Feed aggregator

Over the past week and in response to the recent clash in Charlottesville, Virginia between neo-Nazi/white supremacist fanatics and those opposed to them, some elements on the American political left are now clamoring for the Trump administration to restart a long-ago discredited Department of Homeland Security (DHS) program known as “Countering Violent Extremism” (CVE). This action alert from CREDO Mobile is representative:

DHS established the Counter Violent Extremism (CVE) program to protect communities from domestic hate groups. It provides funding and resources to help communities develop and sustain local prevention efforts. Life After Hate, a group that focuses on breaking down white supremacy, was the only organization of its kind that DHS had approved for CVE federal funding before Trump took office. Just weeks before the terror attack in Charlotteville, anti-Muslim Trump aide Katharine Gorka, wife of known Nazi sympathizer and Trump adviser Sebastian Gorka, pushed the CVE to eliminate Life After Hate’s grant. Now, no CVE funding supports groups that exclusively confront white supremacy. Trump’s administration is planning to refocus the CVE program on extremism connected to Muslim groups only. That is blatantly discriminatory and completely unacceptable.

Tell DHS and key congressional leaders: Restore federal funding to stop Nazis, white supremacists and other right-wing extremists engaging in domestic terrorism.

Yesterday, a Huffington Post headline blared “It’s Not Just Trump, Federal Law Enforcement Is Not Very Focused On Far-Right Extremism.” As I noted last week, a 2014 survey conducted by Duke University in concert with the Police Executive Research Forum (PERF) found exactly the opposite. Indeed, as I noted in that same piece, multiple governmental entities over the course of the last decade have continued to highlight the threat posed by neo-Nazi/white supremacist groups.

Just last year, a bipartisan collection of groups sent a letter to then-President Obama denouncing CVE programs, especially those that target Arab/Muslim-Americans. I’ve also had extremely harsh things to say about federal CVE programs vis a vis their targeting of the Arab/Muslim-American communities. Any federal program that targets people or groups for increased surveillance or other repressive measures because of their political or religious beliefs is, by definition, unconstitutional. It’s also operationally counterproductive.

In his 2007 book Thinking Like A Terrorist, FBI veteran Mike German offers this warning (p. 153):

…when law enforcement officers arrest the white supremacist terrorist his beliefs are featured prominently in the media and in his subsequent prosecution. The prosecutor takes advantage of the community’s disdain for his beliefs to paint him as an evil person. After his conviction the prosecutor asks the judge to sentence the white supremacist more harsly because of his beliefs. The legislature will step in as well, passing hate crime laws to enhance penalties for such race-motivated crims. What these prosecutors and legislators forget is that while this political cause is unpopular, it does have a following. Their reaction is going to be exactly the same as if their cause was popular. When the prosecutors and legislators emphasize the politics rather than the crime, that small following will begin to feel they are being oppressed for their political beliefs, and rightfully so. They would now have a more legitimate reason to resist the government, and a more effective recruiting campaign.

Reviving the discredited CVE framework would only confirm the prevailing white supremacist narrative—in exactly the way it confirmed the ISIS narrative when CVE targeted Arab/Muslim-Americans. 

The authoritarian streak in the American political psyche—which I attribute to Federalists like Hamilton, and to their modern successors who populate and cheer on our gargantuan National Security State—has repeatedly led to efforts to silence speech and ideas deemed by the majority to be at best objectionable, and at worst a threat to the prevailing political and economic order. From the Alien and Sedition Acts to the Anarchist Exclusion Act to the McCarran Act and beyond, attempts to suppress unpopular speech or ideas have been fought in the streets, in legislatures, and in the courts. And it is that last institution, in the form of the United States Supreme Court, that has made it clear that no matter how much you or I may detest a David Duke or a Richard Spencer or any of their adherents, their right to speak their mind—no matter how hateful or repellant their words or ideas may be—is foundational to our republican form of government.

As Justice Douglass noted in his concurrence to the Court’s decision in Brandenburg v. Ohio:

One’s beliefs have long been thought to be sanctuaries which government could not invade. Barenblatt is one example of the ease with which that sanctuary can be violated. The lines drawn by the Court between the criminal act of being an “active” Communist and the innocent act of being a nominal or inactive Communist mark the difference only between deep and abiding belief and casual or uncertain belief. But I think, that all matters of belief are beyond the reach of subpoenas or the probings of investigators. That is why the invasions of privacy made by investigating committees were notoriously unconstitutional. That is the deep-seated fault in the infamous loyalty-security hearings which, since 1947 when President Truman launched them, have processed 20,000,000 men and women. Those hearings were primarily concerned with one’s thoughts, ideas, beliefs, and convictions. They were the most blatant violations of the First Amendment we have ever known.

James Fields crossed the line from a hateful belief to a criminal act—murder—and that is why he will face a jury of his peers for his monstrous conduct, and deservedly so.

So long as America holds to the standard set by the Bill of Rights, the chance to maintain a society that values open debate over even the most repellant of ideas remains. Indeed, forcing the David Dukes and the Richard Spencers of the world to defend their bigoted, fact-free philosophy is the best way to ensure they are discredited and have no chance of ever gaining any power to implement them. This should be the main lesson of Charlottesville, and the main reason why federal CVE programs in any form should be a thing of the past.

The Frederick News-Post has run my opinion piece on Maryland Gov. Larry Hogan’s decision to support relocation of the statue of Chief Justice Roger Taney from its previous conspicuous spot in front of the State House in Annapolis. Excerpt:

…Taney did many things in an illustrious legal career but is remembered for only one: the disastrous Dred Scott decision, which served to entrench slavery….

Change in the display of public memorials is natural and inevitable. Cemeteries may aspire to present an unaltered face over centuries. Both cemeteries and battlefield sites appropriately look backward rather than forward. But much-used public places are meant for the living and are ordinarily under tougher scarcity constraints.….

No one has erased him from the history books—the Dred Scott case itself makes sure of that….

Some conservatives now cry that we are on a slippery slope. What next—take Jefferson off the nickel, rename the city and state of Washington?

But each and every modern society chooses to live partway down the slippery slope of “not remembering history.” (Unless you believe that Warsaw, Prague and Budapest should have left their statues of Lenin in place.) Like Europe, America can take advantage of big, comfortable stopping points short of tearing down and renaming everyone and everything. For example, few if any memorials to Jefferson honor him for being a slaveholder. On the other hand, most statues of Confederate leaders do honor their service to that specific cause. Writes libertarian law professor Ilya Somin: “government should not honor people whose principal claim to fame was fighting a war in defense of the evil institution of slavery.”

On one big point, conservatives are very right: Mob actions and vandalism seldom end well….

Whole thing here, plus some thoughts from Andrew Stuttaford. From Atlas Obscura, displaced statues as a subject of historic preservation. Related, from @david_tanenhaus on Twitter: “My favorite Civil War era monuments are the 13th, 14th, and 15th Amendments.” [cross-posted and adapted from Overlawyered]

I see that my colleagues are referring to the new online Encyclopedia of Libertarianism as “a Wikipedia for libertarianism.” I suppose that’s sort of true, in that it’s an online encyclopedia. But it’s not exactly Hayekian, as Jimmy Wales describes Wikipedia. That is, it didn’t emerge spontaneously from the actions of hundreds of thousands of contributors. Instead, editors Ronald Hamowy, Jason Kuznicki, and Aaron Steelman drew up a list of topics and sought the best scholars to write on each one – people like Alan Charles Kors, Bryan Caplan, Deirdre McCloskey, George H. Smith, Israel Kirzner, James Buchanan, Joan Kennedy Taylor, Jeremy Shearmur, Leda Cosmides and John Tooby, Norman Barry, Richard Epstein, Randy Barnett, and Vernon L. Smith, along with many Cato Institute experts. In that regard it’s more like the Encyclopedia Britannica of libertarianism, a guide to important topics by top scholars in the relevant field.

The Britannica over the years has published articles by Albert Einstein, Sigmund Freud, Marie Curie, Leon Trotsky, Harry Houdini, George Bernard Shaw, Bertrand Russell, Milton Friedman, Simon Baron Cohen, and Desmond Tutu. They may have slipped a bit when they published articles by Jimmy Carter, Bill Clinton, and Lee Iacocca. And particularly when they chose to me to write their entry on libertarianism.

Was it a typo or a Freudian slip? The Washington Post reports:

As president-elect, for instance, Trump took Boeing to task for cost overruns when he tweeted that the Air Force One program’s $4 billion expenditures were “out of control” and suggested the contract be canceled….

Trump was more complementary on Feb. 17, when he made appearance at a Boeing factory in South Carolina and concluded his remarks by saying, “May God bless you, may God bless the United States of America, and may God bless Boeing.”

