Category Archives: Competition

Review of Michael Munger, The Sharing Economy: Its Pitfalls and Promises (Institute of Economic Affairs, 2021)

Transaction costs are one of the most overlooked ideas in economics. They are also one of the most important. The lowering of transaction costs is an engine of modernity itself and key to understanding where future progress might take us. That is Duke University economist Michael Munger’s argument in his new book, The Sharing Economy: Its Pitfalls and Promises (free download from the Institute of Economic Affairs). It follows up his 2018 book, Tomorrow 3.0.

What are transaction costs? As the name implies, they are things that get in the way of making transactions. Think of them as economic friction. Things like waiting in line, searching for a product, comparing prices, driving to and from a store, or resetting another forgotten website password. Transaction costs often cannot be measured in money, but they are still part of the price of everything we buy. A good economist knows, money is not everything.

Countless beneficial transactions never happen because the time and hassle required outweigh the benefits. Successful entrepreneurs can make these lost transactions come to life just by lowering their attendant transaction costs. In a way, they succeed by making the invisible visible. Along the way, they can create new industries, revolutionize existing industries, and topple old ones. Transaction costs are the hidden engine of Joseph Schumpeter’s creative destruction.

Munger’s dissertation advisor was the Nobel laureate Douglass North, one of the pioneers of transaction cost economics, along with other laureates such as Ronald Coase and Oliver Williamson. Munger likes to tell a story about North explaining that, while there are endless questions to ask in economics, many of them have the same answer: transaction costs. Over the years, North’s advice has proved useful to countless graduate students in search of thesis topics. As The Sharing Economy shows, there remains plenty of uncharted territory. Today’s graduate students should take note—as should experienced scholars.

How do transaction costs apply to the new sharing economy? Broken down to fundamentals, Uber doesn’t sell taxi rides or food delivery. It sells access to a platform that drastically lowers transaction costs for some services. The product is the platform. Yes, people use it to buy and sell taxi rides and food delivery, but they could use that type of platform for almost anything.

This is why a lot of sharing economy startups describe themselves as the Uber or the Airbnb for this or that service. They’re selling transaction cost savings, not whatever product or service appears in their marketing materials.

Every transaction requires what Munger calls the three Ts: triangulation, transfer, and trust. Sharing platforms can solve the three Ts quickly and cheaply:

  • Triangulation means coordinating everyone involved in the transaction. In a food delivery transaction, those are the customer, the restaurant, and the delivery driver. Until recently, solving this coordination problem was so difficult that most restaurants did not offer delivery at all. And the ones that did had to hire their own drivers to work exclusively for them. Unless business was both brisk and consistent, this was a risky proposition. Today, sharing platforms can solve triangulation problems in seconds. As a result, restaurants that might not be able to afford full-time delivery staff can now share drivers with other businesses and earn additional sales, while customers gain additional choices.
  • Transfer means making sure everyone gets paid. Uber, for example, has all parties’ payment info stored in the app. It automatically charges customers the right amount, pays restaurants and drivers, and handles tips. That is far easier than in the old days of cash, checks, or reading out your credit card number over the phone.
  • Trust is all parties having confidence that everything will go as it should. This is what ratings systems contribute. Riders can avoid drivers with low ratings and drivers can avoid problem customers. They can do this in seconds just by glancing at their ratings. On the other side of the coin, high ratings can be lucrative for vendors, giving them a greater incentive to keep customers happy than a traditional cab driver. And keeping that five-star rating can encourage more civil behavior from customers. It doesn’t pay to be a Karen.

Sharing economy platforms solve the three Ts so quickly and easily that millions of transactions that would never have happened a decade ago are now routine. This, for Munger, is a reason for optimism. Similar platforms could emerge for all kinds of goods. In fact, they probably are doing so right now in a dorm room or a garage somewhere.

Tools that spend nearly all of their time in storage could be rented out, for example. Other possibilities include office equipment, professional-grade audio and video equipment, and designer clothing. Some of these ideas might be successful. Others might be duds. People will likely find out soon enough.

