September 29, 2020 10:38AM

NATO’s Rogue Member Meddles in Another Conflict

This week, armed clashes erupted between the forces of Armenia and Azerbaijan, exacerbating already serious tensions in the Caucasus. The underlying reason for the latest incident is the long‐​standing struggle between the two countries for control of Ngorno‐​Karabakh. That ethnically Armenian region is legally part of Azerbaijan, but Armenia assumes responsibility for guaranteeing the minority enclave’s self‐​declared political independence. The inherently unstable arrangement has led to several previous outbreaks of violence over the past three decades, but the latest incident seems especially serious. Both countries have declared martial law and commenced full military mobilization. “We are a step away from a large‐​scale war,” warns Olesya Vartanyan of the International Crisis Group, an NGO focused on preventing and resolving deadly conflicts.

The default assumption of most Americans might be that this obscure dispute is a parochial spat that has little or no relevance to the interests of the United States, and normally that would be the case. However, fellow NATO member Turkey has injected itself into the conflict, declaring its “complete support” for Azerbaijan and placing total blame on Armenia for the fighting. Armenian Prime Minister Nikol Pashinyan criticized Turkey for its “aggressive behavior” and demanded that the Turks stay out of the matter. Since Turkey is a much larger power than either Armenia or Azerbaijan, its direct involvement would signal a major expansion of the conflict.

Ankara’s declaration especially creates the danger of a showdown between Turkey and Russia, since Moscow regards Armenia as a client, if not an outright protectorate. The mere possibility of an armed confrontation between a NATO member and Russia is cause for alarm, since Article 5 of the North Atlantic Treaty considers an attack on any alliance member as an attack on all. A military incident between Turkey and Russia would require the United States to sort‐​out which party was guilty of aggression, and if it appeared that Moscow had initiated the clash, Washington would have an obligation to come to Turkey’s defense.

A responsible member of NATO would be cautious about putting its security partners in such a position. But Turkey, under the government of President Recep Tayyip Erdogan, is an increasingly volatile and irresponsible international actor. Thanks to Ankara’s recent behavior in the eastern Mediterranean, NATO’s old nightmare of a war between Greece and Turkey has resurfaced as a prominent worry for the Alliance. Erdogan’s regime also is engaging in high‐​profile meddling in Libya’s civil war—again adopting a stance that puts Turkish strategy in direct opposition to the faction that Russia is backing.

Both Ankara and Moscow can be faulted for their conduct in Libya, but Turkey is clearly the more disruptive player in the Caucasus. The Kremlin is urging both Armenia and Azerbaijan to de‐​escalate the fighting, and Russian Foreign Minister Sergei Lavrov is attempting to play the role of mediator. Erdogan’s government, on the other hand, appears determined to back Azerbaijan, even if that stance leads to an expanded and more deadly war.

The United States and other leading NATO powers need to send Erdogan a blunt message that they have no intention of allowing themselves to become entangled in a conflict with Russia over a chronic dispute between a Turkish client state and its long‐​time regional adversary. This latest problem also would be a good occasion to emphasize to Ankara that its disruptive behavior toward Greece, its ill‐​advised meddling in Libya, and now its destabilizing role in the Caucasus is making it difficult—if not impossible—to regard Turkey as a NATO member in good standing.

As I’ve written elsewhere, the entire rationale for the United States to continue its obligation to defend more than two‐​dozen NATO allies, including strategically irrelevant mini‐​states, makes less and less sense in the twenty‐​first century. But being obligated to defend an “ally” that seems determined to create more, rather than less, international turmoil is the essence of folly. Turkey’s disruptive behavior in the Caucasus is just the latest reason for U.S. leaders to downgrade America’s relationship with that country.

September 28, 2020 4:05PM

Using Artificial Classes To Settle Mass Litigation: What Ginsburg Knew

Last week the Sixth Circuit rejected a federal judge’s novel certification of an unusual “negotiating class” aimed at promoting a global settlement between opiate manufacturers and cities and counties around the country that have sued them. The designated class would have included thousands of cities and counties around the country that have not filed suit, and the way in which it would have handled their legal interests was assailed from many directions as lacking in fairness. Last week’s ruling triumphantly vindicates the prescience of the late Justice Ruth Bader Ginsburg, the modern Court’s most influential proceduralist voice, whose opinion for the Court in Amchem Products v. Windsor (1997) laid out the path correctly followed by the Sixth Circuit majority. (Judge Eric Clay wrote the opinion, joined by Judge David McKeague; Judge Karen Moore dissented.)