The reporters meant “complimentary.” But indeed the point of the article is just how “complementary” big government and its big contractors are. Headlined “Why America’s biggest government contractors balked at criticizing Trump,” the article explores how CEOs started jumping off President Trump’s advisory councils after his disappointing remarks about white supremacists marching in Charlottesville – but not “the four government contractors on the president’s advisory councils — Lockheed Martin, Boeing, Harris Corp. and United Technologies.” After all, 

In many ways, contractors such as Boeing and Lockheed Martin are more dependent on government decision-making than other companies that took part in the councils.

Indeed, if a large part of your business comes from government contracts, you’d better be very careful about criticizing the president of the United States. Especially a president who has little sense of the proper limits of presidential authority:

Those negotiations [over a new fighter plane] were marked by unusually close interactions between Trump and the business executives involved. Bloomberg later reported that Trump allowed Boeing chief executive Dennis A. Muilenburg to listen in on a call with a key government manager for the F-35 program as Trump sought information on the two planes.

President Trump’s tweets, legal problems, chaotic White House management, and other high-profile troubles may have diverted attention from a problem that many of us pointed out before he was elected: his “economic nationalism” that seems to mean in practice protectionism, crony capitalism, and a promise that he’ll personally run the U.S. economy.

Government contractors understand this. Even before he was elected, Trump intervened to “persuade” Carrier to keep a plant open in Indiana. How? Was it the state tax credits? Or something less public? The CEO of Carrier’s parent company United Technologies, Greg Hayes – who was later on the president’s manufacturing council – acknowledged that the deal to keep the plant open probably wasn’t really economic. But:

I was born at night, but it wasn’t last night. I also know that about 10 percent of our revenue comes from the U.S. government.

When companies get in bed with government, that’s the bargain they make. And as we’ve just seen, that bargain not only leads to economic decisions that make us all poorer, it stifles the free speech of those dependent on government decisions. And that’s a problem when government is the biggest landlord, employer, arts patron, and purchaser of goods and services in society.

When those in power make blatantly false claims that could lead to less freedom for the least advantaged members of society, it is imperative that they are corrected. The President of the American Federation for Teachers, Randi Weingarten, gave a speech last month where she stated that school “vouchers increase racial and economic segregation.”

What Does the Evidence Say?

As I pointed out last month in a Cato Policy Forum on School Choice and Democracy, out of the eight rigorous empirical studies existing on the subject, seven of them show that school voucher programs increase racial integration within the United States. As shown in table 1 below, none of these studies indicate that vouchers lead to racial segregation. Why is this the case?

When school choice programs give disadvantaged children the opportunity to exit their already-segregated neighborhood schools, their transitions unsurprisingly result in a more racially and socioeconomically integrated society.

Table 1: Impacts of Voucher Programs on Racial Integration

Note: A box highlighted in green indicates that the study found a statistically significant improvement in racial integration. A box highlighted in yellow indicates that the study did not find any significant differences in racial integration across sectors.

For some reason, school choice skeptics ignore all of the compelling U.S. evidence, and instead point to international studies from countries like Sweden and Chile. However, these studies are methodologically unable to show that vouchers actually led to segregation.

For example, the Sweden study correlates the changes in the degree of school choice and racial segregation across two time periods, 1993 and 2009. Notably, the authors did not control for overall changes in the immigrant population in Sweden over the same years, so they are unable to conclude that the voucher program caused the change. According to the Swedish Migration Agency, yearly residence permits granted to immigrants ranged from around 59,000 in 1993 to around 100,000 in 2009. Obviously, overall immigration leads to more racial stratification, regardless of the school choice program.

More Important Questions?

The scientific evidence overwhelmingly shows that school vouchers lead to racial integration. But what if it didn’t?

Should such a finding allow government officials to control the educational selections of families desperately trying to improve their well-being? If a minority parent wanted to send their child to a prestigious all-black school, should a bureaucrat be able to tell them that they are not allowed to do so?

This is strange to me: in our country’s disturbing history, government officials decided that people ought to be “separate but equal” for the greater good. They thought that having racially diverse people in the same institutions would harm society through racial tension. Today, public officials make the same argument, just in the opposite direction. They claim that government force is necessary to move people, based on skin color, to achieve some greater social goal.

If you do not support the forced racial segregation that started back in 1896, it would be logically inconsistent for you to support forced racial integration today.

We can avoid all of the negative consequences that come along with the use of government force by allowing individual families to voluntarily opt into the schools that work best for their children. As the evidence shows, the self-interested choices of individual parents result in the social benefit of racial integration, without coercion by those in power.

Yesterday, two men drove a van into a crowd in Barcelona, Spain and killed more than a dozen people and injured many more in what Spanish authorities are claiming is a terror attack.  Spanish authorities may have also prevented another attack later in that day and believe these are linked to a recent explosion.  Not all of the facts are public yet and we will learn more in the coming days, but Spain’s experience with terrorism can at least put what happened into perspective.

According to data from the Global Terrorism Database at the University of Maryland and the RAND Corporation, terrorists murdered 1,209 people in Spain from 1975 through the end of 2016 (Figure 1).  The spike in deaths in 2004 was the result of a major al Qaeda attack on the Madrid subway system that murdered 192 people.  Only the United Kingdom has suffered more from terrorism during that time with 2,359 total murders.  There were also 4,738 injuries in terror attack in Spain during this 42-year time span.

Figure 1

Murders in Spanish Terrorist Attacks by Year, 1975-2016

Sources: Global Terrorist Database and RAND Corporation.

From 1975 through 2016, a Spaniard’s chance of dying in a terrorist attack was about 1 in 1.43 million per year (Table 1).  Spain was the fourth most victimized country on the list by that measure and about 2.6 times as deadly as the United States.  Basque separatists are responsible for most of the terrorist deaths in Spain since 1975 but Islamists have been the most deadly since 2004.  Basques are not immigrants and their language is so different from other European tongues than many linguists have theorized that it predates the introduction of Indo-European tongues on the continent of Europe.  Thus, most of these deaths were not caused by recent immigrants.

Terrorism in Spain has trended generally downward since 2000.  Since then until 2016, 247 people have been killed in terrorist attacks on Spanish soil which translates to a 1 in 3.1 million chance per year of dying in that way.  In 2012, there were 364 homicides in Spain which translate into a 1 in 128,729 chance of being murdered in a non-terrorist homicide that year.  Spain’s homicide rate is so low that terrorism actually looks like a serious hazard by comparison.                                                                                

Table 1

Annual Chance of Dying in a Terror Attack and the Number of Terrorist Murders in European Countries, 1975-2016


Terrorist Murders

Annual Chance of Dying in Terror Attack



1 in 765,052



1 in 975,361

United Kingdom


1 in 1,057,248



1 in 1,432,847



1 in 2,069,185



1 in 2,396,564



1 in 3,977,889



1 in 5,030,009



1 in 5,932,942



1 in 7,643,464



1 in 12,810,844



1 in 14,605,254



1 in 16,424,666



1 in 24,047,998



1 in 24,208,884



1 in 26,776,608



1 in 28,499,047



1 in 29,982,906



1 in 31,607,066



1 in 50,301,214



1 in 55,645,834



1 in 72,048,567



1 in 72,509,981



1 in 82,598,090



1 in 143,165,024



1 in 225,306,754



1 in 231,493,613




Sources: Global Terrorist Database, RAND Corporation, United Nations, and Author’s Calculations.

Some pundits argue that that chance of dying in terrorism is too high and that people should be willing to give up some freedoms in exchange for more security.  They only sometimes mention what freedoms we should surrender and how such a move could lead to more security.  They also rarely consider the costs of giving the government extra power, especially costs to our safety.    

The misconception that safety is binary—it is either perfect safety or perfect danger—is on full display in the debate over security and freedom.  The real question is what specific freedoms or other costs are Spaniards, or others, are willing to endure by increasing the size and power of their governments and reducing their own civil liberties in order to diminish the already low annual chance of dying in a terrorist attack to 1 in 4 million a year, 5 million a year, or to some other number.  Even assuming that there is no way more government power can backfire and decrease security, there is a point where the costs of extra security are not worth the benefits.  In the United States, we are well past that point.  Spain shouldn’t similarly overreact.                

Why should a city, state, or federal government put statues in public parks?  Doing so addresses no plausible market failure, while using taxpayers funds and, as demonstrated tragically over the past few weeks, generates controversy, polarization, and violence. Thus governments should take down all statues, regardless of their political implications.

This is not “erasing” history but instead leaving it where it belongs, in the hands of private actors and mechanisms.  Historians, textbook authors, universities, learned societies, the History Channel, and many other individuals and organizations can all present their own views of history and battle for the hearts and minds of the public.  Government statues are government putting its thumb on the scale, which is one step down the slippery slope of thought control.


