’s noncompete case that kind of shocked policymakers. And Biden, as Obama’s Vice President, was very vocal at that point about how concerned he was that noncompete agreements could constrain the upward mobility of low wage workers. And since then, gosh, we’ve seen so many bills.
The one I just want to point out that I think is interesting is Chris Murphy and Todd Young of Indiana and now several other both Democrats and Republicans have signed onto ban noncompete agreements outright. And there’s a twin bill in the House as well. So we have a bicameral, bipartisan bill to ban noncompete agreements everywhere that’s also moving alongside Biden’s request for the FTC to curtail unfair noncompetes.
And the reason I bring that up is because I think part of the draw of noncompete agreements that’s made it an important issue today is that it’s not partisan. Both parties are interested in workers being able to improve their livelihood. And when there are contracts that hold them down, that chain them to their job, it’s hard for policymakers to stomach that, especially when the worker is making minimum wage and could never afford an attorney.
And so I think there is scope for movement on this front because there’s a lot to agree about. Every firm that’s tried to hire a worker who’s bound by a noncompete agreement and gets stifled, has a negative experience with it. This is not a worker/firm issue because firms are on both sides of the equation. I mean, firms would definitely love to protect t
with noncompete agreements. But firms are also hurt by noncompetes because they clog up the labor market. They make it harder for workers to start new firms.
And so this is a classic what we call a prisoner’s dilemma where every firm may have a private incentive to use noncompete agreements. But if every firm does that, then we may be worse off as a society. And the solution to that is the policy, right? And the policy doesn’t require passing huge amounts of money. It doesn’t require millions of dollars or billions of dollars or tax increases or something. It is a policy that just regulates how these contracts are enforced.
MR. TEDDY DOWNEY: Yeah, I wanted to stay on one of the arguments for noncompetes is that it will be so easy for an employee to just start their own company and take everything and take all your business. That’s kind of a completely ridiculous, fictitious view of how hard it is to start a company. Does that come up as like this whole premise for these as being a realistic thing that you’re fighting off as just ridiculous on its face? As someone who started a company, that is not how it works. It’s not easy regardless of how much—I mean, obviously, you can’t steal sensitive trade secrets. It’s a completely different area of law. But there are so many things that make it hard to steal employees and leave unless they’re being treated badly and so many difficulties in starting a business. Like does that come up? Just kind of like an argument about the premise of why these things exist in the first place?
MR. EVAN STARR: Yeah, so I think you raise several good points and let me just highlight a few of them. One is you’re absolutely right that firms have alternative ways to protect themselves. We have trade secrets laws that protect against the misappropriation of trade secrets. We have nondisclosure agreements, which can go further than just protecting trade secrets. They can protect all sorts of confidential information. And with the MeToo movement, we’ve seen kind of a new understanding of how broad noncompetes can really get—sorry, not noncompetes, non-disclosure agreements. The firm has non-solicitation agreements where you can agree that you can leave and go start a competitor, but just don’t solicit my clients for a year. They have those agreements that they can use. They can protect a lot of their assets already.
So the question is why do you need noncompetes people on top of all that? Well, here’s the story that the lawyers will tell you. They will say suppose you have a worker who is a bad actor, who is going to take all of these trade secrets or other information and try to use them against you. And if this worker is able to leave and do this, then it may take a very long time to get that trade secret lawsuit up and running. It may take a while to realize that that worker has solicited all your clients. It may take a while for the firm to realize that the NDA has been violated.
So in this situation, the noncompete agreement gives the firm a really quick and easy way to protect themselves, which is that they don’t have to observe that information was misappropriated. All they
have to observe is that the worker started or joined a competitor. And so that gives them, as you can imagine, an enormous amount of protection, probably too much protection. But this is the argument that they will make.
And so what I’ll say is there’s a lawyer who’s a friend of mine, Russell Beck. And maybe it was him or John Marsh, one of these guys came up with the term springing noncompete, which I think is interesting policy solution. The solution is you don’t include noncompete agreements in employment contracts. They’re gone. They’re not in the employment contract. But if a worker is perceived to be, justifiably has mal-intent, then you can bring a noncompete as a remedy. And so hopefully, if a worker is just leaving, normal course of action, everything’s fine, then we can have all the benefits of free mobility in hiring and all the extra business dynamism. But in the instances that firms are really concerned about where a worker may have bad intentions, there is this remedy that could help them. I think that is a middle ground that has some appeal.
MR. TEDDY DOWNEY: It’s interesting. But who determines what mal-intent is? It could just be that employee is really good at their job and actually represents a risk to the business. And that’s a key man risk.
Mr. EVAN STAR: Absolutely.
MR. TEDDY DOWNEY: Not mal-intent. That has all sorts of unintended consequences, if you think about it.