Munger believes the low-transaction cost sharing economy could transform manufacturing as we know it. Factories would make far fewer goods, and what they do make would tend to be professional grade and more durable. A drill that spends 30 years in someone’s garage might get a couple hours of use over its entire lifespan, but if it’s shared, it will get far more use on a wider variety of tasks. It will need to be more solidly built and easily repaired.

That is one reason why the sharing economy has costs, as well as benefits. Both words in the phrase “creative destruction” are important. Just because the benefits outweigh the costs doesn’t mean costs do not exist. Manufacturing jobs have already declined by about a third since their 1979 peak, from about 18 million workers to about 12 million. If sharing platforms become popular for a lot of goods, that decline will be deeper and steeper. At the same time, the higher-grade goods still being made might require more skills or more automation to make, displacing less skilled workers.

Other jobs would open up in warehousing and delivery, and likely in other sectors. Munger doesn’t know what these might be, and neither does anyone else. Some of these jobs might not be appealing, might not pay as well, or may not work with some workers’ family responsibilities or other personal situations.

On the other hand, the size of the labor force has stayed remarkably consistent relative to population  throughout America’s transition from agriculture to industry, and from industry to services. While the inability to predict the future is scary, that’s no reason to keep things as they are. As Munger says on page 89, “Platforms are disruptive, but outlawing disruption has never worked.”

Transaction Costs and a Policy Revolution

As transaction cost reductions transform the economy, they also transform public policy. The problem is that public policy usually takes long to catch up. Regulations classify many workers as either employees or contractors, and treat them differently. Employee status comes with certain rules for minimum wages, benefits, and working conditions, while the rules for contractors are generally looser. It is also a false dichotomy that poorly fits workers’ needs.

If an accountant uses a platform like Taskrabbit to work for several clients, is she an employee of any of them? Does she count as Taskrabbit’s employee because it handles her payments or is she a customer who pays to access its platform? Does it matter if she commutes to an office or works from home? What if she’d rather choose her own health insurance or retirement plan? My colleague Iain Murray explored this question in his 2016 CEI paper “Punching the Clock on a Smartphone App.”

Current regulations don’t have good answers to these questions. And laws like California’s AB5 gig worker law and the proposed PRO Act at the federal level would entrench the legacy labor law model even further. All this is because some entrepreneurs thought of a way to use smartphone apps to reduce transaction costs.

In the age of COVID, sharing platforms have made it easier for workers to avoid public transportation and crowded offices. When COVID subsides, many workers will still prefer to avoid commutes and offices in favor of more pleasant surroundings like home offices, coffee shops, or smaller shared offices—which some sharing platforms offer. Regulators should think carefully before they take those options away.

Antitrust policy is in the middle of its own revolution, thanks in part to transaction costs. Sharing platforms are just another version of the old make-or-buy decision. If a company needs legal help, does it use in-house counsel or hire an outside attorney? Should a firm employ its own custodian or hire a cleaning service? The answer depends on transaction costs. If it’s cheaper to do something yourself, do that. If transacting with someone else costs less, do that. The answer is different for every company—and can change over time within a company. Sharing platforms and their lower transaction costs provide new possible answers to this age-old problem.

As of this writing, the leading food delivery platforms are DoorDash, UberEats, and GrubHub. They could buy out smaller competitors or merge with each other in the coming years, which might result in antitrust action. It shouldn’t, and transaction costs explain why.

A restaurant that wants to offer delivery has a make-or-buy decision to make. Does it hire its own driver or outsource to a sharing platform? It will go with whichever has lower transaction costs. This provides a built-in competitive check on sharing platforms that will never go away—even if one sharing platform monopolizes the entire market. If its fees cost more than the restaurant hiring its own drivers, then restaurants will opt for the latter. Several restaurants in the same city could even band together and jointly hire a driver—if regulations allow them.

The reason people use sharing platforms in the first place is because their transaction costs are lower than the alternatives. And the alternative of doing something in-house will never go away. Sharing platforms do not have market power, and never will.