The Encyclopedia of Libertarianism, published in 2008 in hard copy, is now available free online at Libertarianism.org. The Encyclopedia includes more than 300 succinct, original articles on libertarian ideas, institutions, and thinkers. Contributors include James Buchanan, Richard Epstein, Tyler Cowen, Randy Barnett, Ellen Frankel Paul, Deirdre McCloskey, and more than 100 other scholars.

A couple of years ago, in an interesting discussion of social change and especially the best ways to spread classical liberal ideas at Liberty Fund’s Online Library of Liberty, historian David M. Hart had high praise for the Encyclopedia

The Encyclopedia of Libertarianism provides an excellent survey of the key movements, individuals, and events in the evolution of the classical liberal movement….

One should begin with Steve Davies’ “General Introduction,” pp. xxv-xxxvii, which is an excellent survey of the ideas, movements, and key events in the development of liberty, then read some of the articles on specific historical periods, movements, schools of thought, and individuals.

He goes on to suggest specific articles in the Encyclopedia that are “essential reading” for understanding “successful radical change in ideas and political and economic structures, in both a pro-liberty and anti-liberty direction.” Here’s his guide to learning about the history of liberty in the Encyclopedia of Libertarianism:

  1. The Ancient World
    1. “Liberty in the Ancient World”
    2. “Epicureanism”
    3. “Stoicism”
  2. Medieval Period
    1. “Scholastics - School of Salamanca”
  3. Reformation & Renaissance
    1. “Classical Republicanism”
    2. “Dutch Republic”
  4. The 17th Century
    1. “English Civil Wars”
      1. “The Levellers”
      2. “John Milton” & “Puritanism”
    2. “Glorious Revolution”
      1. “John Locke” & “Algernon Sidney”
      2. “Whiggism”
  5. The 18th Century
    1. 18thC Commonwealthmen - “Cato’s Letters”
    2. The Scottish Enlightenment
      1. “Enlightenment”
      2. “Adam Smith”, “Adam Ferguson” & “David Hume”
    3. The French Enlightenment
      1. “Physiocracy” - “Turgot”
      2. “Montesquieu” & “Voltaire”
    4. “American Revolution”
      1. “Declaration of Independence” - “Thomas Jefferson” & “Thomas Paine”
      2. “Constitution, U.S.” - “James Madison”
      3. “Bill of Rights, U.S.”
    5. “French Revolution”
      1. “Declaration of the Rights of Man and of the Citizen”
  6. The 19th Century
    1. “Classical Liberalism” - the English School
      1. “Philosophic Radicals”
      2. “Utilitarianism” - “Jeremy Bentham”
      3. “Classical Economics” - “John Stuart Mill”
    2. “Classical Liberalism” - the French School
      1. “Jean-Baptiste Say” & “Benjamin Constant”
      2. “Charles Comte” & “Charles Dunoyer”
      3. “Frédéric Bastiat” & “Gustave de Molinari”
    3. Free Trade Movement
      1. “Anti-Corn Law League” - “John Bright” & “Richard Cobden”
    4. “Feminism and Women’s Rights”
      1. “Mary Wollstonecraft”
      2. “Condorcet”
    5. Abolition of Slavery - “Abolitionism”
      1. “William Wilberforce”
      2. “William Lloyd Garrison” & “John Brown”
      3. “Frederick Douglass” & “Lysander Spooner”
    6. [The Radical Individualists]
      1. “Thomas Hodgskin”, “Herbert Spencer”, & “Auberon Herbert”
    7. The “Austrian School of Economics” I
      1. 1st generation - “Carl Menger”, “Eugen von Böhm-Bawerk”
      2. interwar years - “Ludwig von Mises”, “Friedrich Hayek”
  7. Post-World War 2 Renaissance
    1. “Mont Pelerin Society” - “Friedrich Hayek”, “Milton Friedman”, “Karl Popper”, “James Buchanan”
    2. Institute for Economic Affairs (IEA) & “Antony Fisher”
    3. Foundation for Economic Education (FEE) & “Leonard Read”
    4. Institute for Humane Studies & “F.A. Harper”
    5. The Austrian School of Economics II
      1. post-WW2 2nd generation - “Ludwig von Mises”, “Friedrich Hayek”, “Murray N. Rothbard”, “Israel Kirzner”
    6. “Chicago School of Economics” & “Milton Friedman”
    7. “Objectivism” & “Ayn Rand”
    8. “Public Choice Economics” & “James Buchanan”

I could add more essays to his list, but I’ll restrain myself to just one: Along with the essays on the Constitution and James Madison, read “Federalists Versus Anti-Federalists” by Jeffrey Rogers Hummel.

By the way, you can still get the beautiful hardcover edition. Right now it’s half-price at the Cato Store.

“… equally efficacious, and equally a hoax.” – Benjamin Disraeli, 1848[1]


One of the highlights of the U.S. summer for Fed watchers is the annual ritual in which the Fed’s economic soothsayers peer into their crystal balls, a.k.a. their stress tests, to reassure us that the U.S. banking system is robust and getting stronger all the time.

You see, while the future is uncertain, the results of the stress tests are not. Praise be that the news is always good and getting better.

This year, the news is particularly good. As usual, the key capital metrics across the system are better than ever. And whereas in previous years there were always dunces who failed, the latest set of stress tests are the first in which all the banks passed and this year’s class laggard, Capital One, got only the mildest of slaps on the wrist.

As James Ferguson of The MacroStrategy Partnership notes in a recent commentary on the latest stress tests:

… everywhere you look, the Fed now seems to be bending the rules  in the banks’ favour. … This [stress test] appears to be a test that has been designed to be passed.”[2]

In fact, the Fed is so pleased with the performance of its stress-test examinees that it decided to reward them (or, more precisely, their shareholders) with a big dividend/buyback party that will give them a big windfall.[3] The Fed provides the punchbowl which will be paid for by other bank stakeholders including taxpayers — yes, the same taxpayers who are still being compelled to subsidize the banks (via Too Big to Fail, deposit insurance, and such like) to take excessive risks and overleverage themselves, and who stand to pay the bill if there is another crisis and the banks get bailed out again.[4]

It is curious that these capital distributions are being welcomed by many of the same people who have argued vociferously against higher capital requirements. Advocates of low capital requirements say that high capital requirements would limit banks’ lending capacity, but they fail to note that this is what dividend payouts do too.

Nor is there any sign that the Fed is inclined to take away the punchbowl any time soon. Former Fed chairman William McChesney Martin must be turning in his grave.

Indeed, plans are afoot to make future stress tests even less demanding: the days when banks felt challenged by the Fed’s stress tests are well and truly over.

So this year’s stress tests are great news for bank shareholders, but bad news for everyone else.

Headline results

 The results for the Fed’s stress tests were published in two stages. The results of the first stage – formally known as the Dodd-Frank Act Stress Tests (DFAST) 2017  — were published on June 22. These tests give results showing how the banks involved would fare under an “adverse scenario” and a “severely adverse scenario” projected over the period 2017:Q1 to 2019:Q1. The 34 bank holding companies (BHCs) tested — generally those with $50 billion or more in total consolidated assets — represent more than 75 percent of the assets of all domestic BHCs.

These scenarios project the steep decline in real U.S. GDP shown in Figure 3 of the report.

The “severe adverse scenario” suggests a shock (in real GDP terms) that is more severe than the Global Financial Crisis (GFC). It also posits the U.S. unemployment rate rising by approximately 5.25 percentage points to 10 percent, accompanied by heightened stress in corporate loan markets and commercial real estate. Equity prices fall by 50 percent through the end of 2017, accompanied by a surge in equity market volatility, which approaches the levels attained in 2008. House prices and commercial real estate prices also experience large declines, with house prices and commercial real estate prices falling by 25 percent and 35 percent, respectively, through the first quarter of 2019. Total projected losses across the 34 banks come out to $493 billion.

The results for the second stage — the Comprehensive Capital Analysis and Review (CCAR) 2017  — were published on June 28. The CCAR provides a quantitative assessment of a firm’s capital adequacy and planned capital distributions (or “capital plan”), such as any dividend payments and stock buybacks, and provides a qualitative assessment of the bigger and/or more complex firms. The principal difference between the DFAST and CCAR stress tests is that the former uses a standardized capital plan mandated by the Dodd-Frank Act, whereas the CCAR uses a firm’s planned capital actions under its BHC baseline scenario.

The significant point about this year’s CCAR was that the Fed proved to be in a remarkably generous mood: it approved the capital plans of 33 out of the 34 banks in the stress test exercise. The one exception, Capital One, was given a conditional non-objection to its capital plan, conditional on its plan making certain improvements by the end of the year.

