Monopolies and Antitrust Enforcement
With the FTC suing Amazon for anti-competitive practices, there appears to be a renewed focus on antitrust enforcement. What does economic research think about it?
Thank you for reading our work! If you haven’t yet subscribed, please subscribe below:
If you would like to support us further, please consider sharing and liking this article!
As Nominal News grows larger, we will be able to make this a full-time project, and provide more content. If you would like to suggest a topic for us to cover, please leave it in the comment section.
The US Federal Trade Commission (FTC) recently sued Amazon for undertaking anti-competitive practices that have restricted market competition and adversely affected consumers. Interestingly, the head of the FTC, Lina Khan, prior to her current role, wrote a highly praised scholarly article on the issue of Amazon and antitrust in the Internet age. Let’s look into the key arguments being made in the FTC case and then turn to economic research on the topic of market concentration and the value of antitrust enforcement.
Amazon FTC Case
The FTC is arguing that Amazon has enacted multiple anti-competitive practices that have hurt both consumers and sellers. One of the key points focuses around the role Amazon performs as a ‘marketplace’ – that is where sellers and buyers can find each other and transact. The Amazon platform – Amazon Marketplace – acts as an intermediary where sellers can post their product for sale, while buyers can search and browse for products.
The results Amazon shows to potential buyers are driven by a search and recommendations algorithm that typically suggests products where the seller has paid a fee to Amazon to advertise their product, rather than showing organic search results. Since most suggested products are advertised products, if a seller wants to make any sales, they must pay Amazon to advertise their product (on top of any other platform fees Amazon charges). This has created a ‘pay-to-play’ scenario, as without paying this fee, sellers will, in practice, not have any sales. Moreover, these advertising costs also end up hurting consumers, as these costs end up being pushed onto the consumer.
Amazon also punishes sellers that post lower prices of their goods on other marketplaces or their own website, even if they have lower operational costs on those other platforms. The punishments entail pushing down the appearance of the seller’s product in search results, hiding the product's price on Amazon and disabling certain features for that seller, all with the intent to discourage sales of that seller.
On products that Amazon directly sets the price, Amazon also uses algorithms that are deemed by the FTC to be anti-competitive. One algorithm scours the internet for prices of the same product Amazon offers and matches the price if another marketplace offers a lower price. This prevents other sellers from ever building a competitive presence, as Amazon is willing to incur losses on the sale if need be. A different algorithm, that is occasionally turned on, attempts to determine if competitors would follow suit if Amazon increased a price on a product. The algorithm would then increase the price of a product, and other sellers or marketplaces (that may have their own automated algorithms) would increase prices shortly after. In this scenario, consumers are worse off as the price is permanently increased.
Lastly, Amazon also forces sellers to use Amazon’s distribution service for fulfilling orders. Even if the seller may have their own distribution solution or use a separate warehousing company that may be cheaper, Amazon does not allow them to use these alternatives. Amazon claims that these sellers have a high delivery failure rate, which is why the sellers may end up being removed from the Amazon Marketplace. Sellers, however, claim this is not true.
Monopolies, Competition and Investment
The main reason we are concerned with monopolies and anti-competitive behavior is its impacts on total economic output and societal welfare. Most economic theory predicts that overly large market concentration is detrimental to overall output and welfare. For example, Guttierez and Philippon (2017) looked at how lower market competition impacts overall levels of investment in capital stock (for example, manufacturing machines). Sub-optimal levels of capital stock result in lower output, and reaching optimal levels of capital stock would improve outcomes for all.
In a perfect world, the value generated by capital in a firm (for example, the value of the output generated by a machine in a factory) should equal the cost of the capital (the price of the machine in the factory). The ratio between these two values is referred to as Tobin’s Q. If the generated value of the capital is greater than the cost of the capital (Tobin’s Q above 1), we should give this firm more capital, as it is generating extra value. Firms, however, when making decisions, do not take into account what is the optimal outcome for society and focus on maximizing profits. Thus, they may spend more effort in preventing other firms from entering and investing in the market (for example, by erecting barriers to entry) rather than investing in capital and innovation.
In the US, based on the Tobin Q metric, there is currently under-investment in capital stock (i.e. Tobin’s Q above 1) by about 10%. Under-investment has been occurring every year since the turn of the century. Guttierez and Philippon argue that part of this under-investment can be explained by lower competition. They present the chart below to illustrate the change in competition levels in the US:
The top chart shows that the US has experienced a decline in both entry (creation) and exit (destruction) of firms, meaning there is less competition and innovation occurring.1 The bottom chart shows the Herfindahl index (the green one is the standard index, while the orange one is adjusted for actual shareholders – that is, two firms that have the same owner are actually one firm). The Herfindahl index is a measure that goes from 0 to 1 and aims to capture the size of firms in relation to the industry. Larger values imply higher concentration – that is, fewer firms control more of the output (at a value of 1, all output in an industry is produced by 1 firm). Values above 0.25 typically mean an industry is highly concentrated. In many countries, if firms in industries that have a Herfindahl index above 0.25 attempt to merge, they will come under legal scrutiny.
Increased competition forces companies to innovate and develop by investing in capital, as without this investment, these companies are at risk of being pushed by other companies that are innovating. Thus, if the level of competition drops, firms have less incentive to invest and innovate. Additionally, aggregate investment may be lower since there are fewer companies in the market than there should be due to anti-competitive behavior of other companies. Looking at this data and controlling for other factors, Guttierez and Philippon found that some of this US under-investment can be attributed to falling levels of competition.2
Concentration – Can it be good?
