Antitrust regulation is likely to require more ex post control than in previous decades. Past acquisitions that have led to monopoly power are natural candidates for a breakup. Marketplaces also require strong remedies. In the case of Apple, many recent controversies center around its App Store, where prices are hidden, and rules are obscure and rife with conflicts of interest. Amazon is both a market (the online retail platform) and a market participant (with its own brands), and this situation also creates conflicts of interest.
Two Catch-22s
I see one reason for optimism and two catch-22s with the GAFAMs, all involving the use of data. The first catch-22 involves big data and free entry. As I have argued, big data can lead to powerful price discrimination. This can make markets more efficient, but it can also shift the entire surplus to the pockets of the monopolist. With big data, contestability becomes more important. The catch is that big data is in itself a barrier to entry.
The second issue involves economic footprints and privacy. As I have argued in the previous chapter, one reason the GAFAMs have not boosted the growth rate of the US economy is that their footprint is smaller than that of previous generations of star firms. This leads to the second catch-22. If they remain mostly as they are, their impact on growth will remain small and disappointing. If they branch out and increase their footprint, they could have a meaningful impact on aggregate productivity. But if they increase their footprint, the privacy issues worsen.
Google and Facebook are currently just large advertising machines. They disrupted many advertising firms and nearly all the newspapers. This does not create significant aggregate productivity, but it does create political and democratic issues. The same is true for Apple. The iPhone makes it more convenient to access digital content while traveling. It is nice, but if that’s all it does, it will not move the needle of aggregate productivity. Moreover, the market for iOS apps is opaque, and conflicts of interest are pervasive.
To have a meaningful impact on economic growth, the GAFAMs or their siblings need to improve the markets that really matter: transport, energy, and health. If Google really helps to create a market for effective driverless cars, if Facebook really disrupts the banking system, if other digital technologies enable the efficient provision of health services, then we will see real, widespread benefits.
But here is the catch-22. In all these cases, privacy issues are paramount. In these new markets, the GAFAMs would have access to even more personal data, and data that are even more sensitive than what they can access today. If we already find it difficult to trust them with the data they have now, how will we feel about direct data feeds to private companies from our car or our health providers?
It is not hopeless, however. Technology can help solve the problems that it creates. For instance, artificial intelligence can also be used to fight off hackers and contain data breaches. Technology can make it easier and faster to assign blame for cyberattacks. Tech reporter Adam Janofsky notes that corporations are “using machine learning to sort through millions of malware files, searching for common characteristics that will help them identify new attacks. They’re analyzing people’s voices, fingerprints and typing styles to make sure that only authorized users get into their systems. And they’re hunting for clues to figure out who launched cyberattacks—and make sure they can’t do it again.”e
Competition can also help improve privacy, as Dina Srinivasan (2019) shows: “Facebook tried to renege on its promise not to track users in 2007, and again in 2010, but the market was competitive enough with adequate consumer choice to thwart Facebook’s attempts.” It was only after it acquired a monopoly position that Facebook could afford to disregard the privacy concerns of its users.
The good news is that the sharpest minds in policy circles are focused on this issue. The UK Chancellor of the Exchequer established a panel of experts in September 2018, led by a former chair of the Council of Economic Advisers, Jason Furman. This panel’s influential report tackles some of the complex issues surrounding regulation of the digital economy and proposes a set of principles to guide policy decisions (Furman et al., 2019). Its policy recommendations emphasize “measures to promote data mobility and systems with open standards, and expanding data openness.”
New technologies have the potential to revive productivity growth, but they have yet to deliver. The GAFAMs can be at the forefront of this push if we find a way to deal with the issue of data protection. We must ensure that big data does not present a barrier to entry, and we must make certain our privacy is protected.
* * *
a See Gabriel Zucman’s November 10, 2017, opinion piece in the New York Times, “How corporations and the wealthy avoid taxes.”
b Stack Overflow is a site where developers can ask for help from their peers. It has about 9 million users, 16 million questions, and 25 million answers per year. GitHub is at the heart of open-source software development. It is a code repository with around 25 million users and 50 million repositories. Microsoft announced in June 2018 that it would buy GitHub for $7.5 billion. The move was expected to help Microsoft compete against Amazon Web Services.
c Omarova is quoted in Rana Foroohar, “Banks jump on to the fintech bandwagon,” Financial Times, September 16, 2018.
d The GDPR requires companies to notify regulators of breaches within seventy-two hours. In the US there is no federal breach notification law. Instead, companies navigate a patchwork of state laws.
e Adam Janofsky, “How AI can help stop cyberattacks,” Wall Street Journal, September 18, 2018.
CHAPTER 15
Monopsony Power and Inequality
Masters are always and everywhere in a sort of tacit, but constant and uniform combination, not to raise the wages of labour above their actual rate.