“This year’s results show that, even during a severe recession, our large banks would remain well capitalized,” said the official in charge of the stress tests, Governor Jerome H. Powell, on June 22. “This would allow them to lend throughout the economic cycle, and support households and businesses when times are tough.”

I disagree.

Generic Weaknesses of the Stress Tests

My doubts start with the observation that the primary purpose of the stress tests program is to reassure the public that the banking system is safe. So when the Fed says the U.S. banking system is in good shape, well, they would say that, wouldn’t they? That is what central banks always say and they can realistically do no other. The stress tests are not some independent assessment of the financial strength of the banking system carried out by experts who are free to arrive at conclusions that might not suit the Fed; instead, the stress tests are part of a publicity campaign by a public agency with is own interests and agenda. Therefore, the credibility of the exercise is compromised before it has even started.

A second problem is that the stress tests place far too much emphasis on a single scenario, its “severe adverse scenario.” Recall that stress tests originated as tools for financial risk management in the private sector, and the financial risk management literature recommends against stress tests that rely on a single stress event. The reason is obvious: a firm will face a range of risk scenarios, and if its risk managers assess its vulnerability to only one of these, then their stress tests cannot give them any reassurance that it will be safe in the event of any major stress scenarios that it did not consider.

The same should apply to stress tests carried out by a central bank. Now admittedly, the Fed’s “severe adverse stress scenario” is, indeed, a severe one. However, it is still only one scenario (if one ignores the pointless “adverse scenario”), and there are other scenarios that could have a significant adverse impact on the U.S. banking system or economy more generally.[5] Examples would be a collapse in the Chinese or eurozone banking systems, a major trade war, or a major shock in the oil market. Do the Fed’s stress tests give us any assurance that the U.S. banking system could withstand such shocks and still be in good shape? No. And why not? Because the Fed’s stress tests did not even consider them.

Turning now to the modeling itself, it seems to me that the stress test projections fail a basic reality check: the Fed’s projected losses seem to be low. First, the projected loss of $493 billion from a scenario that in GDP growth or unemployment rate terms is about as severe as the GFC is barely half the losses accumulated so far since the GFC.[6]

More tellingly, Table 2 on p. 25 of the 2017 DFAST Report indicates that the stressed portfolio loan-loss rate is 5.8 percent. This is a very low loss rate for a major stress. Loan-loss rates in crises can easily be double that (e.g., the loan-loss rate in the United States in the period since the GFC is over 12 percent[7]) and sometimes even higher. These low projected loss rates suggest that the link between the severe adverse scenario and projected losses is too low to be plausible.

The modeling of the Fed’s stress scenarios over the years has been subject to a number of other major problems, some of which are set out in Morris Goldstein’s new book, Banking’s Final Exam.[8] The modeling is much  too orderly to capture key features of real-world financial crises. It understates the fat tails and nonlinearities, does not capture the amplification effects, and ignores the chaos, confusion, contagion, funding and firesale problems involved.

In addition to underestimating the impact of real-sector effects on the financial sector, it also underestimates the impact of financial-sector effects on the real sector. These factors make financial crises much more costly than normal recessions and the stress tests greatly understate them. Goldstein also provides a neat example that shows the near-impossibility of getting empirically plausible models of real-world financial crises:

(a) Note that when former Federal Reserve Chairman Ben Bernanke testified to Congress in 2007 about the subprime crisis, he estimated that it would generate total losses in the neighborhood of $50 billion to $100 billion … (b) But … when Bernanke gave testimony in an AIG court case … he explained that, by September and October of 2008, 12 of 13 of the most important financial institutions in the United States were at risk of failure within a period of a week or two. The question for stress test architects and modelmakers is, How do you make your models generate a transition from (a) to (b) in the course of, say, a year or two? This is not a technical sideshow. In stress modeling, it is the main event [emphasis added].[9]

I would also note several other problems with the Fed’s modeling. First, the Fed’s own history suggests that the Fed itself is a major risk to the banking system, e.g., through its own erratic monetary policy, and yet the Fed’s stress tests take no account of Fed risk. Second, the Fed’s prudential policies, including the stress tests, have the effect of standardizing banks’ risk management practices, thereby exposing the whole system to the weaknesses in the Fed’s own risk management models and creating the potential for hidden systemic risks to which the Fed’s and the banks’ risk models are blind.[10]

There is also a sense in which the stress tests are aiming to achieve the impossible. Frank Partnoy and Jesse Eissinger have made the compelling case that it is no longer possible even for a qualified expert to infer a bank’s true financial state from its audited accounts. But in that case one has to wonder how anyone – the Fed included – can possibly work out a bank’s financial condition from exercises such as the stress tests. It strains credibility to suggest that some questionable spreadsheet model of an arbitrary hypothetical shock can be relied upon to compensate for a major weakness in a bank’s accounts. If a bank has a large hidden (e.g., off-balance-sheet) loss that does not show up on its accounts, then how can we expect a stress test to identify that hidden loss? The answer is that we can’t. If the accounting numbers are not right, then the stress tests cannot correct for them.

Biggest Problems with the 2017 Stress Tests: Low Capital Standards and Regulatory Capture

My biggest concerns with the 2017 stress tests, however, relate to the low pass standards and to the evidence that the longstanding tug of war between the banks and the Fed over the stress tests has finally ended with the victory of the banks over the Fed.

Take the headline capital adequacy metric, the CET1 (Common Equity Tier 1) ratio —that is, the ratio of CET1 capital to Risk-Weighted Assets (RWA). I would immediately dismiss these numbers because the RWA metric is unreliable to the point of being discredited.[11]

The more reliable capital ratio metric is the leverage ratio, defined as the ratio of core capital to some measure of total exposure such as un-risk-weighed assets.

In its 2017 stress tests, the Fed uses two different leverage measures each with its own pass standard: the plain “leverage ratio,” defined as the ratio of Tier 1 capital to average assets, with a pass standard of 4 percent; and a newly introduced and more demanding “supplementary leverage ratio” with the same numerator but an expanded denominator equal to average assets plus (some) off-balance-sheet exposures, and which has a pass standard of 3 percent. These pass standards are not high, however. I find it difficult to believe that a banking system with leverage ratios close to these pass standards would be robust and able to service the real economy as we would wish it to in a subsequent stress scenario—especially bearing in mind the unreliability and gameability of the regulatory and accounting numbers, the off-balance-sheet risks that are not captured in those numbers, and the enormous scope for errors in the stress test modeling.

Moreover, both leverage ratios use Tier 1 capital in the numerator, but Tier 1 is an unreliable capital measure because it includes hybrid securities that are unreliable as core capital in a crisis.[12]

Then there is the question of what the ideal minimum required leverage ratio should be. To start with, many economists have called for higher capital requirements and/or criticized inadequate regulatory capital standards as a contributory factor to the severity of the GFC. These include: James Barth and Matteo Miller, Sheila Bair, Charles Calomiris, James Grant, Thomas M. Hoenig, Simon Johnson, Ed Kane, Norbert Michel, Gerald P. O’Driscoll, Jr.[13], The Systemic Risk Council and Sir John Vickers. In his book, The End of Alchemy, former Bank of England Governor Mervyn King wrote that a “minimum ratio of equity to total assets of 10 per cent would be a good start.”[14] Others would suggest higher minimum ratios. A famous example is an important letter drafted by Anat Admati in the Financial Times in 2010, in which no less than 20 renowned experts recommended a minimum ratio of equity to total assets of at least 15 percent. Independently, Morris Goldstein has recommended a leverage ratio of around 15 percent, John Allison, Martin Hutchinson and yours truly have called for minimum capital to asset ratios of at least 15 percent, and Allan Meltzer[15] and Walker Todd[16] have recommended a minimum of 20 percent for the largest banks.

The pass standards in the stress tests (and those in the leverage ratio tests in particular) are far from being an academic issue. Had the latter been higher, then some or all of the banks would have failed to meet the minimum capital requirements and the Fed would have been obliged to object to their capital plans — in other words, the banks would have been required to further build up their capital.

This issue is particularly acute for the 8 big systemic banks. The following table shows their DFAST and CCAR supplementary leverage ratios under the severe adverse scenario:

Table: Minimum Stressed Supplementary Leverage Ratios for U.S. Systemic Banks (Percent) Bank DFAST CCAR Bank of America 5.4 4.3 Bank of New York Mellon 5.5 4.8 Citigroup 5.5 4.5 Goldman Sachs 4.1 3.1 JP Morgan Chase 5.0 3.9 Morgan Stanley 3.8 3.2 State Street 4.2 3.6 Wells Fargo 6.1 5.3 Weighted average 5.4 4.4 Note: The pass standard is 3 percent. Numbers based on Board of Governors of the Federal Reserve System, Dodd-Frank Act Stress Test 2017: Supervisory Stress Test Methodology and Results, Tables 2 and 4; and Comprehensive Capital Analysis and Review 2017: Assessment Framework and Results, Tables 1 and 6.A. Both reports were published in June 2017.


