Category Archives: Uncategorized

Pierre Bayle on Media Bias

Bias is nothing new, and it likely has gotten neither better nor worse over time. People tend to think that media bias and misinformation are getting worse in the social media age, ironically because the human brain is prone to recency bias, availability bias, and pessimistic bias. But as Pierre Bayle wrote in the 1680s in his Nouvelles de la République des Lettres back in the 1680s:

“History is dished up very much like meat. Each nation and religion takes the same raw facts and dresses them in a sauce of its own taste, and each reader finds them true or false according to whether they agree or disagree with his prejudices.”

A little bit of awareness of this tendency can go a long way towards minimizing the problem of bias and misinformation. In today’s time, that means avoiding the cable news channels, purging low-quality sources from one’s social media feeds (or abstaining entirely), putting a little effort into statistical literacy, and employing a little bit of skepticism, especially towards too-good-to-be-true stories that appeal to one’s ideological priors. Keeping an eye on the longer arcs of history, and not just today’s ephemera, is also useful. I often find myself simultaneously a short-term pessimist and a long-run optimist; it turns out that both can be true.

Bayle’s emphasis on tolerance towards people with different beliefs, apparently still as radical in our time as it was in his, is also helpful, both here and in many other parts of life.

DC Antitrust Suit Against Amazon Could Actually Harm Consumers by Making Online Goods More Expensive

This press release was originally posted on

The District of Columbia’s Attorney General filed a lawsuit today against Amazon, alleging the tech company is engaged in anti-competitive behavior by controlling retail prices online.

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

“Amazon should be free to set the terms for third party sellers on its own platform, just as sellers are free to pass on using Amazon’s Marketplace if they don’t like the terms. Amazon is pushing to offer the lowest prices possible and while that may not always please third-party sellers, it’s the dead opposite of consumer harm.”

Senior Fellow Ryan Young said:

“The District of Columbia’s Amazon lawsuit has terrible timing, for two reasons. One, it comes on the heels of the worst pandemic in a century, during which online ordering of groceries and other goods helped keep people out of crowded stores. If successful, the District’s antitrust lawsuit would make online goods more expensive. This would hit lower-income consumers and vulnerable populations the hardest. It would also harm sellers, who are having a difficult time as it is during a tough recovery.

“Two, competition is already increasing. Other retailers such as Walmart now have their own third-party seller programs that compete with Amazon’s. This is on top of existing online options small sellers can use, such as Ebay, Etsy, and Shopify, as well as numerous niche markets, such as Reverb for musical equipment and Newegg for computer products.”

Read more:

This Week in Ridiculous Regulations

Congress passed a $1.9 trillion spending bill, some of which may actually be COVID-related. Agencies issued new rules ranging from eastern hellbenders to reentry licenses.

On to the data:

  • Agencies issued 83 final regulations last week, after 60 the previous week.
  • That’s the equivalent of a new regulation every two hours and one minute.
  • With 642 final regulations so far in 2021, agencies are on pace to issue 3,415 final regulations this year. 2020’s total was 3,327 final regulations.
  • Agencies issued 63 proposed regulations in the Federal Register last week, after 47 the previous week.
  • With 380 proposed regulations so far in 2021, agencies are on pace to issue 1,887 proposed regulations this year. 2020’s total was 2,021 proposed regulations.
  • Agencies published 432 notices last week, after 428 notices the previous week.
  • With 4,167 notices so far in 2021, agencies are on pace to issue 22,165 notices this year. 2020’s total was 22,480.
  • Last week, 1,069 new pages were added to the Federal Register in a three-day week, after 1,300 pages the previous week.
  • The average Federal Register issue this year contains 302 pages.
  • With 14,216 pages so far, the 2021 Federal Register is on pace for 75,638 pages in 2021. The 2020 total was 87,352 pages. The all-time record adjusted page count (subtracting skips, jumps, and blank pages) is 96,994, set in 2016.
  • Rules are called “economically significant” if they have costs of $100 million or more in a given year. There are two such rules so far in 2021, none from the last week. Agencies published five economically significant rules in 2020, and four in 2019.
  • The running cost tally for 2021’s economically significant rules ranges from net savings of $100.7 million to net costs of $362.5 million. The 2020 figure ranges from net savings of between $2.04 billion and $5.69 billion, mostly from estimated savings on federal spending. The exact numbers depend on discount rates and other assumptions.
  • Agencies have published 11 final rules meeting the broader definition of “significant” in 2020, with two in the last week. This is on pace for 59 significant rules in 2021. 2020’s total was 79 significant final rules.
  • In 2021, 111 new rules affect small businesses. Two are classified as significant. 2020’s totals were 668 rules affecting small businesses, 26 of them significant.

