Watch the hearing in full here.
Antitrust reform is a central focus for both Republicans and Democrats who want to reign in big tech. The recently re-introduced “Consolidation Prevention and Competition Promotion Act of 2021” (CPCA) is Congress’ most recent iteration in a shift towards more merger review.1 Amy Klobuchar, “Text – S.3267 – 117th Congress (2021-2022): Consolidation Prevention and Competition Promotion Act of 2021,” legislation, November 18, 2021, 2021/2022, https://www.congress.gov/bill/117th-congress/senate-bill/3267/text. Inherent in this bill are assumptions about how markets and big companies arrived at their current distributions (and concentrations) of power. When compared to economic research of the last two decades, these assumptions do not hold up, and therefore do not support the changes the CPCA 2021 is attempting to institute. Many of the trends that the bill hopes to reverse aren’t from a lack of enforcement, but rather are due to broad changes in U.S. demographics.
This letter analyzes the disparity between rigorous economic research and the assumptions outlined in the CPAC 2021. It will examine the following:
- The rise of market power according to the language in the CPCA 2021;
- The substantive changes the bill would create;
- A more benign alternative for how power became concentrated in current US markets; and
- Potential consequences of changes to FTC enforcement proposed by the
1 The rise of market power
The first six pages of the bill describe why many reformers of antitrust believe a change is needed. According to the account in those opening pages, competition has fundamentally changed since 2008. The bill attributes some of those changes to antitrust agencies being derelict in their duties.
The bill begins by describing the purpose and value of competitive markets with the implication that markets in the United States have lost some of the fundamental benefits of competition: “[C]ompetitive markets, in which multiple firms compete to buy and sell products and services, are critical to ensuring economic opportunity for all people in the United States and providing resilience to the economy during unpredictable times.” When firms compete, they “offer the highest quality and choice of goods and services for the lowest possible prices to consumers and other businesses.” Competition “fosters small business growth, reduces economic inequality, and spurs innovation and job creation.”
Beginning in 2008, industries began to change as market power shifted. “The presence and exercise of market power has grown and become more substantial,” CPCA explains. In turn, market power “makes it more difficult for people in the United States to start their own businesses, depresses wages, and increases economic inequality, with particularly damaging effects on historically disadvantaged communities.” Market power also means “higher prices, lower quality, lessened choice, reduced innovation, foreclosure of competitors, and increased entry barriers.”
The bill also cites shifts in consumer, labor, and supplier markets as being concerning. “Monopsony power or seller market power allows a firm to force suppliers of goods or services to accept below market prices or to force workers to accept below market wages, resulting in lower quality products and services, reduced opportunities for suppliers and workers, reduced availability of products and services for consumers, reduced innovation, foreclosure of competitors, and increased entry barriers.”
Most concerning of all, the bill credits the market power shifts with affecting political power and the quality of life for all Americans. The twin problems of “market power and undue market concentration contribute to the consolidation of political power, undermining the health of democracy in the United States,” which “threatens to place the American dream further out of reach for many consumers in the United States.”
2 How the bill changes merger review
Congress hopes to change course by setting new rules for mergers and acquisitions since “horizontal consolidation, vertical consolidation, and conglomerate mergers all have potential to increase market power and cause anticompetitive harm.” Acquiring “nascent or potential rivals” presents “significant long-term threats to competition and innovation,” especially if the acquisition is “a maverick firm that plays a disruptive role in the market—by using an innovative business model or technology, offering lower prices or new, different products or services products.”
The proposed bill would amend the law such that the Federal Trade Commission (FTC) and Department of Justice (DOJ) would only have to show that the proposed transaction materially lessens competition to challenge an acquisition. It would be a substantial change from the current standard, which makes deals illegal if they significantly lessen competition.
These changes shift the burden of proof for firms with $100 billion in assets, sales, or market capitalization. Instead of a presumption of innocence, companies that meet this threshold angling to buy companies that triggered the normal premerger notification or had $5 billion in assets would have to prove that the transaction does not harm competition.2 “HSR Threshold Adjustments and Reportability for 2021,” Federal Trade Commission, February 17, 2021, https://www.ftc.gov/news-events/blogs/competition-matters/2021/02/hsr-threshold-adjustments-reportability-2021.