Sharing platforms also do not restrain trade, which is another threshold for antitrust enforcement. They enable new trades that would never have happened otherwise, because they lower transaction costs. Before sharing apps, food delivery options in most places were limited to pizza and Chinese. Now, everyone from McDonald’s to mom-and-pop diners offer delivery. Some restaurants, such as Panda Express, are even attempting to undercut sharing platforms by creating their own app-based ordering services to avoid platforms’ fees.

Transaction Costs and the Moral Economy

Contrary to popular belief, morals are an integral part of economics. One of the discipline’s founding works, after all, is Adam Smith’s Theory of Moral Sentiments. Man is an animal that trades, to paraphrase from Smith’s other great work, The Wealth of Nations. People want to exchange, cooperate, and compete. It is our nature. But transaction costs get in the way of our nature, because they get in the way of trade.

Economists Virgil Storr and Ginni Choi, of the Mercatus Center at George Mason University, argue in their book Do Markets Corrupt Our Morals? that markets are moral playgrounds. When people enter those playgrounds, they learn how to trust and earn trust. They learn to keep their word and be polite—the late, great Steve Horwitz delighted in the “double thank you” that accompanies most transactions. People on the playground learn that the best way to get something you value is to give others things they value even more.

These are skills that take practice and repetition to develop. Transaction costs raise the cost of this moral practice. And when something costs more, people consume less of it.

If the goal is an open, civil society, then transaction cost reduction should be an important priority. Sharing economy platforms have the potential to do exactly that on a massive scale. At the same time, they are just another chapter in a long story, and hardly the final one.

Munger, by taking Douglass North’s advice, has used an old and overlooked tool to better understand new technologies and emerging economic changes. Sharing platforms have already changed the way people take cab rides, order food, and go on vacation. In the coming years, they could reshape the manufacturing sector, office culture, and even urban design, if traditional offices and downtowns continue to fall out of favor.

Transaction costs can also lead to fresh insights about labor regulations, antitrust, and other areas of public policy, as well as the overlooked symbiosis of markets and morality. Though The Sharing Economy is geared to a British audience, American readers will still get far more value from this freely downloadable book than they spend in transaction costs. While this admittedly sets a low bar, I intend it as high praise. I could not recommend this book more highly.

Download The Sharing Economy for free from IEA’s website here.

Court Rules Apple App Store Rules Do Not Violate Antitrust Laws

This press release was originally posted on cei.org.

A federal district court today ruled that Apple’s rules regarding payments on its App Store do not violate antitrust laws. The case, brought by video game maker Epic Games, alleged Apple violated antitrust laws by requiring purchases be made on its own system.

Director of CEI’s Center for Technology and Innovation Jessica Melugin said:

“With a court finding it is not a monopoly, the decision is largely a victory for Apple. The company will mostly continue to operate their private property, the Apple App Store, by the rules it wishes. Apple will not be forced to allow outside payment systems from developers and the App Store can remain the exclusive app download method on iPhones and iPads. The finding that Apple is in violation of California state law under the software giant’s prohibition on developers telling users there are alternative and cheaper payment options is along the lines of concessions it has already started to make with internal policy changes and legal settlement offers. Consumers will continue to benefit from Apple’s intact security, convenience and reliability at the App Store.”    

Senior Fellow Ryan Young said:

“The wisdom of Apple’s business practices is constantly being put to the test by consumers. Their size does not protect them from flops like the Newton tablet, its failed Ping social network, or its forgotten Pippin gaming console. Same goes for the App Store’s payment and commission policies.

“The separate question of whether Apple’s App Store is a monopoly is less debatable. Making that case requires defining Apple’s market so narrowly that real-world consumers can escape its boundaries with a dozen keystrokes or less. Before Apple booted Epic’s Fortnite game from its App Store in August 2020, roughly 90 percent of Fortnite downloads came through non-App Store vendors. Epic tried to define Apple’s market this way; the court disagreed.

“Any market is a monopoly if you define it narrowly enough. But those types of language games don’t always hold up in court. Real-world considerations keep getting in the way.” 

FTC Re-Files Facebook Antitrust Complaint

See also a CEI news release with statements from Jessica Melugin and me.

The Federal Trade Commission (FTC) submitted a revised antitrust complaint against Facebook today. In June, a judge threw out the initial complaint for not providing evidence that Facebook had a monopoly in anything. The FTC had until today to give it another try. The text of the amended complaint is here.