Highlights from last week’s new regulations:

For more data, see Ten Thousand Commandments and follow @10KC and @RegoftheDay on Twitter.

Priorities for Commerce Secretary Nominee Raimondo: Tariffs, TPA, Trade Agreements

President-Elect Biden will nominate Rhode Island Governor Gina Raimondo to be the next Commerce Secretary. She will soon be in a position to undo much of the damage the Trump administration’s trade policies have done to America’s economy and foreign policy interests. 

She should work with Congress and President-Elect Biden to repeal all of the Trump tariffs, and to repeal Section 232 of the 1962 Trade Expansion Act and Sections 201 and 301 of the 1974 Trade Act. No future president should be able to abuse executive power the way Trump has, especially as the country goes through a difficult recovery from the COVID-19 crisis.

The Biden administration will likely negotiate important trade agreements with the United Kingdom and European Union over the next few years. Raimondo will likely play a prominent role in the negotiations. Once in office, Raimondo will need to work with Congress to renew the president’s Trade Promotion Authority (TPA), which expires in July. This would greatly speed up negotiations and help remove harmful trade barriers during a difficult economic recovery that could use the boost.

It is also important that the upcoming trade agreements stick to trade to the extent possible. There has been a trend of trade agreements increasingly including trade-unrelated provisions such as regulatory, environment, and labor policies. These are open invitations to cronyism and rent-seeking. The added complexity means more things can go wrong during negotiations, and could scuttle worthwhile agreements.

President Trump’s United States-Mexico-Canada Agreement, which was laden with trade-unrelated provisions and political giveaways, should not set a precedent for upcoming agreements. Raimondo has a wonderful reform opportunity ahead of her that few commerce Secretaries have enjoyed. She should take full advantage of that opportunity.

Several trade reform idea policies are in Iain Murray’s and my paper “Traders of the Lost Ark,” and in CEI’s forthcoming Agenda for Congress.

Toward Simplifying Antitrust Regulation

Antitrust regulation is a complex mess. Multiple agencies have overlapping jurisdiction with no set rules for determining who takes which cases. One of the antitrust enforcement agencies, the Federal Trade Commission (FTC), even has its own court system, where it sets the rules and hire its own judges, and pays their salaries. Over at The Hill, Alex Reinauer and I take a look at two bills from Sen. Mike Lee (R-UT) that would simplify the mess a little bit. One bill is the SMARTER Act:

It would require the DOJ [Department of Justice] and the FTC to follow a uniform interpretation of the Clayton Act, which governs merger cases. Currently, the agencies can follow different interpretations as suits their political needs, confusing judges and defendants alike.

It would also require the FTC to use independent courts.

The other is the One Agency Act:

Lee’s second bill is the One Agency Act. It would remove the FTC from merger cases entirely, though not from other antitrust cases. The Justice Department is perfectly capable of handling merger cases, and one agency in charge would add simplicity, predictability, and expertise. It would also end pointless turf battles between the agencies.

Read the whole thing here. See also CEI’s dedicated antitrust site,, and Wayne Crews’s and my paper “The Case against Antitrust Law.”

More Interconnected than Most People Think

From page 70 of Johan Norberg’s 2020 book, Open: The Story of Human Progress:

What we now think of as Western civilization is a combination of a philosophical heritage from the Greeks, religions from the Middle East, creatively interpreted by Romans in what is now Turkey, and scientific ideas borrowed from the Arabs and the Chinese. We got our alphabets from the Phoenicians, and our numbers are called ‘Arabic numerals’ because we learned them from mathematicians in Baghdad, who got them from the Indians.

This Week in Ridiculous Regulations

The big news is that the Food and Drug Administration is poised to follow several other countries’ lead in approving one or more coronavirus vaccines. [Ed.: this post was drafted last Friday morning. Vaccine approval happened over the weekend. The larger point stands.] If a mutual recognition policy was in place with any of those countries, approval would already have happened, and people would already be receiving treatment. The Fall 2020 Unified Agenda came out last week. It lists each rulemaking agency’s planned rules for the short and long-term. It was also due in October. CEI’s Wayne Crews breaks it down over at Forbes.Regulatory agencies issued new regulations ranging from standards for professional journalism to accountant qualifications.

