Antitrust in a Changing Economy and Changing Economics
This is a lightly edited transcript of remarks at the Federal Trade Commission Hearing on Competition and Consumer Protection in the 21st Century, September 13, 2018
Thank you so much. Chairman Simons' remarks were perfect in three respects. One is you want somebody to be open-minded coming to this question of how antitrust is evolving because thinking in this area really is evolving very rapidly. Second, he provided a really excellent capsule history of antitrust and thinking. And third, I think he made it clear that he was deferring completely to economists in how he was proceeding on this matter.
I am a little bit of an interloper on this panel. I am the only economist. Moreover, anyone that knows any economics would know I am even more of an interloper than that because my main focus has been on macroeconomic issues, labor market issues, inequality, not on industrial organization and antitrust, narrowly defined.
The Slowdown in Productivity Growth and Increase in Inequality
When I was chairing the Council of Economic Advisors, I came to this issue partly out of what I will now admit was a bit of paranoia. There was a crime that had been committed and we were looking for suspects. The crime was low productivity growth and high inequality. Something was clearly going wrong in the economy. Productivity growth was more than a percentage point lower over the prior decade than it had been previously. At the same time, high levels of inequality continued to move higher. And those were the two main factors that were underlying the slowdown of income growth for typical families, which is the central challenge for economic policy.
What could we do to raise productivity growth or to reduce inequality? And we were looking around at a lot of different suspects. Just to be clear, there is more than one cause of this set of phenomena. But one thing we alighted on was competition and antitrust. Part of what motivated this was a few sub-facts under those two big ones. And let me list a few of them.
First, a number of economists had documented that throughout the economy, there was less churn and dynamism; fewer businesses being created; average business age increasing, larger businesses increasingly dominating the economy; fewer people moving from job to job, so a little bit more of a sclerosis than we would like to think is the case for the U.S. economy.
Second, there has been a downward trend in business investment, which was partly the result of a shift to intangibles but which cannot fully be explained by their increase (Crouzet and Eberly 2018, Gutiérrez and Philippon 2017b, Farhi and Guorio 2018).
Third, on the inequality side, there was a fall in the share of income going to labor.
And finally, there has been both an increase in markups, a rise in the rate of return to capital relative to the safe rate of return and an increasingly skewed rate of return to capital, with some very successful companies having persistently very high returns, much higher relative to the median than was previously the case.
The Increase in Concentration in the Macro and Micro Data
Based on this fact pattern in the aggregate data we decided to look beneath the aggregates to see what was going on at the firm and the industry level.
One way to look at what is going on at the firm and industry level is to use aggregate industrial data and to divide up the economy into 10 industries, or 800 industries, and look at what is happening to concentration in each one—and a number of people did that, Grullon, Larkin, and Michaely (forthcoming), Gutíerrez and Philippon (2017a), the Economist (2016). We also did this at the Council of Economic Advisors, (2016) and these types of studies generally find that in about 75 percent of industries defined in this way concentration increased.
The antitrust community had a divided view on the merits of this approach. Some thought that industrial organization economists had understood it was an idiotic procedure 35 years ago while others disagreed and said it was 50 years ago (Shapiro 2018). Now, the people using crude concentration measures understood this from the beginning. No one would or should bring an antitrust case based on these types of aggregate data. Every approach has its pluses and minuses. But because we were trying to understand economywide phenomenon, we really needed to use economywide data. The type of relevant antitrust market analysis is available for parts of the economy, but not all of it, blocking the ability to really aggregate up, synthesize, and add it all together.
The aggregate data is definitely flawed but it is currently the best we have. The question, however, is not whether it is perfect but whether it is useful. Does it help explain some of what we are trying to explain? And subsequent research by Gutiérrez and Philippon (2017a), among others, has found that at this aggregated level, increases in concentration are tied to reductions in business investment and also are associated with rising markups and rates of profit in those industries. So you see that these different measures seem to, in a broad sense, work and explain some of what we are interested in.
The next set of measures that one could look at are not the aggregate macro data, but as in an antitrust case, data for a particular relevant market, properly defined, and studying whether the level of concentration is high, or has increased.