MR. EVAN STARR: And I’ll just say one thing on starting a new firm that is interesting, that kind of reflects this challenge, which is that if you look at some of the advocates for noncompete reform, for getting rid of noncompetes, you’ll see folks like the New England Venture Capital Association who were active with the Obama administration. And they want dynamic entrepreneurial labor and product markets. And so they wanted reform in this area. But if you look at some older studies, if you look at the contracts that venture capitalists sign with the companies that they’re funding, seventy percent of those contracts also have noncompetes in them. So they don’t want this early-stage startup someone to run off and take the idea and do something else. But they also want people to be free to start startups. So you can kind of sense the tension with which those venture capitalists are trying to navigate this environment.
MR. TEDDY DOWNEY: Yeah, that’s interesting. If you hear children screaming, we’re just piping that in for authenticity during the COVID era. But if people have questions, please send them to firstname.lastname@example.org. I have a couple more here. One thing that I’ve come across—and I’m curious if you have noticed this as well—in looking into consolidated industries is that it seems like they’re HR departments all know each other and talk to each other pretty frequently, which
seems like that should not be allowed and that should be really policed. Does that come up in your research as a problem that these consolidated industries would be sort of like, you know. It seems like ripe for potential no poach agreements and collusive conversations. They also go to the same conferences and things like that. Are there any tips or ideas that you’ve had from the criminal standpoint like things that DOJ should look into? I mean, that’s kind of one that I’m curious about, but I don’t know if there were others.
MR. EVAN STARR: You are right on when you say that collusion is more likely in highly concentrated markets. And that holds for product markets. That also holds for labor markets. And you are also right that a lot of these top folks, they know each other. They’ve interacted with each other. I used to teach in a school at the University of Illinois, the Labor and Employment Relations School, where we gave degrees to HR folks who then went into HR roles at Boeing and several companies. And they all came from the same institutions, of course. They know each other. And there are several of these schools throughout the country. So absolutely they know each other.
And I’ll just say one more thing that we’ve learned in the last few years is that economists for a long time have thought that labor markets were perfectly competitive. If you wanted to sell your labor from firm A to firm B to firm C to firm D, you can go ahead and do that. It doesn’t matter what industry they’re in. You can go flip burgers. Or you can go do swim lessons. What we’ve learned over the last really seven years or so is that labor markets are very much imperfect. And, of course, it takes a long time to even find a good fit for a job. You’ve got to have your preferences satisfied. You’ve got to have your employer’s preferences satisfied. You’ve got moving costs that you’ve got to deal with.
There’s one study that just came out this last year, which showed that if you define labor markets by an occupation and a commuting zone, that if you just look at the number of vacancies that exist in a given occupation and commuting zone, that many, if not most, of the labor markets in the U.S. would be defined as highly concentrated by the DOJ’s own standards. And so I think that we are coming to this realization, like not only are the HR folks connected, that they know each other, but actually labor markets are a lot smaller than we thought they were.
And so I do think that raises concerns about collusive behavior. And the DOJ has taken a very strong approach to this. They are not only going after these cases, but they are also pursuing criminal sanctions for those who have been perpetrating them. I think that sends a pretty strong signal to me and I’m not sure if companies have totally taken notice yet because we’re still seeing these cases trickle in. But I do think that they are doing what they can. I don’t know what I would advise them to do more, except to increase funding to the DOJ and the FTC so they can pursue these activities.
MR. TEDDY DOWNEY: I mean, this is the second example you’ve given of neoclassical economists and their theories being wrong. I’m just curious, are there any other examples of economic theories in this area that we should look at as just like wrongheaded based on the evidence that you’re saying? Just because it just seems like a theme. It seems like there are a lot of theories out there and there’s just no evidence to support them, particularly neoclassical economic theory. Do any others jump out at you? I mean, we’ve got two here so far.
MR. EVAN STARR: I think the big one that we saw in the last, well, I guess is the minimum wage. Where back in 1991, the classic debate over minimum wages if you increase the minimum wage and labor markets are competitive, then basically all you’re going to end up doing is getting a whole bunch of workers laid off from their firm because firms don’t want to pay more money for these workers. They can’t afford to. And so they lay off workers. That has been like the neoclassical prediction for a long time.
And then Alan Krueger and David Card in 1991 wrote a paper about New Jersey and Pennsylvania, where the minimum wage went up in one state, but not the other. And so they used as a natural experiment to see what happened to employment at these food establishments. And they found no effect of the minimum wage, at least no negative effect. If anything, the effect was positive on employment. And so what we’ve seen, I mean, that was 1991. So that shook a lot of people’s feathers. People dismissed it.
And then what we’ve seen since then in the last few years are several very high-quality studies exploring lots of minimum wage changes over the years, finding largely similar things that when you increase the minimum wage, that workers are not laid off, that workers’ wages do rise at the bottom of the distribution. And so the question is why is that? And the answer that most have come to is that’s because labor markets are imperfect. And what that means is that firms can hold down wages below what they would have been in a competitive market. There’s space. And so when you increase those wages a little bit, the firm doesn’t have to lay people off. And in fact, they could actually hire more people because more people are willing to join that company. That’s the big fear. I mean, that’s part of the reason that David Card won the Nobel Prize in economics this year.