On to the data:

  • Last week, 67 new final regulations were published in the Federal Register, after 65 the previous week.
  • That’s the equivalent of a new regulation every two hours and 30 minutes.
  • Federal agencies have issued 3,157 final regulations in 2020. At that pace, there will be 3,316 new final regulations. Last year’s total was 2,964 regulations.
  • There were 34 proposed regulations in the Federal Register last week, for a total of 2,055 on the year. At that pace, there will be 2,168 new proposed regulations in 2020. Last year’s total was 2,158 proposed regulations.
  • Last week, agencies published 456 notices, for a total of 21,260 in 2020. At that pace, there will be 22,327 new notices this year. Last year’s total was 21,804.
  • Last week, 1,881 new pages were added to the Federal Register, after 2,278 pages the previous week.
  • The 2020 Federal Register totals 80,580 pages. It is on pace for 84,626 pages. The 2019 total was 70,938 pages. The all-time record adjusted page count (which subtracts skips, jumps, and blank pages) is 96,994, set in 2016.
  • Rules are called “economically significant” if they have costs of $100 million or more in a given year. Five such rules have been published this year. Four such rules were published in 2019.
  • The running cost tally for 2020’s economically significant regulations ranges from net savings of between $2.04 billion and $5.69 billion. 2019’s total ranges from net savings of $350 million to $650 million, mostly from estimated savings on federal spending. The exact number depends on discount rates and other assumptions.
  • Agencies have published 74 final rules meeting the broader definition of “significant” so far this year. 2019’s total was 66 significant final rules.
  • So far in 2020, 622 new rules affect small businesses; 24 of them are classified as significant. 2019’s totals were 501 rules affecting small businesses, with 22 of them significant.

Highlights from last week’s new regulations:

For more data, see Ten Thousand Commandments and follow @10KC and @RegoftheDay on Twitter.

The Relevant Market Fallacy and Facebook’s Antitrust Cases

Facebook was hit by two separate antitrust complaints this week. One is from the Federal Trade Commission (FTC) and the other is from a group of attorneys general (AGs) from 46 states. As most antitrust complaints do, they both commit the relevant market fallacy. This is defining a company’s market unrealistically narrowly, so it looks more dominant than it is. The Justice Department also made this error in its recent Google complaint.

The FTC complaint slips up in its second paragraph, stating that “Facebook holds monopoly power in the market for ‘personal social networking services.’” The states’ complaint uses the same term, which they define starting on page 11.

If you’ve never heard the term “personal social networking services” before, you’re not alone. The FTC and attorneys general made it up just for their Facebook cases. And they both define it in such a way that excludes TikTok—a curious oversight. TikTok is a competing social network that is popular among the younger people Facebook is actively courting to freshen up its aging user base. Facebook is not only competing against other networks, it is also competing against demographics.

Facebook’s popularity with the over-40 crowd is a major reason for its unpopularity with teenagers. They generally prefer to be where their parents and teachers are not. Because younger people’s friend groups are malleable and still forming, and because typing a website’s name into a browser is so easy, the network effects argument the complaints rely on is not very strong. Network effects basically means that there is strength in numbers. People will tend to join networks where their friends already are, rather than try to cajole them into moving to a brand new network. But this theory does quite explain why many people are members of multiple networks. It is common for people on Facebook to also be active on Twitter, Discord, LinkedIn, and other networks.

While the network effects argument has some sway with older people, whose networks are long-established and for whom inertia is stronger, it is weaker with younger people. In the case of Facebook, its network effects among the gray-haired set may in fact be pushing younger people away.

This may be why a text search for “TikTok” turns up zero results in both the FTC’s and the states’ complaints—out of sight, out of mind. So even if Facebook holds a monopoly in “personal social networking services” as defined by the complaints, Facebook clearly does not hold a monopoly in its real-world relevant market. Because of that, consumer harm is almost impossible to prove—and proof of consumer harm is the threshold for antitrust enforcement.

But this concedes too much. Facebook almost certainly does not hold a monopoly even in the narrow “personal social networking services” market definition. One common test whether a company has a monopoly is to see if it is able to jack up prices, restrict supply, and still remain dominant. Extra-high monopoly profit margins naturally attract competitors; successfully keeping them out is a sign a company may have monopoly power.

Facebook does not charge its users, but it does charge advertisers. Has Facebook been able to raise ad prices and squeeze supply? It has not. Online advertising prices have gone down by about half over the last decade—precisely the period when Facebook became a major player. Over the same period, print advertising prices have been going up. Some newspapers have doubled their ad prices.

If Facebook has a monopoly and is acting to illegally maintain it, cutting its ad prices while competing ad sellers raise theirs is an unlikely way to go about it.

There are a lot of other arguments in play in the Facebook antitrust cases. But the relevant market fallacy is a major one, and it doesn’t look good for the FTC’s or the state AGs’ cases.