There have been a range of studies—some done by the FTC; a number done by economists—for a lot of markets, ad services, health insurers, hospitals, refrigerators, airlines, telecommunications, beer, financial services, agriculture, all of which have consistently found very high levels of concentration, and in many cases rising levels of concentration, well in excess of the levels that would trigger a review if there was a merger under the merger guidelines (see e.g. Corbae and D'Erasmo 2013; Shields 2010; Gaynor, Ho, and Town 2015; FCC 2015; and literature reviewed in Abdela and Steinbaum 2018).
Moreover, a new set of research on common ownership is raising the possibility that when the same few companies own all of the airlines and own all of the banks, that increases concentration above and beyond what you would measure if you thought that American Airlines, United Airlines and Delta were three different companies (Azar, Raina, and Schmalz 2016; Azar Schmalz, and Tecu 2018). This is still a very new line of research and the results are contested so I would not necessarily go and make policy on it with certainty tomorrow. But, so far, it is turning out to be empirically more convincing than I would have expected.
Why Has Concentration Increased?
Why have we seen this increase in concentration and what are its consequences? There is not any single answer to the why question. In some cases, the increase in concentration may be for good reasons and reflect increases in efficiency and increases in competition that weed out some of the less effective firms, globalization and the like. This is an explanation that has been stressed by David Autor et al. (2017). That story probably works pretty well in the retail sector where, although there have been a few big mergers, the largest increase in concentration has come about because Walmart figured out how to have better supply chain management and grew, and then Amazon did the same online. As a result there is more concentration in that sector and it reflects that increase in efficiency.
For a lot of the economy, though, the story is less benign and it has its roots in what Chairman Simons described as a large change in the way we thought about antitrust. Kwoka (2017) has documented, for example, the FTC's oversight challenge looking into mergers. The FTC used to look at mergers bringing the number of competitors down from six to five, but now it would never look at something like that. Grullon, Larkin, and Michaely (2017) documented that the number of Sherman Act Section 2 cases filed by the Department of Justice has fallen from around 16 a year from 1970 to 1999 to fewer than three a year since 2000. So there certainly have been changes in antitrust enforcement.
We should also be looking also at regulations and rent seeking that allow companies to create rules that benefit themselves at the expense of others, certainly in questions like intellectual property. And I think a lot of these competition issues are about antitrust, but they also go more broadly.
And then if you look at labor markets, you want to look at occupational licensing, something the FTC has been at the forefront of for a long time. There are also land use restrictions and many other policies and practices that reduce competition in the economy.
So concentration has increased for good reasons and for bad reasons. In some sectors the balance of these is more ambiguous. Take the tech sector. You have seen a lot of innovation, but you also have platforms with network effects that lend themselves to scale. That might say it is efficient to have a single producer at scale. It is also efficient to have a single municipal water company, but that does not mean we would want to let it go off and charge whatever it wanted to charge.
We certainly would not want to regulate technology the same way we regulate municipal water. It is much more complicated and it is an issue that I am currently looking at as head of the Digital Competition Expert Panel for the UK government.
In general these sectoral examples are just about motivating us to try and understand the combination of good reasons and bad reasons that concentration has increased and what its implications are.
Why Do We Care That Concentration Has Increased?
Traditionally in industrial organization we are focused on prices, whether concentration drives prices being higher. That issue matters. Airline prices and cell phone bills are higher in the United States than they are in Europe because European competition enforcers have been more vigorous. The price issue is important. But in the parlance of economics, it is about triangles of deadweight loss and these usually are not huge.
The price issue may be a lot smaller than some of the other issues I started with. One is innovation. What increased concentration does to the incentives for business investment, for research and development, for productivity growth. There is a longstanding debate between a view of Arrow and Schumpeter in economics about the impact of competition on innovation, but there is a number of ways in which it could be deleterious—with fewer upstarts innovating to challenge larger businesses and larger businesses less on their toes for lack of competition.
And then, finally, inequality. At the same time that there has been this increased thinking about these types of macro issues in competition, there also has been a similar effort in labor markets, as well. And that is grounded in the observation that every employment relationship has a bit of monopoly power and a bit of rent that is being divided between the two parties because there is a cost of finding a new job or new employee. So market power matters a lot.