Concentration is not necessarily bad. Concentration and monopolies may arise naturally if firms are very ‘productive’ – that is, they efficiently convert inputs such as raw materials and labor into outputs. Boar and Midrigan (2023) theoretically demonstrate that it may be in everyone’s interest to increase concentration under certain circumstances. By providing more capital to the most productive firm, overall output will be higher than if the capital was distributed among many firms, as these other firms are less productive. Since output is higher, everyone in the economy is better off. However, this rests on a key assumption – that firms in this economy are owned entirely by all citizens in this economy. Without this assumption, the result no longer holds, as the most productive firm will generate profits that might not be distributed back to the citizens equally.
Department of Justice (DOJ) and Antitrust
Although the question of optimal levels of concentration have not yet been resolved by economists, we can study what has been the impact of antitrust enforcement. Babina, Barkai, Jeffers, Karger and Volkova (2023) looked at the impacts of US DOJ antitrust cases. By manually collating data on DOJ antitrust cases, Babina et al. were able to study how these antitrust cases impacted the industries in which these cases were brought. DOJ antitrust enforcement actions resulted in an increase in employment by 5.4% and new business formation by 4.1% in the industries antitrust actions were undertaken compared to industries that did not have any antitrust enforcement. Additionally, total payroll increased by more than employment, implying that worker’s average wages increased.
Babina et al. were not able to directly measure whether antitrust action reduced prices of goods or increased quantities sold, which would both be positive for society. However, they did find that sales, measured in dollars, in the impacted industries did not change. Given that the industries employed more people, Babina et al. argue that output must have gone up (more inputs should lead to more outputs). Since output is higher, but total sales measured in dollars remained the same, it must then mean that prices went down. Overall, it appears that DOJ’s antitrust enforcement has positively impacted economic activity in the targeted industries.
Conclusion
The Amazon lawsuit brought by the FTC may be one of the most important antitrust cases in recent history, as it may set a precedent for future antitrust enforcement actions. It is unclear how this particular legal case will pan out, especially since law doesn’t always follow economics. On average, the DOJ antitrust cases have had a positive impact on the economy. This is a crucial finding, as recent history suggests that the number of antitrust cases in the US has fallen significantly – Grullon, Larkin and Michaely (2016) show that in the period from 1970 to 1999, there were on average 16 antitrust cases a year, while in the period from 2000 to 2014, that number dropped to 3 cases a year. The damages from lack of effective antitrust enforcement can be quite significant. Guttierez and Philippon discuss research which has shown that if the mobile telecommunications sector in the US was as competitive as in Germany, consumers would gain about $65bln a year in value.
Although economists have been able to measure impacts of competition on particular industries, the question of the optimal levels of concentration is unresolved. Market concentration due to higher productivity can potentially be a benefit to all market participants. Preventing such concentration might have unintended consequences whereby innovation and investment is reduced. At the same time, effective antitrust enforcement can improve economic and welfare outcomes.
In the US, the evidence suggests that increasing competition (and reducing concentration) currently would improve welfare. Economy-wide, we’re not yet at the situation where preventing concentration discourages innovation and entry of new companies, which would be welfare reducing. However, antitrust enforcement might not be beneficial in a particular industry, which is why determining when to pursue and when not to pursue antitrust actions is an important decision.
Interesting Reads from the Week
- discusses the latest CPI data by removing the Owner’s Equivalent Rent (which is an imputed calculation) suggesting that October CPI is at 1.5%.
Note: My note on an interesting discussion of the issue of Universal Basic Income and Inflation.
Tweet/X: The recent recovery from the pandemic has not been good for people with college degrees, as real wages for this category of individuals are still below pre-pandemic levels.
News: The removal of Sam Altman as CEO of OpenAI and his hire by Microsoft shows how economically beneficial it is to not enforce non-competes. We previously covered why non-competes hurt everyone:
Photo by Ylanite Koppens.
If you enjoyed this article, you may also enjoy the following ones from Nominal News:
The End of Inflation in the US? (June 25, 2023) – after the inflationary surge that occurred during 2021-2023, the June 2023 inflation data suggested that US inflation has come back down to the 2% target. Did it come down due to the Federal Reserve interest rate hiking policy or was it driven by the normalization of supply chains?
Work Requirements (May 29, 2023) – work requirements are considered to be a helpful policy to incentivize people to work. Empirically, the data shows that this is not the case. It does not increase participation in the workforce and hurts the most vulnerable individuals.
Community Civilian Interventions (June 12, 2023) – cities have always been interested in improving safety in neighborhoods. Civilian government interventions have been proposed as a solution instead of police. However, civilian interventions work best if local government presence is strong.
New firms are typically more productive and innovative than old, established firms. Therefore, in a competitive market, we should expect to see high firm turnover – that is a lot of new firms appearing on the market, and a lot of old firms disappearing from the market.
This result was even more pronounced in industries where US companies faced competition from Chinese companies in the 90s. Although competition from China in particular industries reduced the number of US firms in these industries, the surviving US firms significantly increased their investment levels, demonstrating that increasing competition increases capital investment.
For hundreds of years, this nation did lots of antitrust at the state and sometimes even local level, that should be restored. It was also (and I mean compared to today, I don't mean some kind of Borg Collective type "universal democracy") done at the unmeasurably and unknowably generated directions of a great deal of different interests groups and, despite what they say, sometimes groups of "regular" people who studied issues and sometimes had great successes. It was a democratic republic once, an imperfect one, but an improving one, and by the late 1970s we'd abandoned it for a technocratic dictatorship that has performed far worse.