Antitrust in the Information Age
Among the arguments being laid out by the U.S. Department of Justice in its opposition to AT&T’s acquisition of Time Warner is the notion that the mega deal would stifle innovation from online streaming firms. The showdown over the proposed mega merger highlights an issue of rising importance for antitrust in the information age: the relationship between market power and innovation. What type of market environment best spurs innovation? Is it a highly competitive market with many companies, all of which are trying to innovate to gain market share and increased profitability? Or, is innovation more likely when an industry is dominated by a few large companies that can afford to set up expensive research labs and navigate the necessary regulatory procedures?
Understanding the market conditions that best support innovation has important implications for antitrust policy.
- Antitrust laws are meant to preserve competition. One primary concern has been that the existence of monopolies and oligopolies leads to higher prices for consumers. An obvious violation of antitrust laws is an explicit price-fixing conspiracy. Antitrust laws also often prohibit predatory actions of dominant companies that drive or keep competitors out of a market, and mergers between two (or more) large rivals that might create a monopoly.
- Technical innovation — new products or new ways of making old products — is a key driver of economic growth and has been credited with roughly two-thirds of per capita GDP growth in the United States over the last 100 years. Technological innovation drives economic activity in areas as diverse as bio-tech, energy production and on-line shopping. The information age has ushered a wave of technological innovations rapidly transforming almost every aspect of our daily lives. While notable innovations were famously born in a home garage or college dorm, their growth and proliferation has been accompanied by the rise of new mega companies with tremendous market power such as Amazon, Alphabet (Google's parent company), Facebook and Apple. The proposed merger of AT&T and Time Warner would lead to yet another mega company.
- Increasingly, antitrust authorities have come to recognize that dominant firms and mergers have potentially important effects on innovation as well. The Department of Justice (DOJ) made this clear at the dawn of the information age 15 years ago when it prosecuted Microsoft for suppressing the emergence of rival software. The case was resolved when Microsoft consented to stop requiring that computers using Windows exclude other programs and to cease making Windows non-compatible with rival software applications. By 2010, the Merger Guidelines issued jointly by the DOJ and the Federal Trade Commission (FTC) explicitly recognized the need to consider the competitive innovation implications in evaluating the overall impact of a proposed merger. Such analysis played a big role in the FTC’s 2016 approval of the merger between Teva and Allergan — the largest and third largest generic drug producers. While the FTC permitted the merger to proceed, it did so only on the condition that the companies divest over 75 of their pharmaceutical products, including many in the pipeline stage. These divestures reflected concerns about the merger’s impact in markets where the two firms currently competed. But, importantly, they also reflected the impact in markets where one firm was active and the other was developing a new drug for entry into that market, as well as in markets where neither was currently active but where both were developing products that, if successful, would compete in the future.
- Economists debate how the nature of market competition affects innovation. One approach follows the work of Kenneth Arrow and argues that highly competitive markets spur innovative activity because firms under competitive pressure will vie to produce better or more cost-efficient goods in order to gain market share and increased profitability. And because they do not have monopoly power to begin with, firms in highly competitive markets do not have to worry that developing a new product will cause the loss of profit from displacing an existing one. The other approach reflects the earlier work of Joseph Schumpeter. In this view, the large expenses incurred in setting up research labs, testing, and obtaining regulatory approval means that firms that innovate must have enough market power to appropriate most of the profit that their innovation generates. And these firms will pursue innovative activities precisely because if they do not, their profit, and their very existence will be threatened by other large firms that do innovate. In the Schumpeterian world, sustaining innovative activity requires large, oligopolistic or dominant firms. There’s also the problem that many technological upstarts compete vigorously precisely because they hope to someday become dominant. So, an antitrust policy that protects small firms by limiting the actions that dominant firms can take may end up dulling the incentives for startups to launch in the first place.