From my point of view, the opiates litigation demanding recoupment of public funds spent on addiction should never have been filed at all, and its proper settlement value would be $0.00. (More herehere, here, etc.) It’s not clear the courts will rule that way, however, and many legal observers expect that, as with many mass tort actions, this one will eventuate in a settlement. Such a settlement would extract money from drug company investors (as well as bystanders such as pain patients who may pay more for medication in future) and redistribute it to mayors, county executives, and their lawyers (many of the latter on contingency fee). As advocates of the “negotiating class” idea saw it, a settlement would be speedier and more generous to the extent that it could be more comprehensive and final. Federal judge Dan Polster in Cleveland agreed (more), and the appeal in In Re: National Prescription Opiate Litigation followed.

Not dissimilar arguments fueled the would‐​be global asbestos settlement brought before the Court in Amchem. There, too, a judge had been talked into what has been called a managerial approach to mass litigation, one that would convert court process into a sort of administered compensation scheme, like black lung, disability, or others commonly situated in the executive branch, while according short shrift to the individual procedural interests of some parties not (yet) suing.

The practical arguments for global disposition back then, as Ginsburg was the first to acknowledge, were weighty. Asbestos litigation was and remains enormously expensive, slow, and chancy, the reverse in all respects of how a beneficent thinker would design a compensation program. But the answer was not to cut corners in the process rights enjoyed by even the humblest federal court litigant by herding all into class outcomes without full and proper individual notice and chance to decide. Federal court procedure holds out a promise of individual hearing and individual adjudication that must not be lost in the felt practical need to aggregate litigant groups and move them by the hundreds and thousands as if on a game board.

Ginsburg and the Court were right then, and the Sixth Circuit panel is right now.

September 28, 2020 12:47PM

Trump and Biden Tax Policies

Penn-Wharton’s budget team has produced new estimates of the effects of Joe Biden’s fiscal proposals. This post looks at PW’s tax distribution estimates.

PW presents estimated effective tax rates by income group, including individual income taxes, corporate income taxes, and payroll taxes. The estimates assume that corporate taxes land partly on capital income and partly on labor income. Effective tax rates are taxes paid by an income group divided by the group’s income.

Joe Biden’s campaign site says, “President Trump spent the remainder of his first year in office fighting for a $1.5 trillion tax giveaway primarily for large corporations and the wealthy … But middle‐​class Americans were largely left out.” This refers to the Tax Cuts and Jobs Act (TCJA) signed into law by Trump in 2017.

Related Event

Focus on Fiscal Leadership: Release of the 2020 Fiscal Report Card on America’s Governors

Going into 2020, the United States was in its 11th year of economic expansion and state governments were enjoying robust revenue and spending growth. Then COVID-19 hit and triggered a deep recession, which has created new budget challenges for the nation’s governors. That is the backdrop to this year’s 15th biennial Fiscal Report Card on America’s Governors, which grades them based on their tax and spending policies. Please join us for a conversation with New Hampshire Governor Chris Sununu about this year’s report and the importance of fiscal restraint in state government.

Join the Conversation  

In my view, the PW estimates show that this Biden claim is not correct. The table below shows the PW estimates of effective tax rates in 2021 with and without the TCJA. I added a column showing the percentage tax cut for each group. For the bottom group, for example, the law cut the tax rate from 2.5 percent to 1.1 percent, which is a 56 percent tax cut.

The TCJA provided the largest percentage tax cuts to the bottom two quintiles—that is, the bottom 40 percent of U.S. households. All other groups received roughly similar percentage cuts ranging from 8 to 13 percent. It is not true that middle‐ or lower‐​income groups were “largely left out” of the Trump tax cuts.


Biden talks a lot about fairness in tax policy. His campaign site says, “As President, Biden will require corporations and the wealthiest Americans to finally pay their fair share.” At the Democratic convention, Biden said “it’s long past time the wealthiest people and the biggest corporations in this country paid their fair share.” Leftist politicians never say what that fair share is exactly. Fairness seems to be whatever the wealthy are paying now plus a lot more.

The chart below shows the PW estimates of effective tax rates in 2021 under current law and the Biden plan. Fairness is an opinion, but it seems far left to not perceive today’s highly slanted tax distribution as insufficiently progressive. Under current law, households at the top pay an 81 percent higher tax rate than households in the middle quintile. They pay three times the rate as those in the second quintile and more than 20 times the rate as those in the bottom quintile.