Several points jump out.

First, the stressed supplementary leverage ratios in the DFAST column are low even before the bank-specific distributions in the CCAR column. Had the Fed applied a higher minimum pass standard for CCAR plans then most if not all of the banks’ capital plans would have been rejected: at a 4 percent minimum, the capital plans of only four banks (BOA, NY Mellon, Citi and Wells Fargo) would have been approved; at a 5 percent minimum, only Wells Fargo’s plan would have been approved; and at a 6 percent minimum none of the big banks’ capital plans would have been approved.

In this context, it is interesting to note that the Fed is in the process of imposing a 5 percent minimum “enhanced supplementary leverage ratio” on the 8 big systemic BHCs, and a 6 percent minimum on their federally insured subsidiaries, to become effective on January 1, 2018. One must wonder then why the Fed chose to impose a 3 percent supplementary leverage ratio on them in the 2017 stress tests. As a general rule, the pass standards in any central bank stress tests should be at least the minimum required standards under the capital adequacy rules, otherwise banks can be (and in the case of the 2017 CCAR seemingly were) deemed to have passed the stress tests even though their stressed leverage ratios appear to fall below the regulatory minima.

Second, the difference between the two columns — the distributions approved by the CCAR vs. those assumed in the DFAST — are remarkably high, with most of them being at least 100 basis points. Not reported in the table are the banks’ plans to increase their payout ratios — their ratios of distributions to net earnings — to nearly 100 percent on average, with some banks planning even higher payout ratios. For the Fed to approve distributions on such a scale, when the big systemic banks have such low leverage ratios and are being subsidized to run down their capital and take excessive risks too, seems reckless in the extreme.

Goldman Sachs gamed the system to near perfection: with a CCAR supplementary leverage ratio of 3.1 percent, it managed to persuade the Fed to approve a capital plan that left it with 10 basis points to spare over the Fed’s regulatory minimum. We can now presumably expect that in future CCARs, the banks’ stressed supplementary leverage ratios will converge to the minimum possible, 3 percent.

Why is the Fed approving such generous distributions when the banks’ leverage ratios are so low? Two words: regulatory capture.

Or consider the following live blog feed from MarketWatch on the Fed’s press conference on June 28th announcing the results of the CCAR:

4:30 pm EDT

One thing that has emerged from the stress tests is that banks are running very close to the minimum thresholds on capital and leverage. A Fed official didn’t deny that. That means the banks are getting good at calculating the maximum payouts they can make to investors without failing the stress tests.

4:30 pm EDT

A Federal Reserve official said banks have substantially increased their payouts, as they’ll be paying out close to 100% of projected net income over the coming four quarters, compared to 65% last year. The Fed sees that as a sign of health. “I’m pleased that the CCAR process has motivated all of the largest banks to achieve healthy capital levels and most to substantially improve their capital planning processes,” said [Governor] Powell in a prepared statement.”

I hope that I am not the only one to sense a disconnect here.

Is the U.S. Banking System Really in Good Shape?

So the Fed is confident that the U.S. banking system is in good shape – and, so much so, that on June 25 this year, Chair Yellen said that she did not expect to see another financial crisis in her lifetime. Let’s wish her a long life and hope that her prediction does not go the way of Keynes’s “We will not have any more crashes in our time” (1926) or Irving Fisher’s “Stock prices have reached what looks like a permanently high plateau” twelve days before the October ’29 crash.[17] But still the doubts creep in. For a start, she has often gotten her forecasts wrong before, even with a large staff of economists from top schools grinding them out. Consider this little beauty from January 22, 2007:

While the decline in housing activity has been significant and will probably continue for a while longer, I think the concerns we used to hear about the possibility of a devastating collapse—one that might be big enough to cause a recession in the U.S. economy—have been largely allayed.

History suggests that crises always recur and that each crisis always catches the central bank off-guard. “This time is different,” they always say, but it never is.

We should be concerned at the obvious blindspots in the Fed’s stress tests: the scenarios not considered, the low pass standards, the danger of hidden systemic risks, and the worry that the stress tests have now become an easy-to-pass compliance ritual that provides little useful information about the true state of the banks. One cannot dismiss the possibility that the good performance of U.S. banks in the latest stress tests merely indicates that banks have mastered the now routine annual stress test game. They have had plenty of practice.

We should also be concerned about how the Fed is going to achieve monetary normalization without inadvertently bursting the “everything bubble” that it has blown so lovingly since 2008. Elementary back-of-the-envelope calculations suggest that losses from a collapse in asset prices could run into trillions, i.e., multiples of the $493 billion losses in the Fed’s “severely adverse scenario” which point also suggests that the Fed’s projected losses might be well on the low side.[18]  Even a natural optimist like Robert Shiller is now warning that his CAPE index is higher than at any time since 2000, the eve of the dotcom crash, and 1929.

The Fed may regard the U.S. banking system as well nigh unsinkable but the Fed’s capital adequacy metrics and stress test pass standards on which this view is based are clearly inadequate. The U.S. banking system is also sailing in iceberg-infested waters and one worries that the Fed’s stress test radar system does not detect even those icebergs that are in plain view.


I thank Anat Admati, Jim Dorn, James Ferguson, Morris Goldstein, Martin Hutchinson, Gordon Kerr, Jerry O’Driscoll, Sir John Vickers and Basil Zafiriou for helpful feedback.

[1] M. Hutchinson, “St Januarius’ blood,” The Bear’s Lair, January 6th 2010.

[2] J. Ferguson, “Annual Fed Stress Test: Turning the dividend ATM on full,” The MacroStrategy Partnership, June 26th 2017, p. 2.

[3] As an aside, one has to ask why share repurchases by deposit-insured banks are even legal? Given bankers’ incentive to game the system by running their capital down to the lowest possible level, share repurchases are highly undesirable because they give bankers an additional and easy-to-implement means of decapitalizing their own banks. Buybacks also destabilize the banking system by making it more pro-cyclical. Banks will repurchase shares at the top, and do emergency issues or lobby for bailouts at the bottom, thereby also helping to destroy their shareholders’ wealth and inflict losses on taxpayers as well. Share buybacks may be justifiable for a company under laissez-faire conditions, but when banks are incentivized to take excessive risk and run down their capital, they become a self-dealing handout to stock-optioned management that undermines the stability of the banking system.

[4] The definitive reference on this issue is A. Admati and M. Hellwig, The Bankers’ New Clothes: What’s Wrong with Banking and What to Do about It, Princeton University Press, 2013. See also L. Gambacorta and H. S. Shin “Why bank capital matters for monetary policy,” BIS Working Papers No. 558, April 2016.

[5] The Fed also applied a global market shock to the trading portfolios of six BHCs with large trading and private equity exposures, and it applied a counterparty default component, which assumes the default of a BHC’s largest counterparty under the global market shock, to the same six BHCs and two other BHCs with substantial trading, processing, or custodial operations. However, these components are add-ons to the economic conditions and financial market environment specified in the adverse and severely adverse scenarios, and the former is merely a watered-down version of the latter. So in essence, the stress tests are still dependent on one key scenario, the severe adverse one.

[6] Ferguson, op. cit., p. 5 estimates that the cumulative loss inflicted on U.S. banks by the GFC was as much as $880 billion, excluding fines. I suppose a defender of the stress tests might argue that the Fed’s projected losses are so low because of the relatively short horizon assumed in the stress test, but this is hardly an adequate defense: the logical implication of this position is that the Fed should use a longer horizon. Remember that many of the losses associated with the GFC took much longer than 2 years to feed through into publicly disclosed realized losses.

[7] Ferguson, loc. cit. estimates that U.S. banks’ loss rate was equivalent to 12.4  percent of peak 2008:Q3 loans of $7.1 trillion.

[8] M. Goldstein, Banking’s Final Exam: Stress Testing and Bank-Capital Reform. Washington DC: Petersen Institute for International Economics, May 2017. See also here and here.

[9] Goldstein, op. cit., p. 251.

[10] Indeed, concerns about the stress test program have recently been voiced by the GAO  and even by the Fed itself.

[11] See, e.g., here, here or here.

[12] Sir John Vickers, “Response to the Treasury Select Committee’s Capital Inquiry: Recovery and Resolution,” March 3rd 2017.

[13] Gerald P. O’Driscoll, Jr., “The Financial Crisis: Causes and Consequences.” The Intercollegiate Review (Fall 2009): 4-12.