The new complaint has the same problem, and relies heavily on wordplay. The FTC argues that Facebook dominates the market for “personal social networking services,” which it defines in a way that excludes TikTok, Twitter, Clubhouse, Discord, YouTube, and others. By the FTC’s boutique market definition, Facebook’s biggest competitor is Snapchat.

Any market is a monopoly if you define it narrowly enough, and that is the only thing the FTC’s complaint successfully proves.

Real-world monopolies, as opposed to semantic monopolies, are characterized by rising prices, restricted supply, and slowed innovation. Facebook and its competitors show none of these characteristics. They are largely free to consumers. Advertisers pay to show their ads on Facebook and competing networks, but the prices they pay have fallen by half over the last decade—akin to a permanent 50 percent off sale compared to before Facebook got big.

Facebook is not able to restrict the supply of social networking services. New social networks are constantly rising, falling, and evolving, and Facebook cannot stop people from using them. Signing up for a competing service takes a minute or two and maybe a few dozen keystrokes. Many people also have multiple accounts on multiple social networks—how many people do you know who use both Facebook and Twitter, for example?

As for innovation, Facebook spent $21 billion on research and development over the past year. It is constantly experimenting with new features on its site in an ongoing trial-and-error process—because its competitors are, too. This is not monopoly behavior.

Nobody but lawyers are benefiting from the FTC’s ideologically charged word games. For example, a 2019 Inspector General report found that the FTC routinely pays outside experts as much as $750 per hour. In years-long antitrust cases, that can add up to millions of dollars, without creating any consumer value.

Another concern is regulatory capture. An antitrust settlement against Facebook would likely include expensive new policies involving privacy, content moderation, and more. Facebook can absorb these compliance costs; smaller startups cannot.

Facebook, with its aging user base, and seeing people under 25 moving to TikTok and other competitors, would likely be happy to negotiate such a settlement down the road. In the world of regulation, intentions and results are often very different things. Antitrust policy is no exception. The FTC’s ideological campaign is harming both consumers and the competitive process. At the very least, it should drop the Facebook case, if a judge doesn’t drop it first again.

Longer term, it is time to reconsider antitrust regulation altogether.

One of Google’s Antitrust Cases Dismissed, for Now

A District judge on Thursday dismissed a private antitrust case against Google brought by a group of advertisers. It does not affect separate cases brought by state attorneys general and the federal Department of Justice.

The dismissal is rooted in the relevant market fallacy. Essentially, the advertisers’ lawyers defined Google’s relevant market too narrowly, which leaves out important details. As the judge writes, “The Court is particularly concerned that Plaintiffs’ market excludes social media display advertising and direct negotiations.”

Essentially, the attorneys argued that Google has a monopoly over Google ads. This is true, in the same way that Ford has a monopoly over Ford-branded cars. But just as car buyers can easily buy a Toyota or a Chevy despite this monopoly, advertisers can easily turn to other options, both online and in print.

The plaintiff’s lawyers until June 14 to revise and resubmit their lawsuit with a more realistic definition of Google’s relevant market.

The other antitrust complaints against Google commit their own versions of the relevant market fallacy, as I’ve noted before:

Google’s relevant market is larger than a traditional search engine page. Every Uber ride involves an Internet search to pair riders and drivers. These searches do not use a Google algorithm, and would not work if their customers’ information was “being concentrated in one company.” Netflix, Hulu, and Spotify searches do not use Google. Nor do dating sites, which compete with each other based on proprietary search algorithms, as do many other popular search-based Internet services.

The relevant market fallacy also applies to allegations of anti-conservative bias against Google. If Google acquires even the reputation of serving unreliable search results, consumers can turn to competing options by simply typing a web address into their browser. And the relevant competitive market, as noted above, is not limited to search engines. News aggregators, consumer review sites, and other relevant content sites are legion, and easy to find, even for relatively uninformed users.

I call this the dozen keystrokes argument, because that’s roughly how difficult it is to type in another website’s address.