Senator Klobuchar has in the past called out these as megadeals, but the bill would shift proof and the presumption of innocence for all companies over a certain threshold.3 Liz Crampton, “Klobuchar to Sponsor Antitrust Bill That Requires Higher Fees,” U.S. Senator Amy Klobuchar, accessed December 14, 2021, https://www.klobuchar.senate.gov/public/index.cfm/2017/3/klobuchar-to-sponsor-antitrust-bill-that-requires-higher-fees.But it is not the size of the deal that’s at issue. Because the law doesn’t scale down the notification threshold, deals that fall below this line would face no review at all. It would be a wholly unique standard for large companies since nothing else like it exists in tort law, contracts, or even criminal law.
The $100 billion market cap casts a big tent, capturing 88 U.S. companies. The list is extensive and involves more than just big tech companies, including: Apple, Microsoft, Alphabet/Google, Amazon, Tesla, Meta/Facebook, NVIDIA, Berkshire Hathaway, JPMorgan Chase, Visa, Unit- edHealth, Johnson & Johnson, Home Depot, Walmart, Procter & Gamble, Bank of America, Mastercard, Adobe, Pfizer, Oracle, Walt Disney, Netflix, Exxon Mobil, Nike, Salesforce, Broad- com, Thermo Fisher Scientific, Cisco, Costco, Coca-Cola, Abbott Laboratories, Pepsico, Eli Lilly, Chevron, Danaher, AbbVie, PayPal, Comcast, Verizon, QUALCOMM, Intel, Wells Fargo, McDonald, Intuit, Merck, United Parcel Service, Texas Instruments, Nextera, Morgan Stanley, Lowe’s, AMD, AT&T, Union Pacific Corporation, Charles Schwab, Honeywell, T-Mobile US, BlackRock, Philip Morris, Applied Materials, Starbucks, CVS Health, Goldman Sachs, Estee Lauder, American Express, Raytheon Technologies, ServiceNow, Bristol-Myers Squibb, Ameri- can Tower, Intuitive Surgical, Boeing, Citigroup, Amgen, Prologis, Airbnb, S&P Global, Target, Snowflake, IBM, Deere & Company, Caterpillar, Charter Communications, Zoetis, General Electric, Anthem, Moderna, 3M, and Rivian.
The $100 billion threshold would be indexed to the Gross Domestic Product (GDP) to ensure it keeps up with changes over time. But the indexing creates a mismatch since GDP grows more slowly than market capitalization. A decades-long trend suggests that market value tends to grow about 1.2 times faster than the GDP.4 “Buffett Indicator Shows Stock Market Is Strongly Overvalued,” accessed December 14, 2021, https://www.currentmarketvaluation.com/models/buffett-indicator.php. Over time then, the $100 billion line would grow more slowly than the market, encompassing even more firms. Even now, the $100 billion mark is probably too expansive, since it includes Rivian, a company that just went public last month. At a minimum, the law should be indexed to a bundle of market prices instead of GDP.
3 A more benign alternative
In a hearing before the FTC, former director of the Bureau of Economics, Jonathan Baker laid out what he called the benign alternative.5 “Competition and Consumer Protection in the 21st Century,” transcript, Federal Trade Commission Hearings Session 1, September 13, 2018, https://www.ftc.gov/system/files/documents/public_events/1398386/ftc_hearings_session_1_transcript_9-13-18_1.pdf Instead of being an economics problem that could be solved by regulation or changes in policy, Baker suggested that market power might be best explained by larger changes in demographics. He went on to argue against it, but the idea of a benign alternative shouldn’t be dismissed out of hand. It goes far to explain a lot of features of our economy.
This less nefarious story fits all of the facts as well. It is one in which larger demographic changes have given rise to large firms that seemingly possess market power. These long run demographic changes act as confounding variables, affecting both measures of market power and economic outcomes. While laxer antitrust enforcement might have contributed to the rise of market power, the impact is probably still small given that changes in demographics, education, and software go far in explaining the rise of large firms.
There is very little disagreement on the broadest of trends regarding concentration. More than 75 percent of U.S. industries have seen increases in concentration levels throughout the 2000s.6 Gustavo Grullon, Yelena Larkin, and Roni Michaely, “Are U.S. Industries Becoming More Concentrated?,” 2018, https://doi.org/10.2139/ssrn.2612047. These measures are typically expressed by the Herfindahl-Hirschman Index (HHI), which is calculated by squaring the market share of each firm in a market and then summing them all.
European countries have also seen a general rise in concentration at the industry level, albeit to a lesser extent than the U.S.77 “Concentration in the EU: Where It Is Increasing and Why,” ProMarket (blog), May 19, 2021, https://promarket.org/2021/05/19/concentration-eu-antitrust-market-power-barriers-entry. Not to be left out of the party, Japan is facing a rise in firm concentration as well.