[14] Mervyn King, The End of Alchemy: Money, Banking and the Future of the Global Economy, London: Little, Brown, 2016, p. 280.

[15] Cited in A. Admati and M. Hellwig, op. cit., p. 311.

[16] “Start with 20 percent on a leverage basis, not risk adjusted, for the big boys, and then we’ll talk.” Personal correspondence, May 21st 2017.

[17] New York Times, October 17th 1929.

[18] See, e.g. K. Dowd and M. Hutchinson, “From Excess Stimulus to Monetary Mayhem,” Cato Journal Vol. 37, Number 2, Spring/Summer 2017, p. 316, note 16.

[Cross-posted from Alt-M.org]

In 2012, various properties in Van Buren County, Michigan became subject to foreclosure for property tax delinquencies. In 2014, the properties were subject to an order of foreclosure and were auctioned off to satisfy the delinquencies. Wayside Church owed $16,750 in back taxes on a parcel it used as a youth camp. When the property was sold for $206,000, Van Buren County kept the $189,250 in surplus as required by Michigan’s General Property Tax Act. Other taxpayers were similarly situated. For example, Myron Stahl and Henderson Hodgens had their properties auctioned for $68,750 to pay a $25,000 debt and $47,750 to pay a $5,900 debt, respectively.

Michigan law doesn’t recognize a right to surplus proceeds from tax sales, so the property owners sued in federal court, alleging that the county violated the Fifth Amendment’s Takings Clause when it kept the surplus proceeds from the sale of their properties. The district court dismissed the suit, precisely because Michigan law doesn’t recognize a right to surplus proceeds in such cases. On appeal, a divided Sixth Circuit dismissed the case for lack of jurisdiction. Citing the Supreme Court’s ruling in Williamson County Regional Planning Commission v. Hamilton Bank of Johnson City (1985), the court held that plaintiffs’ failure to first pursue avenues of relief in state court barred the door to federal court.

Wayside Church and the other property owners filed a petition asking the Supreme Court to take the case and clarify takings law. Along with the National Federation of Independent Business, Southeastern Legal Foundation, and Prof. Ilya Somin, Cato has filed an amicus brief supporting that petition. We argue that this case provides an excellent opportunity to preferably overrule, but at least reconsider, Williamson County’s requirement that a property owner must first sue in state court to ripen a federal takings claim.

The reality is that Williamson County’s state-remedies requirement results in constitutional absurdity: the very state court decision that a property owner must receive in order to ripen their claim simultaneously bars the owner from (re)litigating the issue in federal court. The Williamson County rule has also proven to be a potent weapon in the hands of manipulative defendants. Since the Supreme Court ruled in 1997 that a takings claim filed in state court could be “removed” to federal court (because of the federal constitutional issue), governmental defendants have removed claims to federal court, and then argued that they should be dismissed as unripe!

The state-remedies rule has no doctrinal basis and is antithetical to the Fourteenth Amendment, which was ratified to secure constitutional rights against the states and was seen as necessary to curb state government abuses. Fearing state courts could not be trusted to enforce the U.S. Constitution against their own state governments, a federal civil rights law 42 U.S.C. § 1983 was then enacted to ensure a federal forum for vindicating federal rights. Yet Williamson County has effectively gutted the protections of both of these Reconstruction-era reforms.

Before Williamson County, there was no rule that required a property owner to resort to litigation in order to ripen a takings claim, and nothing in the text of the Fifth Amendment suggests that litigation in state court is necessary to ripen a takings claim. Instead, the text should be read to recognize a ripened claim the moment property is taken if there isn’t a readily available administrative procedure for obtaining just compensation. 

The Supreme Court will decide this fall whether to take up Wayside Church v. Van Buren County.

The North American Free Trade Agreement has been a source of controversy since well before its implementation in 1994.  It was the first trade agreement involving the United States and a “developing” country, so it raised concerns that a giant sucking sound from south of the border would hoover up U.S. investment and jobs.  Ross Perot, Pat Buchanan, and most Democratic presidential candidates beginning with John Kerry all lamented the imminent or unfolding devastation wrought by NAFTA.

Even though the U.S. manufacturing sector has continued to attract more investment than every other countries’ manufacturing sectors ever since NAFTA was implemented, and even though that implementation did not accelerate the trend of U.S. manufacturing job decline, which had been underway for 14 years since employment peaked at 19.4 million in 1979 (2.6 million decline between 1979 and 1993; 2.7 million decline between 1993 and 2007; 600,000 increase between 1993 and 1999), NAFTA became a symbol of corporate excess and a rallying cry for organized labor, environmental organizations, and other anti-business groups over the years.  It also made it nearly impossible for Democrats in Congress to support trade liberalization in the ensuing decades.

During the 2008 presidential election campaign, Democratic candidates John Edwards, Hillary Clinton, and Barack Obama all vowed to re-open NAFTA to make it less unfair for U.S. workers.  Within a few weeks of assuming office, President Obama let the president of Mexico and the prime minister of Canada know that he wasn’t about to follow through on his NAFTA pledges and risk disrupting North American production and supply chains that have enabled regional producers to compete more effectively against Asian and European rivals, while delivering better goods and services at more affordable prices to consumers.

Probably owing to the anti-trade agreement fervor that brewed during the debates over Trade Promotion Authority and the Trans-Pacific Partnership over the last few years, killing NAFTA (and the TPP) became a central plank in Donald Trump’s presidential campaign. Although, regrettably, he withdrew the United States from the TPP, Trump seems to have been talked off the ledge about jettisoning NAFTA , which (as of this morning) is being renegotiated.

As a guide to better understanding what’s on the table and what’s at stake, my colleagues Simon Lester, Inu Manak, and I produced this working paper: Negotiating NAFTA in the Era of Trump: Keeping the Trade Liberalization In and the Protectionism Out.


  • When NAFTA came into force in January 1994, it was a groundbreaking achievement. It eliminated nearly all tariffs among three significant trading partners and achieved liberalization on a wide range of other issues (some of which had never before been included in trade agreements).
  • Total U.S. trade in goods and services with Canada and Mexico reached over $1 trillion in 2016, growing 125.2% in real terms since 1993.
  • NAFTA’s liberalization not only encouraged more trade, but more cross-border investment leading to deeper integration of production networks.  
  • On a historical cost basis, foreign direct investment in the United States from Canada increased from $40.4 billion in 1993 to $269 billion in 2015, and from Mexico it increased from $1.2 billion to $16.6 billion.
  • Today, 40 cents of every dollar of U.S. goods imports from Mexico and 25 cents of every dollar imported from Canada is U.S. value added (material, labor, and overhead contributed in the United States).

Still, the agreement can afford some updating to incorporate issues that were not included originally, to fix some of the flaws that have become evident over nearly a quarter of a century, and to reduce trade barriers in sectors that were originally untouched by liberalization.



NAFTA was drafted in the pre-Internet era, when E-Commerce was virtually non-existent.  Since then, we have seen the growth of online product sales, the conversion of certain products into services delivered electronically, and the development of various online “platforms.”  E-Commerce is now a standard way of doing business for many companies, in both domestic and international trade.  The free flow of information is essential to free trade in electronic commerce, as well as to the industries for which data are crucial components of the product or service.  While governments have grappled with various responses to the growth of E-Commerce, trade policy has been slow in keeping pace with the changes.  But in recent years, U.S. trade negotiators have been making some progress on this issue. 

State-Owned Enterprises

State-owned enterprises are commercial enterprises in which the state has majority ownership, controlling ownership, or the ability to appoint a majority of the board of directors. SOEs have become more prominent actors in the global economy in recent years. Concerns about trade distortions and unfair competition have grown, as SOEs that had previously operated almost exclusively within their own territories are increasingly engaged in international trade and cross-border investment. According to the Office of the U.S. Trade Representative, there was only one SOE on the list of the Fortune Global 50 largest companies in the world in 2000.  By 2015, there were close to a dozen.  The purpose of trade agreement rules on SOEs is to curtail their market distorting effects.

Regulatory Cooperation

Since NAFTA reduced almost all tariffs to zero, some of the largest remaining barriers to North American trade take the form of non-tariff barriers, or so-called “behind the border” measures. Differences in standards and regulations are frequently cited as key obstacles to trade, and many of these regulatory divergences are often the result of regulation occurring in domestic silos, without giving much thought to how those rules may impede trade.

Trade Facilitation

In broad terms, trade facilitation includes reforms aimed at improving the chain of administrative and physical procedures involved in the transport of goods and services across international borders. Countries with inadequate trade infrastructure, burdensome administrative processes, or limited competition in trade logistics services are less capable of benefiting from the opportunities of expanding global trade.  Like tariff cuts, improvements in trade facilitation procedures can help reduce the cost of trade and increase its flow. 