Households at the bottom pay virtually nothing, while households at the top pay almost a third of their incomes in federal taxes, on average. The New York Times reports that Donald Trump pays little in income taxes. If true, Trump’s taxes are a rare exception to the heavy burden that most high earners face, as reflected in the overall statistics.

What is tax fairness? My starting point would be the simplest possible system with a single rate and equal treatment of individuals and industries. The simpler the system, the easier it would be for the IRS to collect taxes on high earners such as Trump. Unfortunately, both parties in Congress have conspired to impose a tax system of ghastly complexity, which has made it easier for some high earners to slip through the cracks.


Note: in its appendix, PW presents estimates of the Biden plan including proposed changes to tax credits, which makes the Biden plan’s distribution even more progressive than shown here.

September 28, 2020 10:05AM

Survey: Roughly Half of Private Schools See Decreased Enrollment

During the first few months of COVID-19 lockdowns, supporters of private schooling feared that the sector would take a huge enrollment, financial, and ultimately existential hit. During the summer, however, fears subsided. Cato’s tracker of COVID‐​related private school closings grew to almost 120 schools, but nowhere near mass extinction. And as public schools increasingly announced that they would only operate online to start the new year, anecdotal evidence suggested ramped‐​up private school demand. Meanwhile, national surveys offered mixed clues as to what might be happening to private schools.

To get a quick but more concrete sense of the start‐​of‐​the‐​year enrollment situation, Cato’s Center for Educational Freedom conducted a survey from September 9 to the 18th of 400 private schools with a highest grade of kindergarten or higher, randomly selected from the federal Private School Universe database. We asked only if a school’s enrollment had increased, decreased, or stayed the same from the previous school year, and by how many students it may have changed including and excluding pre‐​K students.

As shown below, we found that when pre‐​K students are included, 57 percent of schools lost enrollment between last school year and the current one, 24 percent saw an increase, and 19 percent remained unchanged.

Private school enrollment changes with pre-k

Excluding pre‐​K students—so just looking at students in the K-12 range—47 percent lost students, 33 percent increased, and 21 percent were unchanged.

Private school enrollment changes without pre-k

What do enrollment numbers look like? Including pre‐​K students, the average school lost slightly fewer than 14 students, and without pre‐​K the average loss approached 6 students. To put those losses in context, the average school that responded to our survey enrolled about 225 students, including pre‐​K, according to the most recent enrollment data we could find. So they experienced, on average, roughly a 6 percent enrollment loss. The range of changes was large, from a gain of 90 students to a loss of 137. Not including pre‐​K students, the range was a gain of 75 to a loss of 100.

Note that one school that replied to our survey closed due to COVID-19. It is included in calculations of shares of schools with increasing, decreasing, or unchanged enrollments but not specific numbers. That school has been added to the closure tracker.

There are important limits to this survey. We conducted it quickly to begin filling a vacuum of information on how private schools were faring as the school year began. We only received 62 usable surveys—a 16 percent response rate—yielding a margin of error of +/- 12 percentage. We did not receive changes specifically for non‐​pre‐​K students from 4 schools that had pre‐​K students, so results excluding pre‐​K students are for only 58 schools, yielding a margin of error of +/- 13 percent. Also, we gathered total previous enrollment data from several sources that might be older than the previous year—we asked schools only for net gain or loss figures—and a handful of losses would be larger than the possibly old total enrollment recorded. We kept numbers as reported. Finally, while the 400 schools contacted were randomly selected, it is not clear that responses were not skewed in some way. That said, one check—the share of schools that are Catholic—suggested responses were close to representative: 21 percent were Catholic, almost exactly matching the Catholic share of all private schools according to recent federal figures. Similarly, Baptist schools composed about 5 percent of our respondents and account for roughly 5 percent of all private schools.

Of course, better polling as the year goes on and deeper analysis are needed, but these findings help provide some broad, early insight into 2020–21 effects of COVID-19 on private schools.

September 28, 2020 9:55AM

Mexican Immigrant Population Drops to Lowest Level Since 2004

From 2017 to 2019, the immigrant share of the U.S. population failed to increase at all for the first time since the Great Recession and probably for the first time in decades during an economic expansion. According to recently released data from the Census Bureau’s American Community Survey, a major reason for the stagnation is that America’s Mexican immigrant population has started to decline rapidly, dropping nearly 642,000 from 2016 to 2019.

As seen in Figure 1, Mexican immigrants reached 11.7 million in 2010 before fluctuating between 11.5 and 11.7 from 2011 to 2016, but then, starting in 2017, the numbers rapidly declined to 10.9 million. The last time the population was this small was in 2004. Meanwhile, the Mexican share of the immigrant population declined from a peak of 31 percent in 2005 to 24 percent in 2019.