It will be months before court dates are set in any of the Google antitrust suits. They are still in the process of deciding relevant market definitions for the purposes of the cases. As we’ve seen, plaintiffs often try to bias antitrust cases in their favor by suggesting unrealistically narrow market definitions. It is good that at least one judge is wise to this semantic trick.

Hawley Antitrust Plan Would Limit Innovation and Harm Consumers

This news release was originally posted at cei.org.

Senator Josh Hawley (R-MO) today touted a new proposal he calls a “trust busting plan” that calls for a new standard for antitrust intervention to replace the legal principle of consumer harm. While the plan is not accompanied by specific legislation, Competitive Enterprise Institute experts weighed in on the Senator’s proposal.

Director of CEI’s Center for Technology and Innovation Jessica Melugin said:

“Senator Hawley claims that allowing the tech industry to operate in a relatively free market ‘hasn’t been a success for the American consumer,’ but if there’s consumer harm to point to, why does he advocate for abandoning the consumer harm standard in U.S. antitrust law? Perhaps it’s because consumers have enjoyed consistent innovations in products and services from ‘big tech,’ especially while quarantining during the pandemic and often at no monetary cost to them.  Similarly, his claims that the industry, ‘hasn’t been a success…for the American economy,’ don’t ring true for so many Americans that are employed by or invested in these economic powerhouses, not to mention the millions of consumers who enjoy tech products.”

Senior Fellow Ryan Young said:

“One of the most problematic parts of Sen. Hawley’s antitrust plan is its proposed ban of mergers and acquisitions for companies larger than $100 billion in annual revenues. Startups need capital to compete in the big leagues. But financial regulations, especially in the post-Dodd-Frank era, make it difficult for smaller companies to hold IPOs or attract other forms of investment. So, they instead get their capital by being bought out by one of the big tech companies.

“The regulatory situation is so bad that many promising startups are founded with the explicit goal of selling out to a bigger company. If Sen. Hawley wants fewer acquisitions by big companies, he should focus on the root cause of bad financial regulations, rather than the feel-good populism of banning mergers and acquisitions.

“Then again, the feel-good populism is likely the point. Hawley’s proposal is unlikely to become law. For him, antitrust policy is just another culture war issue. He wants to fire up his base and provoke his opponents. In this sense, his antitrust proposal is no different than his similarly unserious proposals to ban infinite scrolling in social media apps and to have the federal government regulate political speech.”

For more information on CEI’s position on antitrust, please visit cei.org/antitrust.

Monopoly Is Not the Same as Big

Ball State University economist Steve Horwitz posted to YouTube an excellent clarification/gentle rant about the difference between having a monopoly and being big. Though aimed at one his undergraduate classes in which many students were making repeated slips, it is a good reminder for just about everyone. This is what good teaching looks like.

The seven-minute video is here. It is even shorter than that if, as I often do, you play the video at 1.5x speed or so.

Book Review: Marc Levinson – The Great A&P and the Struggle for Small Business in America

Marc Levinson – The Great A&P and the Struggle for Small Business in America

This is an excellent history that is playing out again in today’s antitrust revival. A&P was the first nationwide grocery store chain. Though it barely exists today, in its prime it was the nation’s largest retailer. A&P inspired fear among its competitors and outrage among populists.

People made many of the same arguments against A&P in the popular press and in antitrust cases that people make today against Walmart, Amazon, and other big companies. The word choices, hyperbole, and breathless tone are almost identical. And yet, A&P was no match for consumer preferences, which eventually shifted elsewhere. The company chose not to adapt, and today exists on roughly the same scale as Blockbuster Video, which is down to a single store in Oregon.

Some of the very same charges, such as A&P’s selling self-branded products at lower prices than outside brands, are being revived today against Amazon. A&P-era arguments are even being repurposed to argue against Apple and Google’s app stores and search results. Not only were their business practices never anti-competitive, they clearly weren’t enough to save A&P from the competitive process. Nor will it be enough to save today’s big tech companies. Consumers are harsh sovereigns, and as soon as someone does it better, they’ll move on.

History does not repeat itself, but it often rhymes. Levinson digs up some of the lost stanzas of a poem being rebooted all over Washington today. There are lots of lessons here for people on both sides of the antitrust revival.