ADAM SMITH, THE WEALTH OF NATIONS
THERE ARE TWO TYPES of market power: monopoly and monopsony. Monopoly power is better known. A firm has monopoly power when it can charge a high price for its products because its clients have few other choices. It’s easy to visualize, and we have discussed its implications.
A firm has monopsony power when it can exert market power on its employees and suppliers because they have few other places to sell their labor or their goods and services.
When I was in graduate school, monopsony power was deemed so irrelevant that the subject was dropped from the standard coursework. The modern models that we use to think about the economy do not include monopsony power. Monopsony was “so nineteenth century.”
Much to my surprise, however, the topic has come back. There is evidence of growing monopsony power in several local labor markets across the US. And the pricing power of internet platforms—just like that of credit card companies—is also a form of monopsony, directed toward suppliers more than toward consumers.
Monopsony and monopoly have different sources but similar implications for the broad economy. Imagine a world where production requires only labor: one worker produces one unit of a good. Think about a monopoly charging a 50 percent markup over the wage. If the competitive wage is 1, then the price of the good is 1.5. If there are 100 workers in the economy, they produce 100 units of the good. The nominal GDP is $150, labor income is $100, and capital income is $50. Workers earn two-thirds of GDP, and capital owners earn one-third. But workers are also consumers. Each worker earns 1 but can only purchase 1 / 1.5, or 2 / 3 of a unit of the good.
Now imagine a monopsony. The price of the good is 1, but the wage is pushed down to 2 / 3 instead of 1 because workers have nowhere else to go. Notice that the outcome is exactly the same as before. Workers earn two-thirds of GDP and capital owners one-third. The worker can only buy two-thirds of the good, though in the perfectly competitive economy she would be able to buy a full unit.
At the aggregate level, therefore, monopsony and monopoly have the same implications for workers’ standard of living. It does not matter whether firms mark up their prices by 50 percent or whether employers p
ush down wages by 30 percent. In both cases the purchasing power of workers is 30 percent lower than it should be.
Labor Market Concentration
A few recent papers argue that monopsony power is coming back in the US labor market. The idea is relatively straightforward. If potential workers have the choice between only a handful of employers, then the employers have market power over the workers and can offer lower wages.
The first question, then, is this: how concentrated are labor markets? José Azar, Ioana Marinescu, Marshall Steinbaum, and Bledi Taska (2018) look at job vacancies posted online and collected by an analytics software company in 2016. They compute an index of labor market concentration using the Herfindahl-Hirschman index (HHI) for each commuting zone by six-digit federal Standard Occupational Classification codes. The average market has an HHI of 3,953, or the equivalent of 2.5 recruiting employers in the market. More than half of labor markets are highly concentrated with an HHI above 2,500—the cutoff in the DoJ / FTC guidelines we discussed in Chapter 2. Highly concentrated markets account for 17 percent of employment. When the authors consider other, plausible alternative market definitions, they find that the fraction of highly concentrated markets is never less than one-third.
The next question is whether concentration pushes down wages. It is difficult to measure, but recent papers suggest this might be the case. Economists Efraim Benmelech, Nittai Bergman, and Hyunseob Kim (2018) study the effect of local labor market concentration on wages using Census data over the period 1977–2009. They argue that local employer concentration has increased over time, and there is a negative relation between local employer concentration and wages. This is consistent with the idea that employers have monopsony power in concentrated labor markets. Moreover, they find the relation between labor market concentration and wages is more negative at high levels of concentration and when unionization rates are low. But wage growth is more tightly linked to productivity growth when labor markets are less concentrated.
Interestingly, they also find that concentration has increased in markets with greater exposure to import competition from China. These markets have become competitive on the product side, and we expect monopoly rents to have decreased. Unfortunately, the concentration might have triggered an increase in monopsony power, thus canceling some of the benefits of free trade. This is another example of the broad point that I made in the Introduction. There is a difference between domestic and foreign competition. Promoting domestic competition should be an absolute no-brainer. Foreign competition is more complicated and less clear-cut.
The literature on monopsony power is still in its infancy, and there is much more we need to learn. There is broad agreement that labor market power hurts workers. David Berger, Kyle F. Herkenhoff, and Simon Mongey (2019) estimate welfare losses from labor market power that range from 2.9 to 8.0 percent of lifetime consumption. There is no consensus, however, regarding the evolution of labor market concentration. Berger, Herkenhoff, and Mongey find that local labor market concentration has declined over the last thirty-five years despite the overall increase in national concentration. This brings us back to the issue that we discussed in Chapter 2. In Box 2.2 we showed that national concentration measures can differ from local concentration measures and give a misleading picture of the economy.