MR. TEDDY DOWNEY: And that gets me to my final question here which is despite the evidence, despite these theories being wrong, most people are still taught that these are sort of basic truths about how to think about the economy. And I’m curious to get your thoughts of whether or not this sort of new thinking about labor, you know, we’re seeing it at the FTC. We wrote about this Lockheed/Aerojet merger where they were analyzing data about how it would affect wages for engineers, aerospace engineers, in certain markets.
So I think from an expert standpoint, from a policy making standpoint, we’re seeing this have an effect. But are there any other things that you’re seeing that are more mainstream that will get into the sort of mainstream thinking, for the evidence to start making its way into the general public understanding of how things work? Or just how long are you thinking it will take before it’s kind of conventional wisdom that these economic theories have been proven wrong? Because you still read articles that, oh, well, they’re raising the minimum wage. Will jobs be cut? I mean, that’s still a huge debate at the local level and in the local press. That’s not by any stretch sort of well understood that these sort of theories have been debunked. Like what are you seeing in terms of trends like the mainstream buying into this new evidence?
MR. EVAN STARR: Well, it’s slow. It’s definitely slow. I mean, I think that every economist would agree with you, that if you raise the minimum wage enough, eventually firms will start shedding workers. Every economist would agree that if you raise the minimum wage to $200 an hour, then fast food restaurants would be letting some workers go. So the question is really one of magnitude and where in the middle is it going to be better for workers without hurting the workers who are now laid off? And I think ultimately, that’s an empirical question that you need policy shocks to study. You need big changes in the minimum wage to see. Did this really big shock get workers fired? Or did it keep workers there and just raise their wages?
I think that a big part of the problem is that the way we teach economics in undergraduate courses, in high school, on the AP exams, even in advanced graduate classes, sometimes is based really heavily around this perfectly competitive model. It’s a nice benchmark where the math works out pretty, where the graphs look really nice. And I think that where the econ profession has come is to realize that the perfectly competitive model is reasonable in some cases maybe, but in most cases, markets aren’t perfect. And I think that we’re moving towards a place where the default is—really in the last few year—is becoming an imperfect markets, both in product markets and in labor markets. And I think that move for labor markets has really lagged product market. If you’re involved in antitrust litigation or you’re involved in industrial organization research, rarely, rarely, rarely do you think about a perfectly competitive product market. Because in those markets, there’s no harm to consumers. But if you allow for imperfect markets, then all sorts of things can happen. And Iowa economists have been there and they’ve known that for a long time. It’s that labor economists have taken a real long time to catch up and to learn that labor markets actually operate in the same way.
MR. TEDDY DOWNEY: And in terms of bringing that back home to antitrust, is it more obvious to everyone now that consolidation and power of the employer plays a role here? I mean, that’s really the imbalance of power between the worker and the employer, the concentrated market, how power makes it imperfect. Is that something that’s kind of becoming a little bit more well understood or studied?
MR. EVAN STARR: I think we’re getting there. I think we’re getting there. I’ll tell you—so Matt Gibson is the economist at Williams, who wrote a really nice study on the Silicon Valley no poach case. And let me just summarize again, the main finding there is that when you take two firms who have agreed not to hire from each other, that single agreement reduced the salaries of those workers at those firms by, on average, 2.5 percent. So if you just like zoom out, and I said to you, let’s imagine the two firms and the labor market agree not to compete for each other’s workers, just two random firms, you probably wouldn’t think there’d be much of a wage effect.
But what we have here is that in a case where these firms can select each other, when they know who their labor market competitors are, they can select just one other employer in the market. And when they do so, they can decrease their wage bill by a substantial amount. Just one employer. To me that has implications for how we think about antitrust. Because we do see, of course, two firms merging all the time. That would be the same idea just with the merger, if Apple and Google were to merge, for example. Obviously, that wouldn’t happen, but as a hypothetical. And so you have these situations where firms, they know their labor market competitors. And those competitors are the ones who drive wage growth, who drive mobility. They drive careers. And if you can just get an agreement with one of those not to hire from each other, you can start wage growth for a whole bunch of people. I think that finding in and of itself should tell us a lot about how modern labor markets are working and how small they really are.
MR. TEDDY DOWNEY: Yeah. Well, I mean, every single aspect of this conversation is really fascinating to me. I love learning about it. I love all these anecdotes. I love learning about this and I really can’t thank you enough for doing that. Your research is really interesting. It’s, I think, a dynamic space where we’re learning a lot and there’s a lot of new evidence and people are really benefiting a lot from the work that you’re doing. So thank you so much for doing this. Thanks for all the interesting research that you’re doing. And thanks to everyone for joining us on today’s call.
MR. EVAN STARR: Thank you so much, Teddy, for having me.