For more, see Iain Murray, Jessica Melugin, and Mario Loyola’s statement, Wayne Crews’s and my paper, “The Case against Antitrust Law,” and visit

Book Review: Judith Herrin – Ravenna: Capital of Empire, Crucible of Europe

Judith Herrin – Ravenna: Capital of Empire, Crucible of Europe (Princeton: Princeton University Press , 2020).

Less a history of Ravenna, than a history of Europe from about 390 to 813 AD. Herrin’s history ranges from Late Antiquity (Early Christianity in Herrin’s terminology) up to Charlemagne. Ravenna is more of a constant background character in a larger narrative than the star.

Ravenna has a fascinating place in history, and I would have loved to have learned more about the city itself. As the Roman Empire’s focus moved east, the city of Rome lost its luster. Ravenna became something of a second capital city on the Italian peninsula. Emperors would live their entire lives in or near Ravenna, perhaps visiting Rome once or twice in their reign to give a ceremonial appearance before the Senate, which still existed, but had no purpose other than to keep Rome’s remaining wealth squabbling with each other rather than with the Emperor.

But the Empire’s center of gravity continued to move east past Ravenna, to Constantinople. Ravenna never really got its due as the capital of a major empire. First, Diocletian split the Empire into separate Eastern and Western halves in the late third century AD. This would have been Ravenna’s best time to shine, but it was always overshadowed by Constantinople, the Eastern capital. Then the Western half collapsed in 476, and Ravenna slowly descended into obscurity—though as Herrin shows, for this entire period, and for centuries to come, it was still home to fascinating figures and power struggles.

Herrin does not go into great detail about Ravenna’s layout, architecture, daily life and culture, economy, intellectual life, geography, or much else about he city. But she does an excellent job on her narrower focus of monarchs and politics. The amount of times Ravenna changed hands between Romans, Byzantines, Goths, and eventually proto-national dynasties is astounding. Ravenna might rarely have been the center of attention, but it was nearly always part of the action. Most of Herrin’s narrative centers around powerful rulers.

Galla Placidia (d. 437 AD), the daughter of the Gothic emperor Theodosius I and regent to Valentinian III, emerges as a powerful figure at a time when women rulers were extremely rare. She spent part of her early life in the household of the Roman general Stilicho (d. 408), who became a de facto emperor. She was captured by the invader Alaric’s army, and married the Visigothic king Ataulf, becoming their queen. After he was murdered, she eventually married the Roman emperor Constantius II, with whom she had a son, Valentinian III, and served as his regent.

Theoderic the Great (d. 526) was an Ostrogothic king who filled the power vacuum left by the fall of Rome, and fought off the Byzantines, as Eastern Empire had come to be called. As an Arian Christian, he played an outsize role in early Church schisms, which the Arians lost.

Justinian (d. 565) was Byzantine emperor about a generation after Theoderic’s time. He came as close as anyone to reuniting the two halves through his general Valisarius, though he ultimately fell short. He also issued an influential law code in 525, and the Hagia Sophia was built during his reign—though far from Ravenna.

After Justinian’s death, the Lombards (“long beards”), thought to be of Scandinavian origin, took over Northern Italy, including Ravenna. They were in turn displaced by the Merovingian dynasty, which ruled over large pats of what is now France, and then the Carolingian dynasty, which takes its name from its founder Carolus Magnus, which translates from the Latin as “Chuck the Great.” He is today known as Charlemagne.

Charlemagne represents a lot of things. Two of the most important are the power struggle between church and state, and the power dynamics between East and West. Ravenna was home to the Byzantine papacy from 537 to 752, when it moved back to Rome under Stephen III. This represented a shift in the center of gravity from the East back to the West. In 800, the pope crowned Charlemagne on Christmas Day in St. Peter’s Basilica—in Rome, not Ravenna. This was another data point for the Western revival. It also marked a shift in power from church back to state.

A third Carolingian theme is European unification. After centuries of squabbling between Romans, Byzantines, barbarians, the Catholic Church, the Orthodox Church, and Muslims, Charlemagne centralized power over the whole region in himself. And again, Ravenna did not play a starring role. The main locations for this drama were in Rome and Aachen, Charlemagne’s rising capital to the North that had its own symbolic significance. But Ravenna was right there in the middle, taking it all in.

Herrin’s book might have done with either a different title, or with more attention paid to the city in its title. But it is still an excellent history about a period and a city that do not get enough attention from either historians or their readers.

On the Radio: the Google Antitrust Case

Last week I appeared on Jim Blasingame’s Small Business Radio to talk about the Google antitrust case and competition policy more generally. Audio of our conversation is here.