Dispute Settlement

The enforceability of the rules of a trade agreement is of great importance because if governments cannot be held to honor their commitments, the value of those commitments is significantly diminished. Inevitably, disputes arise over the meaning of the obligations in trade agreements. Typically, dispute settlement involves ad hoc panels of experts who hear claims from either governments or private actors, and then issue rulings on those claims. Essentially, these panels act as the judicial arm of the trading system.  If trade agreements are to be reliable and effective, they must include enforcement mechanisms. The degree of enforceability varies depending on the design of the particular system.  In this section of the working paper, we discuss how Chapters 20, 19, and 11 may be addressed in the NAFTA renegotiation.

Rules of Origin

Rules of Origin (RoO) establish the parameters used to determine whether an imported good “originates” within the region, thereby qualifying for the preferential treatment accorded under the agreement. Generally, a product is considered originating if it was wholly made within the region (in the countries party to the agreement), if it was significantly transformed within the region from imported materials and components, or if the relative value of originating materials and manufacturing performed in the region meets a specific threshold.

RoO are necessary components of preferential trade agreements. When products from different countries receive different tariff treatment, importers, exporters, and customs officials must have a way to determine which tariff rates apply.  Rules that permit greater use of non-originating inputs or broader definitions of what constitutes product transformation tend to be more trade liberalizing than more proscriptive rules, which impose greater restrictions on qualification for the agreement’s preferential tariff rates. In today’s globalized economy, strict rules of origin impede the evolution and operation of more efficient supply chains and can be used to privilege existing producers by limiting competition.

Canada’s Supply Management System

Canada’s supply management system for dairy, poultry, and eggs was excluded from NAFTA. The renegotiation offers an opportunity to bring these highly-protected sectors into the agreement, so as to increase market access for U.S. producers, but also to improve purchasing choices for Canadian consumers. For instance, Canada maintains a 270 percent tariff on milk, and the above-quota tariffs on cheese and butter can reach as high as 245 percent and 298 percent, respectively.

Softwood Lumber

The 35-year old lumber dispute between the United States and Canada must be resolved, somehow.  U.S. producers have long complained about Canadian land-management practices that, they claim, provided unfair subsidies to Canadian producers.  Three and a half decades of litigation, temporary supply management agreements, and, frankly, defiance of the trade rules by the U.S. government have strained trade relations and burdened U.S. lumber users and home buyers.  Whether an agreement can be reached within NAFTA is unclear – given that it probably would require the U.S. industry to disavow use of the antidumping and countervailing laws.


NAFTA was the first international trade agreement with significant commitments to liberalize trade in services, and one of the most prominent NAFTA disputes concerns the provision of trucking services in the United States by Mexican firms.  However, in the ensuing years, services liberalization in other trade agreements has evolved to include broader obligations covering more areas of trade, leaving NAFTA’s rules insufficient and outdated.  In the NAFTA renegotiation, there is as an opportunity to take a fresh look at services, such as legal, medical, and educational services, that are more easily traded across borders than was the case in the early 1990s. With growing concerns over the absence of sufficient competition in the provision of U.S. healthcare services, the United States could benefit by allowing more foreign competition in the medical services and health insurance markets.


Trade Deficits

President Trump and members of his administration seem to believe that trade is a zero sum game played between Team USA and – in the case of NAFTA – Team Canada and Team Mexico.  They consider exports to be Team USA’s points, imports to be the foreigners’ points, and the trade account to be the scoreboard.  Since the United States runs a persistent deficit with Mexico (and an occasional deficit with Canada), they conclude that America is losing at trade.  And it’s losing because of poorly negotiated trade deals or outright cheating on the part of the foreign teams.

Secretary of Commerce Wilbur Ross and National Trade Council advisor Peter Navarro have suggested that our trade agreements should include some sort of trigger mechanism, so that if trade balance or some other objectives related to reducing deficits or increasing surpluses are not reached, the deal would be reopened and renegotiated.  That idea, frankly, is absurd.  What would motivate businesses to invest in cross-border relationships or to commit to capital investments if the terms of the underlying trade agreement were prone to such uncertainty and potential upheaval?

Moreover, the bilateral trade balance is not a measure of the success or failure of a trade agreement.  It is a meaningless statistic.  A better measure of the success of a trade agreement is its effect on total trade and investment, and its effect on economic growth.  By those metrics, NAFTA has been an unbridled success.  Furthermore, “despite” 41 straight years of registering annual trade deficits, the U.S. economy has grown significantly, in part because it has benefitted from the capital account surplus (foreign purchases of U.S. assets exceeding U.S. purchases of foreign assets) that goes along with a current account deficit.

The Trump administration’s obsession with trade deficits—especially bilateral trade deficits—is misguided.  The point of trade agreements is to reduce artificial impediments to trade and to provide greater certainty.  It would create vast uncertainty and be a serious mistake to push for any provisions in NAFTA that use the trade account as a trigger of some future action.

Border Taxes

Another problematic suggestion from the Trump trade team is that they would like to use trade agreements to take on the “border adjustment taxes” associated with value-added taxes used by many countries.  Trump advisors have expressed concerns over alleged advantages bestowed upon Mexican producers by way of the rebate of value-added taxes upon export and the assessment of taxes on U.S. products upon import.

A proper understanding of these border adjustment taxes is that their impact on trade is much more benign. The general understanding, according to WTO rules, is that value-added taxes are a consumption tax, similar, in nature, to a sales taxes, and thus a non-discriminatory application of such taxes to domestic and imported products is permitted.  Not that it would necessarily be a good idea, the U.S. government is free to adopt a similar VAT system with rebates upon export or some variation, such as a destination based cash flow tax, which was under consideration in Congress (although such a measure would, of course, have to be applied in a non-discriminatory way).  As of this writing, that controversial proposal seems to have been abandoned.

Regardless, this is not a NAFTA problem or a Mexico problem. Much of the world uses value-added taxes or some similar form of consumption taxes. If the U.S. government has concerns, it can address them at the WTO. There is a long history of disagreement over this issue in the GATT, and finding a way to raise it again at the WTO would not be out of line. Raising it in NAFTA, on the other hand, would be a pointless distraction.

Currency Manipulation

Another controversial issue the Trump administration is likely to press is currency manipulation. While economists are divided in their opinions about how to define currency manipulation and whether it constitutes an especially pernicious offense worthy of counteractions, even many of those who think currency manipulation should be discouraged are skeptical of the wisdom of including punitive provisions in trade agreements.  Even though Canada and Mexico do not generate much concern in the United States as likely currency manipulators, the Trump administration sees NAFTA as an opportunity to set a precedent for future agreements by crafting and incorporating such provisions without much resistance in this agreement.

China and Japan are the countries most often cited as the reasons for including currency manipulation provisions in trade agreements.  But we should be skeptical about the assumptions underlying the relationship between currency manipulation and trade flows.  Given that intermediate goods trade predominates total trade, as a result of the proliferation of global supply chains,  the relationship between currency values and trade is not straightforward.

If only 50 percent of the value of a country’s exports is domestic content and labor (and the other 50 percent is foreign value), as is the approximate case with China, the impact of currency values on trade flows is mitigated.  This helps explain why, despite a 38 percent appreciation of the renminbi against the dollar between 2005 and 2013, the bilateral U.S. trade deficit with China didn’t decrease. Rather, it increased by 46 percent. 

Even if deemed desirable, crafting the right currency manipulation provision would be more difficult than the administration thinks.  There is vast disagreement among economists as to how to identify and measure the effects of currency manipulation. Meanwhile, U.S. government agencies have conflicting views about incorporating currency provisions in trade agreements.  While the Commerce Department—with its mission to protect U.S. industries in the global market place—might favor another weapon in its trade remedy arsenal, the Treasury Department, which has long held sway over financial and economic issues related to currency, is less inclined to take actions in currency markets to achieve trade objectives.

And while the Canadian and Mexican governments might not be overly concerned that their actions would be the targets of such provisions, they will certainly have their own opinions about whether and how such rules should be crafted. Thus, balancing domestic and international views on this issue will take time. If the Trump administration chooses to push this issue, the NAFTA renegotiation may not go as quickly as it wishes.

Buy America

Trump and his trade team have spent a lot of time promoting the idea of Buy America, and have criticized trade agreements that restrict the ability of the U.S. government to buy exclusively from Americans. In the leaked draft memo about NAFTA, it was suggested that the NAFTA renegotiation would provide more flexibility for such Buy America programs. What those pushing for expanded Buy America do not seem to realize is that this issue works both ways.