But it is the magnitude of this change, and how it impacts innovation and consumers prices, that is vitally important and hotly debated. According to Atkinson and de Sousa’s (2021) analysis of Census data, the average sales that the top four firms captured, known as the C4 ratio, increased just 1 percent from 2002 to 2017.8 “Just 4 Percent of US Industries Are Highly Concentrated, ITIF Finds in Analysis of New Census Data, Contradicting Antitrust Alarmists” (Information Technology and Innovation Foundation, June 7, 2021), https://itif.org/publications/2021/06/07/just-4-percent-us-industries-are-highly-concentrated-itif-finds-analysis-new. Since Census data is considered the most reliable source available, this analysis is critical as a starting point to understanding changes in concentration. In their sample, Atkinson and de Sousa found that 55 percent of sectors increased in concentration, while 45 percent of sectors decreased. Among those sectors that did see increases in concentration, 18 percent increased by more than 10 percentage points. Importantly, more concentrated industries in 2002 tended to become less concentrated by 2017.
Mergers, however, aren’t judged by their impact on national-level concentration ratios which is where academic papers have focused. Instead, they will undergo analysis at the local market level. In contrast to national level figures, local market concentration has been on the decline across all major sectors over the past 25 years.9 Esteban Rossi-Hansberg, Pierre-Daniel Sarte, and Nicholas Trachter, “Top Firms and the Decline in Local Product-Market Concentration,” VoxEU.Org (blog), October 19, 2018, https://voxeu.org/article/top-firms-and-decline-local-product-market-concentration. The trend seems contradictory at first, but as economists Rossi-Hansberg, Sarte, and Trachter describe it, “Top firms expand their national market share by opening establishments in new locations, thereby increasing local competition.” Local industrial concentration in particular has been on the decline for even longer, according to work by the Census, such that “by 2015, average local concentration had declined to about three quarters of its 1976 value.”10Kevin Rinz, “Labor Market Concentration, Earnings Inequality, and Earnings Mobility,” n.d., 114, https://www.census.gov/content/dam/Census/library/working-papers/2018/adrm/carra-wp-2018-10.pdf.
Competition agencies and courts have long recognized that concentration measures should be used as blunt instruments that trigger further investigation. Because market power is a vague and ill defined term, concentration ratios often become the rationale for the existence of market power. But concentration increases alone aren’t evidence of harm.
Since market power is held by the supplier of a good or service and instead, if it were the case that increases in demand caused concentration, then concentration measures would be a poor proxy for market power. Hubbard and Mazzeo (2019) discovered recently that the hotel and motel industry competed on quality by investing in swimming pools from the 1960s through the 1980s, but the effect was a rise in industry concentration.11 Thomas N. Hubbard and Michael J. Mazzeo, “When Demand Increases Cause Shakeouts,” American Economic Journal: Microeconomics 11, no. 4 (November 2019): 216–49, https://doi.org/10.1257/mic.20180040. Demand increases throughout this period were associated with fewer hotels, particularly in warm places where outdoor amenities were more valued by consumers. The demand-led concentration effect didn’t occur in other industries that serve highway travelers, like gasoline retailing or restaurants, where quality competition is either less important or quality is supplied more through variable costs. As the two economists note, these results highlight the importance and challenge of distinguishing between “natural” and “market-power-driven” increases in concentration. Hubbard and Mazzeo’s natural forces are the characters in Baker’s vision of a benign alternative.
To understand how changes in competition law will affect markets, it is important first to understand the impact of these external forces. One trend driving concentration is worker productivity. Nobel winning economist Robert Lucas predicted back in 1978 that firms would likely grow as workers got more productive.12 Robert Lucas, “On the Size Distribution of Business Firms,” Bell Journal of Economics 9, no. 2 (1978): 508–23, https://econpapers.repec.org/article/rjebellje/v_3a9_3ay_3a1978_3ai_3aautumn_3ap_3a508-523.htm. Assuming that workers in the U.S. continued to get more productive over time, Lucas anticipated that productivity increases would result in wage increases, which in turn would incentivize marginal entrepreneurs to become employees of a firm. Companies would get more productive and larger in size as workers choose those firms. Over time, as productivity and wages inch upwards, working at a firm would be incentivized over starting a company. It’s not, as the CPCA says, solely market power that makes it more difficult for people in the United States to start their own businesses. They freely choose the alternative, working at a company.