- But separating innovative activity from predatory behavior by dominant firms is challenging. Consider for example, a drug company that has a monopoly over a medical treatment that requires a monthly injection. Suppose that a rival is in the process of developing and marketing an equally effective alternative also administered by monthly injection. Is it predatory if the monopolist incumbent preempts the rival and introduces a new version of its drug that can be administered orally in a weekly pill? If the authorities say yes, society may lose a quality enhancement of significant value to consumers. If they say no, they may open the door to tactics deliberately aimed at reducing competition in the market. The monopolist may have had the oral version ready for some time and only introduced it when a rival appeared in the wings.
- The underlying uncertainty regarding how market conditions affect innovation is reflected in the diversity of antitrust policies. The recent cases against Google and the 2017 merger of Dow Chemical (the second largest chemical company in the world) with the DuPont Company (the fourth largest), offer good examples. In both cases, European and U.S. authorities had essentially the same market information. Yet in both cases, they reached markedly different conclusions on the threat to innovation. Google became the preeminent search engine fairly early on in the history of the Internet. Over time, Google modified its algorithms to produce the modern universal search. Initially, a query about “new york hotel prices” would only list a series of websites, but now the same prompt brings up both a list of travel websites such as Expedia or Trivago, as well as a quite prominent box with a map of Manhattan hotel locations, each showing a price, and three or more reviews of individual hotels. This innovation helps Google consumers, providing them more information more quickly. But it is also a strategic response to potential rivals like Trip Advisor or Amazon, where consumers might simply go first and then start their search within that site. A web site’s advertising revenue is directly tied to the frequency of its use. Therefore, one could argue that Google’s universal search algorithm had predatory effects vis-à-vis its competitors like Expedia and Trip Advisor. The U.S. and European authorities differed when evaluating this question. The U.S. Federal Trade Commission declined to pursue Google for abuse of its monopoly power. But in June, 2017, the European Commission found Google had abused its dominant position to suppress other search innovators. They stipulated new limits on Google’s practices and hit the company with a record €2.4 billion ($2.7 billion) fine.
- The Dow-DuPont merger was also handled differently by the European and U.S. authorities. Dow had numerous products in agriculture (herbicides, pesticides, fungicides), in basic chemicals (chlorine and vinyl chlorine), and in plastics. DuPont had a heavy investment in crop protection and plastics as well, and was also the leading producer of the paint pigment, titanium dioxide. Together, Dow and DuPont are the only two U.S. suppliers of acid copolymers and ionomers, both of which are important inputs for food packaging and other plastics applications. The two companies also share common research activities in developing new seeds and new crop protections. Both the European Commission and the U.S. Department of Justice (DOJ) approved the merger subject to the divestiture of several existing product lines. However, the European authorities went one step further. Expressing concern about the merger’s impact on the development of new products, they required that the new company sell off DuPont’s assets used for research and development of new crop protection products. In contrast, while the DOJ recognized these same concerns, it did not find that conditions in the U.S. market warranted such a remedy.
What this Means:
The proposed merger of AT&T and Time-Warner is a vertical merger in which a telecommunications firm (AT&T) is acquiring a producer of programming — news and entertainment. While a major fear is that the merged firm will restrict access to its programming like HBO and CNN to rival distributors, another fear is that AT&T will favor its programming over newer, innovative shows. Indeed, the two fears are related. If AT&T is able to leverage its programming content to gain power in the distribution market, there may be even less reason to seek out new programming. Promoting innovation and future technological progress is important and so getting antitrust policy right in this area is also important. Economic analysis can and will contribute to this goal. We now have good theoretical models of innovative competition. We know that some competition — what economists call contestability — is needed to spur innovation. We also know that appropriating the gains of innovation requires that successful firms subsequently grow large. Differences in European and U.S. policy stem from disagreements about the evidence and about what the facts say is the right theoretical conclusion, and hence, the right policy choice. In antitrust, as in other areas, economic data are a lot messier than economic models. With time, we can hope that experience with the data will facilitate recognizing empirical patterns and reaching consistent conclusions. But no one ever said that this would be easy. In this day and age “the future isn’t what it used to be!”