As Figure 2 shows, the share of immigrants from Mexico rose rapidly from 1970 to 2000 before declining slightly by 2010, and then sharply by 2019 with most of the decline from 2017 to 2019. The 5‐​percentage point decline from 2010 to 2015 is by far the largest in history for Mexicans. No decade before the 2010s saw a decline of more than 0.2 percentage points. The era of Mexican immigration is wrapping up in a hurry under President Trump.

Mexico is just the fourth country in American history since 1850 to be the most common origin country for U.S. residents. Figure 3 compares the Mexican share of the immigrant population to the other three countries. Ireland led all countries from 1850 to 1870, followed by Germany from 1880 to 1920, followed by Italy from 1930 to 1970. Mexico’s five decades dominance is comparable to both Italy and Germany, but its peak was lower than the peak of 42.8 percent for Ireland in 1850. Unfortunately, the Census data on origin countries only date to 1850, so it’s unclear how long Irish dominance lasted, but the Irish share may have peaked in 1850 after the potato famine.

Mexicans falling numbers are not primarily a consequence of increasing deportations. Indeed, ICE removals to Mexico have slightly fallen under the Trump administration. The better explanation is that fewer Mexicans want to come to the United States, and more are returning. The surprising element is that unlike the stagnation in the immigrant population during the Great Recession, this decline occurred during a period of almost unparalleled economic expansion in the United States, while Mexico’s economy shrank in 2019. Perhaps the president’s anti‐​Mexican rhetoric is influencing Mexicans’ personal behavior. 

September 25, 2020 4:18PM

The One Policy Proven to Help the U.S. Semiconductor Industry (and Hamstring China) Is the One Nobody’s Mentioning

A few weeks ago, I noted that Congress had just earmarked billions of taxpayer dollars for American semiconductor manufacturers to counter the alleged (and thus far empty) national security threat posed by heavily‐​subsidized Chinese competitors. The U.S. subsidies, as you can imagine, have been cheered by both conservative fans of industrial policy and the semiconductor industry, which just released a new report on how “government incentives” will (obviously) help (historically healthy and productive) U.S. semiconductor companies somehow “turn the tide” against China. A recent writeup in The Protocol sets out the industry’s response to critics like me:

[Semiconductor Industry Association CEO John] Neuffer sought to address some of the biggest criticisms of the plan. “There’s this concern that it’s a race to the bottom with subsidies,” he said, referring to the idea that China and the U.S. could keep upping subsidies ad infinitum. “Well, that race began 20 years ago, and we’re still standing at the starting line saying ‘Gee, I wonder if we should get into the race.’ ”

Well then. Now, leaving aside the questionable (at best) efficacy of both the current Chinese subsidies and past U.S. government support for our semiconductor industry, as well as the economic and legal (trade law) concerns raised by the “global subsidies race” more broadly, what’s perhaps most striking about the current push by the politicians, the economic nationalists, and the lobbyists to bolster U.S. chipmakers is the absence of any discussion of one actual policy that has been actually proven to supercharge the industry and is being hobbled by the federal government as we speak: high‐​skill immigration. Indeed, a new Georgetown University report underscores the importance of immigration for the semiconductor industry, finding that–

  • Approximately 40 percent of high‐​skilled semiconductor workers in the United States were born abroad. India is the most common place of origin among foreign‐​born workers, followed by China.

high-skill workers by country of origin
  • In 2011, 87 percent of semiconductor patents awarded to top U.S. universities had at least one foreign‐​born inventor. Between 2000 and 2010, the United States enjoyed a net influx of about 100,000 electrical engineering patent holders, while India and China saw large net outflows.

EE Inflow Outflow
  • International students comprise around two‐​thirds of graduate students in electrical engineering and computer science, the top educational fields feeding into the U.S. semiconductor industry among green card applicants. The number of American students enrolled in semiconductor‐ related graduate programs (around 90,000) has not increased since 1990. In that same period, the number of international students nearly tripled from 50,000 to 140,000.

International Students
  • More than 80 percent of international Ph.D. graduates from semiconductor‐​related fields at U.S. universities stay in the country after completing their degrees. Stay rates are highest among Indian and Chinese doctoral recipients.