Recently Kozeniauskas (2018) found that the decline in startups has been more pronounced for higher education levels.13 Nicholas Kozeniauskas, “What’s Driving the Decline in Entrepreneurship?,” n.d., 57, https://nicjkoz.com/files/JMP_Kozeniauskas.pdf. This implies that at least part of the force driving the changes is not skill-neutral, supporting Lucas. Competition doesn’t drive small business growth. Competitives drives out businesses from the market. The formation of new companies, or entrants, is orthogonal to competition and instead is driven largely by education and population growth.
Work from Karahan, Pugsley, and Şahin (2019) helps to round out the story.14 Fatih Karahan, Benjamin Pugsley, and Ayşegül Şahin, “Demographic Origins of the Startup Deficit,” Working Paper, Working Paper Series (National Bureau of Economic Research, May 2019), https://doi.org/10.3386/w25874. As they found, the long-run decline in the startup rate was caused by a slowdown in labor supply growth since the late 1970s. Summarizing their findings, the authors explain that, “This channel explains roughly two-thirds of the decline and why incumbent firm survival and average growth over the lifecycle have been little changed.” Hopenhayn, Neira, and Singhania (2018) put a fine point on this by diving into aging firm distribution trends even further.15 Hugo Hopenhayn, Julian Neira, and Rish Singhania, “From Population Growth to Firm Demographics: Implications for Concentration, Entrepreneurship and the Labor Share,” Working Paper, Working Paper Series (National Bureau of Economic Research, December 2018), https://doi.org/10.3386/w25382. Their work, which is supported by others, suggests that these trends largely explain the concentration of employment in large firms as well as changes in average firm size and exit rates, which are key determinants of the firm entry rate.16 Michael Peters and Conor Walsh, “Population Growth and Firm Dynamics,” Working Paper, Working Paper Series (National Bureau of Economic Research, October 2021), https://doi.org/10.3386/w29424. Slowing population growth means fewer startups, which means large companies are less likely to be challenged by an upstart.
Meanwhile, the information technology revolution caused some firms to adopt new technologies and become more productive. Since the late 1970s, firm inequality has grown, largely due to between-firm variance.17 Jae Song et al., “Firming Up Inequality,” The Quarterly Journal of Economics 134, no. 1 (February 1, 2019): 1–50, https://doi.org/10.1093/qje/qjy025. Broadly speaking, above-average firms have gotten better at managing assets, while the laggards in an industry tend to be doing worse over time.
Companies on the edge of the productivity frontier pay their employees higher wages, resulting in earnings inequalities. Driven by productive teams and technology, frontier firms were able to take off in retail, manufacturing, mining, services, telecommunications, and in the information sector throughout the late 1990s and into the 2000s. Frontier firms might be bigger, but they tend to be more productive as well. 1818 Dan Andrews, Chiara Criscuolo, and Peter N. Gal, “Frontier Firms, Technology Diffusion and Public Policy: Micro Evidence From OECD Countries,” The Future Of Productivity: Main Background Papers (Organisation for Economic Co-operation and Development, 2015), https://www.oecd.org/economy/growth/Frontier-Firms-Technology-Diffusion-and-Public-Policy-Micro-Evidence-from-OECD-Countries.pdf. Pay differences within a company have remained virtually unchanged. What’s changed is that some firms have charged to the front of productivity. Indeed, virtually all of the rise in earnings dispersion from 1978 to 2012 is due to some firms sorting.1919 Song et al., “Firming Up Inequality.” So it is not a market power story that explains wage inequality, but a firm sorting story.
Economists tend to describe market power in a narrow way, as being synonymous with increasing markups. Markups are a measure of a good’s price over its marginal cost. A good that costs $1 to produce but is sold for $1.10 has a 10 percent markup. Looking at markups across products reveals trends in the pricing power of a company. Simply put, pricing power in all of its variations means market power.20 The term price means broadly the terms of trade, which are outcomes of the equilibrium market process, and that may include, for example, prices, auxiliary fees, qualities, product variety, and service. But measuring costs poses difficulties, which explains why there is vast disagreement about markup trends. By and large, however, these firms expanded output of goods.
In work just restricted to large firms, Hall (2018) found substantial heterogeneity in the rise of markups across sectors and in growth rates. Like the rest of the research in this genre, it presents a muddled picture. While there is no evidence that mega-firm intensive sectors have higher mark-ups, there is some evidence that markups grew in sectors with rising mega-firm intensity.21 Robert Hall, “New Evidence on the Markup of Prices over Marginal Costs and the Role of Mega-Firms in the US Economy” (Cambridge, MA: National Bureau of Economic Research, May 2018), https://doi.org/10.3386/w24574. Since these situations tend to trigger merger review, it is unclear what current enforcement is actually missing in the current regime.