If you have one hospital in town, it is a lot harder for a nurse to threaten to move to another hospital to get a pay raise. If you have two hospitals in town, it is much easier for the two of them to collude tacitly or even illegally to hold down the pay of nurses. Even in the fast food industry, there is evidence that anti-poaching and non-compete agreements have a deleterious impact on workers' bargaining power, help to hold down wages, and have been part of the reason that the labor share has been reduced (Krueger and Posner 2018).
In summary, evidence is coming from a variety of different places and a variety of different perspectives. If you are trying to ask a question about the economy as a whole, you are not going to have one definitive data source or one definitive study that is going to answer that question. You have to take a collage of views, and I think that collage involves looking at the pattern of what we have seen in the data that I have talked about in terms of falling labor share, falling investment, rising markups.
Looking at the industry level, we have seen those phenomena tied to concentration. Looking in a deeper, more careful way where we can, we have found those phenomena as well. No single story comes out of this, but on balance and on average, this does seem to add up to a reduction in competition, a reduction in dynamism and one that I think that we need to be concerned about. We will likely need to update our competition policies to address these issues if we want to have more investment, more dynamism, more productivity growth, less inequality, in addition, of course, to the traditional focus on lower prices for consumers.
Jason Furman is a professor of practice at the Harvard Kennedy School and a nonresident senior fellow at the Peterson Institute for International Economics. He was chairman of the White House Council of Economic Advisers from 2013-17.
Abdela, Adil, and Marshall Steinbaum. 2018. "The United States has a Market Concentration Problem." Issue Brief. New York: Roosevelt Institute. September.
Autor, David, David Dorn, Lawrence F. Katz, Christina Patterson, and John Van Reenan. 2017. "The Fall of the Labor Share and the Rise of Superstar Firms." NBER Working Paper 23396. Cambridge, MA: National Bureau of Economic Research.
Azar, José, Sahil Raina, and Martin C. Schmalz. 2016. "Ultimate Ownership and Bank Competition." Working Paper.
Azar, José, Martin C. Schmalz, and Isabel Tecu. 2018. "Anticompetitive Effects of Common Ownership." Journal of Finance 73(4): 1513-65.
Corbae, Dean, and Pablo D'Erasmo. 2013. A Quantitative Model of Banking Industry Dynamics. Working paper.
Council of Economic Advisers (CEA). 2016. "Benefits of Competition and Indicators of Market Power." Issue Brief. May.
Crouzet, Nicolas, and Janice Eberly. 2018. "Intangibles, Investment, and Efficiency." AEA Papers and Proceedings 108: 426-31.
The Economist. 2016. "Too Much of a Good Thing." March 26.
Farhi, Emmanuel, and Francois Guorio. Forthcoming. "Accounting for Macro-Finance Trends: Market Power, Intangibles, and Risk Premia." Brookings Papers on Economic Activity.
FCC (Federal Communications Commission). 2015. Eighteenth Mobile Wireless Competition Report. Washington: Federal Communications Commission. https://apps.fcc.gov/edocs_public/attachmatch/DA-15- 1487A1.pdf.
Gaynor, Martin, Kate Ho, and Robert J. Town. 2015. The Industrial Organization of Health-Care Markets. Journal of Economic Literature 53, no. 2: 235–84.
Grullon, Gustavo, Yelena Larkin, and Roni Michaely. 2017. "Are U.S. Industries Becoming More Concentrated?" Working Paper.
______. Forthcoming. "Are U.S. Industries Becoming More Concentrated?" Review of Finance.
Gutiérrez, Germán, and Thomas Philippon. 2017a. "Declining Competition and Investment in the U.S." NBER Working Paper 23583. Cambridge, MA: National Bureau of Economic Research.
______. 2017b. "Investmentless Growth: An Empirical Investigation." Brookings Papers on Economic Activity 2017(2): 89-169.
Krueger, Alan B., and Eric A. Posner. 2018. "A Proposal for Protecting Low-Income Workers from Monopsony and Collusion." Policy Proposal 2018-05. Washington: The Hamilton Project.
Kwoka, John E. 2017. "U.S. Antitrust and Competition Policy amid the New Merger Wave." Washington: Washington Center for Equitable Growth. July.
Shapiro, Carl. 2018. "Antitrust in a Time of Populism." International Journal of Industrial Organization.