Stay Rates

Clearly, high‐​skill immigration (both workers and students) has been a boon for the American semiconductor industry — a finding consistent with analyses showing how educated immigrants boost U.S. innovation and productivity more generally. (As well as past work by the SIA.) By contrast, new research reveals that past high‐​skill immigration restrictions push U.S. multinational corporations to move their R&D activities offshore, to increase these activities in other countries (including in China), and thus to diminish the United States’ own innovative capacity. This outcome is particularly relevant for U.S. chipmakers and their current concerns about China: human capital is arguably the Chinese semiconductor industry’s biggest hurdle, and the Georgetown paper finds that “China has so far struggled to attract talent from the United States” and to retain that talent once it’s on the mainland. As such, “[l]arge reductions in the flow of talent between China and the United States would likely be welcomed by Chinese policymakers.” Indeed.

Unfortunately, the Trump administration and its allies are not just ignoring high‐​skill immigration, but actively working to restrict it further — even as COVID-19 hasn’t boosted unemployment in the U.S. tech sector. Given these facts, as well as the alleged threat facing the U.S. semiconductor industry (and by extension national security), you’d think that the ones making such allegations — in Congress, the industry, and the policy community — would at least mention immigration instead of just cheering for subsidies.

Or maybe the China threat just isn’t as big as they claim?

September 25, 2020 4:15PM

America’s Foreign Share Is In The Bottom Third Among Richest Nations

America’s immigrant share of its population stagnated under President Trump for multiple years for the first time since the Great Recession—possibly for the first time during an economic boom since the 1960s. This fact should be assessed against a backdrop in which the United States already had one of the most restrictive migration policies in the world.

New statistics from the United Nations for 2019 confirm that the United States ranks in the bottom third among wealthy countries both in terms of foreign‐​born share of its population and per‐​capita increase in foreign‐​born population. President Trump’s closure of the borders in 2020 has only added restrictions to an already extremely closed system.

Many people call the United States the “most generous country in the world” on immigration, but while it does have the most immigrants in an absolute sense, America is also one of the wealthiest and most populous countries on earth, so a proper comparison would control for wealth and size.

According to the United Nations, 15 percent of the U.S. population was foreign‐​born. Unlike the Census Bureau estimates, the United Nations includes Americans born abroad to American parents who are not “immigrants” in the legal sense since they became U.S. citizens by birth, but for cross‐​country comparisons where citizenship laws vary, it’s necessary to include them.

Of the 217 countries and semi‐​autonomous territories that the United Nations provides population counts for (plus Taiwan), the United States ranks 61st in the world for both foreign‐​born share of its population and its per‐​capita increase in its foreign‐​born population from 2017 to 2019.

As seen in Table 1, among the 57 countries or territories with a per‐​capita GDP over $20,000, the United States ranked 40th—or the 30th percentile—for foreign‐​born share of the population. The average wealthy country had a foreign‐​born population share of 30 percent, double the U.S. share. The United States similarly ranked 39th for per‐​capita increase in its foreign‐​born population from 2017 to 2019. Its population grew 0.27 percent from foreigners compared to an average of 1.8 percent for all wealthy countries.

Other countries maintain foreign‐​born shares far higher than the United States. The United Arab Emirates had the highest foreign‐​born share in the world at 88 percent. Other Gulf states—Kuwait and Qatar—also had most of their populations born abroad. A number of Island nations and city states rank highly on this list, but Australia and Switzerland both had foreign‐​born shares of about 30 percent. Israel, New Zealand, Canada, and Austria had between 20 and 23 percent.

Its worth noting as well, as seen in Figure 1, that the United States is “cheating” even to rank as highly as it does because so many of its immigrants are illegally present in the country. In other words, if the U.S. had enforced its laws as written, it would have an even smaller foreign‐​born population. About a fifth of the U.S. foreign‐​born population was born abroad, according to the Pew Research Center estimate, so taking out those admissions would drop America even further back. Of course, some immigrants live illegally in other countries, but none to the same extent as the United States.

The idea that America is the world’s “most generous” country on immigration is among the world’s most revealing phrases. Yes, it’s misleading even in the sense that it is intended, but more importantly, America doesn’t need to be “generous” to those coming to enjoy the American dream. It just needs to not round them up, not deport them, and not keep them out. That’s not “generosity” or a handout. That’s basic respect for the liberty of other human beings.

As I’ve written before, it is also in the country’s self‐​interest not to prohibit foreigners from living and working in America. Allowing people to freely move and work where they want is not charity. It is an expansion of the free market and allows people to contribute to the economic prosperity of the country and expand the pie for everyone. Unfortunately, our country’s policies—even before Trump—don’t reflect these principles. It’s long past time that they should.