Third Antitrust Suit against Google since October Based on Flawed Argument

This press release was originally posted on cei.org.

A coalition of more than 30 states and territories today filed an antitrust lawsuit against Google, alleging the search engine has abused its power in markets ranging from voice assistants to digital advertising in an attempt to maintain a monopoly over internet searches. The antitrust lawsuit is the third filed against Google since October.

CEI Senior Fellow Ryan Young said:

“Today’s antitrust lawsuit, the third against Google since October, has a major flaw: the dozen keystrokes argument. It is not difficult to type bing.com or duckduckgo.com into your browser. Google pays Apple as much as $12 billion per year to Apple to have Google be its default search engine. This is apparently not enough to prevent Apple from reportedly building up its own search engine.

“Nor was Microsoft’s similar default status for its Internet Explorer browser enough to stave off competition from Firefox, Google Chrome, Apple Safari, and other browsers. Just as Microsoft never actually controlled the browser market, Google does not control the search market. Consumers do.”

Read more:

New EU Tech Rules will Chill Innovation and Harm Consumers

This is a press statement originally posted at cei.org.

The European Union today announced new rules it claims will change the way technology companies operate. The EU says the Digital Services Act and the Digital Markets Act “will create a safer digital space for users” and “level the playing field so that digital businesses can grow.”

Vice President for Strategy Iain Murray said:

“The European Union’s proposed new powers allow it to treat American tech firms as cash cows, to be fined whenever it finds them guilty of providing too much discretion to consumers or allowing too much speech. Its proposed veto on acquisitions will also chill innovation in the European tech sector as it will make the prospect of significant rewards for an acquisition-based business strategy less likely. Europe will act as an anchor on tech innovation, slowing progress and reducing consumer welfare worldwide. The incoming Biden administration should avoid making the same mistakes.”

Senior Fellow Ryan Young said:

“The European Union’s two proposed tech regulation bills have two fatal flaws. One is that, on purpose or not, they are trade protectionism under another name. Many of their provisions are aimed at the large U.S. tech companies. Taking them down a notch would give an opening to EU-based tech companies, the thinking goes. As with President Trump’s trade wars, this will harm consumers without actually helping the industry. The two bills also leave in place the EU’s stifling regulatory culture that is the root cause of Europe’s lack of tech sector innovation.

“The second fatal flaw is that the EU’s proposals would actually lock in the existing American firms’ dominance. They are the only companies that can afford the massive content moderation costs the EU is demanding, or the large fines. Startups that might one day dethrone today’s giants cannot afford these costs, and may not even bother trying to compete.”

Read more:

A Big-Picture View of the Antitrust Debate

In this month’s issue of Reason magazine, I have a feature-length article on the bipartisan push to revive antitrust enforcement. If you don’t have the print edition, it is now online. Here is the introduction:

Mark Zuckerberg was having one of 2020’s worst Zoom meetings. It was July 29, and one of the most influential men in the world was sitting, pale and perspiring, in a sparse white room getting attacked by members of Congress from both parties. Rep. Matt Gaetz, a Florida Republican and close ally of President Donald Trump, was scolding the Facebook CEO about the “content moderators that you employ [who] are out there disadvantaging conservative content.”

But before Zuckerberg could offer much in the way of a response, he was attacked from the left, as Rhode Island Democrat Rep. David Cicilline castigated Zuckerberg for not taking down the same content. For Cicilline, “the problem is Facebook is profiting off and amplifying disinformation that harms others because it’s profitable.”

For good measure, Rep. Jim Sensenbrenner, a Wisconsin Republican, asked Zuckerberg why Facebook temporarily took down Donald Trump Jr.’s account over a post promoting hydroxychloroquine as a COVID-19 treatment. Zuckerberg pointed out that the incident happened on Twitter.

After discussing how conservatives’ and progressives’ ideological priors are warping the antitrust debate, I point to a better way: abolish antitrust regulation outright. Or at the very least, require proof of consumer harm before unleashing it.

Read the whole thing here. See also CEI’s dedicated antitrust site, antitrust.cei.org.