4 FTC enforcement and the cost of going public
While there is a growing body of evidence on concentration and large firms, empirical research on the effectiveness of strong merger enforcement is seriously lacking. It is important to separate out the various roles of competition agencies. Merger enforcement is a distinct part of competition enforcement. The DOJ and FTC could get stronger on other areas of competition enforcement like price fixing or unilateral effects without touching mergers.
Similar to research on concentration, there is broad agreement in the enforcement trends in mergers. Both FTC and DOJ enforcement actions on company deals have gone down over the past three decades, stretching back to the beginning of publicly released data in the 1990s.22 “The State of U.S. Federal Antitrust Enforcement,” Equitable Growth (blog), accessed December 11, 2021, http://www.equitablegrowth.org/research-paper/the-state-of-u-s-federal-antitrust-enforcement/. But the number of enforcement actions taken in the 2000s has largely remained unchanged. While the bill singles out 2008 as the moment when enforcement shifted, the FTC data suggests that there was little change around those years. In the 10 years before 2008, the FTC engaged in 16.3 enforcement actions each year; in the 10 years after, there was an average of 17.5 actions.
Fewer cases might mean that merging parties and the federal competition agencies know what kinds of deals will win in court. Even with its different standards for mergers and arguably bigger staff, the European Union brings about the same number of actions against companies merging as the United States.23 European Court of Auditors, The Commission’s EU Merger Control and Antitrust Proceedings : A Need to Scale up Market Oversight, Special Report No. 24/2020 (European Court of Auditors. Online) (LU: Publications Office, 2020), https://data.europa.eu/doi/10.2865/10520. It could be that the current level of merger enforcements is the optimal number.
Predicting the bill’s impact on mergers poses some difficulties because the law would substantially change standards across the board while also shifting the burden completely for large companies. Still, we should expect that a major channel of innovation would be shut off.
As much as there are broader forces at work contributing to the big trends, the U.S. startup market is incredibly vibrant compared to other countries because it embraces risk to spark innovation.24 Jad Esber, “Entrepreneurship Ecosystems in Europe vs the US,” The Startup (blog), October 6, 2019, https://medium.com/swlh/entrepreneurship-ecosystems-in-europe-vs-the-us-c0d096f05ca6. Much of this is driven by the exit. Entrepreneurs expect that they are going to be bought up, some 58 percent, while only 17 percent expect to take their company public.2525 “2020 Global Startup Outlook,” Silicon Valley Bank Financial Group, n.d., https://www.svb.com/globalassets/library/uploadedfiles/content/trends_and_insights/reports/startup_outlook_report/suo_global_report_2020-final.pdf. It is unlikely that CPCA’s changes to the law would spark the creation of new startups or cause nascent companies to compete with larger firms. More than likely, those companies wouldn’t form in the first place because they wouldn’t have the possibility of exiting.
There is nothing inherently worrying about the current merger and acquisition system. Indeed, the vast majority of acquisitions by large tech companies are made either for the tech or the talent of the target company, not to stifle a future competitor.2626 Will Rinehart, “Welcome to the Kill Zone?,” The Benchmark (blog), February 27, 2020, https://www.thecgo.org/benchmark/welcome-to-the-kill-zone/. As I wrote previously,
Both parties benefit from these deals. Startups often have great ideas but lack the technical and marketing resources to bring a product to a wider audience. Large companies, on the other hand, have the resources and consumer base for a new product but often lack innovative ideas. Yet, the true benefactors of these deals are consumers who can now choose an innovative product that may not have existed otherwise.
Empirical research in the pharmaceutical industry suggests that about 6 percent of acquisitions aim to acquire innovative projects and preempt future competition.27 Colleen Cunningham, Florian Ederer, and Song Ma, “Killer Acquisitions,” SSRN Scholarly Paper (Rochester, NY: Social Science Research Network, April 19, 2020), https://doi.org/10.2139/ssrn.3241707. Because this industry relies so heavily upon patents, the 6 percent finding should be considered a high level mark for the total number of concerning deals. Importantly, the authors explain that these acquisitions disproportionately occur just below current thresholds for antitrust scrutiny. If their findings are accurate, enforcement standards don’t need to change. Rather, more attention should be focused below the current threshold lines.
Merger enforcement doesn’t need dramatic changes to adapt to the changing times. Given that most large advanced economies are experiencing increased concentration, in spite of different competition regimes, leaders should be skeptical of proposals to rewrite antitrust law. Given the emergence of superstar firms and the related demographic trends, CPCA faces serious headwinds. It is unlikely to change industries and reverse those trends because lax merger enforcement is not likely the cause.