You might think that local labor market concentration matters less today because of online labor markets. Arindrajit Dube, Jeff Jacobs, Suresh Naidu, and Siddharth Suri (2018) study exactly this issue. They examine one of the largest on-demand labor platforms, Amazon Mechanical Turk. Online platforms make it easier to search for a job, and one might have conjectured that they would lead to near-perfect competition. But the authors find a surprisingly high degree of market power, even in this large and diverse spot-labor market, suggesting that much of the surplus created by this online labor market platform is captured by employers.
Restricted Contracts
We have already seen how restrictive contracts are used by large hospitals to reduce competition in the health-care market. Similarly, large franchise employers use restriction in labor contracts to limit competition in the labor market. Princeton economists Alan B. Krueger and Orley Ashenfelter (2018) studied the role of covenants in franchise contracts that are commonly used by large firms, such as McDonald’s, Burger King, Jiffy Lube, and H&R Block.
The covenants are meant to restrict the recruitment and hiring of employees from other units within the same franchise chain. In other words, the covenant prevents a McDonald’s franchise in a particular location from poaching employees from another nearby location. Krueger and Ashenfelter find that almost 60 percent of major franchisors’ contracts include these “no-poaching agreements” and that these agreements are more common for franchises in low-wage and high-turnover industries. They clearly decrease competition in the labor market and thus decrease wage growth.
The good news is that these labor market issues seem to finally be getting some attention. Twenty state attorneys general are looking into franchise noncompete hiring clauses in McDonald’s franchise contracts based on Krueger and Ashenfelter’s work.
Occupational Licensing
Geographic mobility has been declining for thirty years in the US. Workers are less likely to move between states and metropolitan areas now than they were in the past. There are several plausible explanations for this trend. One of them is the steady increase in the number of workers whose occupations require some sort of license or certification (Davis and Haltiwanger, 2014).
Morris M. Kleiner and Alan B. Krueger (2013) track the historical growth in licensing from a number of different data sources. They find that the share of the US workforce covered by state licensing laws grew fivefold in the second half of the twentieth century, from less than 5 percent in the early 1950s to 25 percent by 2008. State licenses account for the bulk of licensing, but if we add locally and federally licensed occupations, the share of the workforce that was licensed in 2008 reaches 29 percent.
The increase in licensing has two causes. The first is the growth in the number of employees in occupations that typically require a license over the last few decades. The field of health care contains many such occupations. But this only accounts for about one-third of the increase. At the same time, according to analysis from a 2015 report from the Council of Economic Advisers, there has been a large jump in the number of occupations that require licensing because of newly imposed licensing requirements. This expansion of licensing requirements across occupations explains two-thirds of the growth.
Licensing is always “officially” motivated by concerns for health, safety, and consumer protection. And sometimes it is legitimate. Often, however, it is the perfect way for incumbents to protect their rents. Indeed, they actively lobby for the extension of licensing requirements because they understand that these are efficient barriers to entry.
We have seen earlier that Europe has reduced barriers to entry in many industries. Some of that decrease was driven by a rollback of unnecessary licensing. There is still much illegitimate licensing in Europe, but at least the trend is in the right direction. Over the same period, however, the US has increased its licensing.
Inequality: The Rise of the Club Economy
There has been a large increase in inequality in the US economy, and there are many ways to account for it. For instance, we can look at the returns to education. They have increased over time. This implies that the earnings gap between educated workers (say, those with a college degree) and less-educated ones (say, high school dropouts) has increased over time.
We can look at the split of income between capital and labor. We have shown that the labor share has decreased while firms’ profits and payouts have increased. The concentration of equity ownership is higher than that of human capital. When profits increase, they disproportionately benefit the few households with large equity portfolios. Concentration and market power therefore increase inequality.
Another way to break down the i
ncrease in inequality is to look at within-versus-between-firm inequality. Jae Song, David J. Price, Fatih Guvenen, Nicholas Bloom, and Till von Wachter (2019) use a massive, matched employer-employee database for the United States to analyze the contribution of firms to the rise in earnings inequality from 1978 to 2013. They find that one-third of the rise in the variance of earnings occurred within firms, whereas two-thirds of the rise occurred due to a rise in the dispersion of average earnings between firms.
However, this rising between-firm variance is not accounted for by the firms themselves but rather by a widening gap between firms in the composition of their workers. This compositional change can be split into two roughly equal parts: high-wage workers became increasingly likely to work in high-wage firms, and high-wage workers became increasingly likely to work with other high-wage workers. In other words, sorting increased and segregation rose.
They also find that two-thirds of the rise in the within-firm variance of earnings occurred within mega (10,000+ employee) firms, which saw a particularly large increase in the variance of earnings compared to smaller firms.
The Dangers of Monopsony Power
We have seen that there is a rise in monopsony power in the labor market. There is also the risk of a rise in monopsony power in online platforms. Large platforms have monopsony power with respect to their suppliers, even if they do not have much monopoly power on the consumer side.
The Great Reversal Page 30