Cordoning off more of the estimated $1.7 trillion U.S. government procurement market to U.S. suppliers would mean higher price tags, fewer projects funded, and fewer people hired. In today’s globalized economy, where supply chains are transnational and direct investment crosses borders, finding products that meet the U.S.-made definition is no easy task, as many consist of components made in multiple countries. And by precluding foreign suppliers from bidding, any short-term increases in U.S. economic activity and jobs likely would be offset by lost export sales – and the jobs that go with them – on account of copycat protectionism abroad. If Americans are compelled by law to buy more U.S. product from U.S. firms, then the Canadians and Mexicans will insist on the same.

Buy America may sound good in a campaign speech or on a bumper sticker, but in practice it just means higher price tags for taxpayer funded projects and fewer opportunities for U.S. exporters.

On the campaign trail, Donald Trump claimed that he would cancel President Obama’s Deferred Action for Childhood Arrivals (DACA) program, which allows young unauthorized immigrants—known as “Dreamers”—to live temporarily without fear of removal and work legally. To many people’s surprise (including mine), President Trump decided in January to maintain the program, issue renewals, and even allow new applicants into the program. But this will likely change soon.

Now several states led by Texas are attempting to force the president’s hand, requesting in a letter that he terminate the program by September 5 or face a lawsuit. Texas already successfully challenged President Obama’s 2014 attempt to expand DACA to a broader range of immigrants who came here as children and to create a new program for undocumented parents of U.S. citizens called DAPA. This makes a lawsuit very likely to succeed. Even if it were possible for the Trump administration to defend DACA legally, it is not clear that Attorney General Jeff Sessions would want to defend a policy that he has called constitutionally “very questionable.” In July, he refused to say he would defend it.

After the election, President Trump promised to “work something out” with the Dreamers, and in July, President Trump said that he alone will decide the future of DACA. But the politics just became much more difficult. Texas has essentially forced him to defend in court something that he had characterized as illegal amnesty on the campaign trail and something his biggest supporters hate. Moreover, Trump’s record with the courts is already making him appear inept, so he would likely not want to take the political hits when he could easily lose the case anyway.

If the president does rescind the DACA memorandum on September 5, the program will likely not disappear overnight. Rather, it will slowly wind down over the next two years.

Will Dreamers become priorities for removal?

DACA has three different aspects—deprioritization for removal, “deferred action,” and employment authorization. First, the DACA memo tells agents to prevent Dreamers who may be eligible for DACA “from being placed into removal proceedings or removed from the United States.” This provision deprioritizes their removal. Under new expansive enforcement priorities laid out in a February 20th memo from then-Department of Homeland Security (DHS) Secretary John Kelly, many DACA recipients would be targets for removal if Kelly’s successor rescinds the DACA memo. That’s because the Kelly memo creates “priorities” so expansive that they include nearly all unauthorized immigrants, while specifically rescinding all “conflicting” memos except for the DACA memo and the memo that expanded DACA and provided for the never-implemented DAPA program. 

Like the DACA memo, the DAPA memo instructed Immigration and Customs Enforcement (ICE) “to… prevent the further expenditure of enforcement resources with regard to these individuals” (i.e., undocumented parents of U.S. citizens and DACA children). After initially leaving it in place, Secretary Kelly rescinded the DAPA memo on June 15. Since then, DAPA-eligible immigrants have been eligible for removal, though it appears that ICE was already ignoring those provisions of the DAPA memo. If the DACA memo were rescinded without further DHS guidance, the same thing would happen to the Dreamers that it protects.

Texas and the other states threatening a lawsuit claim that their “request does not require the Secretary to alter the immigration enforcement priorities contained in his separate February 20, 2017 memorandum.” Nor does it “require the federal government to remove any alien.” But their letter simultaneously calls for the full rescission of the DACA memo, which would effectively alter the enforcement priorities in the February 20th memo and, absent any additional guidance, require the federal government to target Dreamers for removal.

Will Dreamers immediately lose deferred action?

The second aspect of DACA is “deferred action.” Deferred action provides protection against arrest by formalizing the decision not to remove them. If an ICE agent encounters a DACA recipient and enters their name into the agency’s database, they will find their status listed as “lawfully present.” DACA applicants submit a form I-821D application to U.S. Citizenship and Immigration Services with a $495 application fee, which covers the cost of a biometric background check. As of March 2017, 886,814 young immigrants had received an approved I-797 Notice of Action form under DACA. Each I-797 form is valid “unless terminated” for two years.

Figure 1: DACA Form I-797 Notice of Action

Source: Imgur

If DHS cancels the DACA memo, it’s unclear whether the underlying deferred action grants would disappear. The threat letter from the states only requests that “the Executive Branch will not renew or issue any new DACA or Expanded DACA permits in the future.” The form itself states that it will “remain in effect for 2 years” unless “terminated.” Does this require an individualized termination for some violation of the program’s rules or would the termination of the program under which the form was issued suffice?

When Secretary Kelly rescinded the DAPA and Expanded DACA memo, he specifically stated that the memo would not “alter the remaining periods of deferred action under the Expanded DACA policy granted between the issuance of the November 20, 2014 Memorandum and the February 16, 2015 preliminary injunction order in the Texas litigation.” This refers to the 3-year renewals issued to original DACA recipients who, both before the November 20th DAPA/Expanded DACA memo and after the injunction, received only 2-year renewals. It is possible that the administration could duplicate this language in the DACA repeal itself.

Another option would be for the administration to leave it unclear, and if it does so, I suspect that individual ICE agents will end up making the determinations. According to the Feb. 20th Kelly memo, “the individual, case-by-case decisions of immigration officers” will have the final say over whether to make arrests.

Will Dreamers immediately lose employment authorization?

The final aspect of DACA is employment authorization. Every DACA recipient receives a 2-year (or 3-year, if they renewed in the period mentioned above) employment authorization document (EAD). DACA applicants must separately submit a Form I-765 to receive an EAD. You can see an approved DACA EAD in Figure 2. The card contains a two-year validity period.

Figure 2: DACA-based Employment Authorization Document (EAD)

Source: WangLawOffice

Because employment authorization is dependent on a grant of deferred action, canceling part 2 of DACA would legally end part 3 as well. However, the memo that rescinded the DAPA/Expanded DACA memo also clarified that it does not “affect the validity of [3-year] Employment Authorization Documents” granted before the injunction in the Texas case. Again, it’s possible that the administration will repeat this clarification in the DACA rescission. In any case, the EADs contain no indication that they were issued under DACA, and employers must accept any facially valid, non-expired ID when screening applicants for employment authorization. I have previously written about the Obama administration’s efforts to cancel certain DACA EADs and how difficult it was in my pre-inauguration post explaining how DACA could end. Given that this circumstance seems quite unlikely, I will not repeat myself here.

How DACA will end

We know roughly how DACA will wind down because we know about when DHS approved the DACA applications. DHS publishes quarterly approval figures, and we know that it issued 108,000 3-year renewals from November 20, 2014 to February 16, 2015. Figure 3 shows the approval schedule for DACA as well as the expected expiration schedule if DHS cancels the program on September 5, 2017, as requested in the states’ letter. As it shows, the DACA grants would expire over a two-year period ending in September 2019.

Figure 3: DACA Approvals—Initial and Renewals—and Projected Expirations

Source: USCIS Data Set: Form I-821D Deferred Action for Childhood Arrivals. April 2017 to September 2017 figures are not published yet and are projections based on the number of two-year renewals in those months in 2015.

Unfortunately, the September 5 deadline will occur just before the expiration of the 108,000 three-year permits. Had President Obama issued only two-year renewals, these individuals would have already renewed their status a year ago, receiving an additional year of protection. Figure 4 provides the expirations by year. More than 60 percent of DACA recipients will still have protection and lawful employment through June 2018. More than a third will still be participating in DACA through January 2019—well over a year after the rescission of the DACA memo.

Figure 4: Projected DACA Expirations by Year

Source: Author’s calculation based on USCIS DACA Approval Figures.

The most likely scenario for DACA’s demise is that it will slowly wind down. President Trump has shown no willingness to move more aggressively against DACA recipients than is necessary, although certain ICE agents seem zealous about targeting them. If the president does determine that DACA should end, or the states receive an injunction against the program, Congress will need to act quickly as the permits expire. I suggest that they look seriously at the Recognizing America’s Children (RAC) Act, which would provide legal status and a pathway to citizenship for these immigrants brought to the U.S. as children.

Add yesterday’s rage-spasm of a press conference to the growing list of reasons reasonable people are inclined to worry about Donald Trump’s proximity to nuclear weapons. In addition to what it suggested about Trump’s moral compass (“Very fine people” aren’t attracted to posters that look like this), his performance also highlighted questions about the judgment, temperament, and impulse control of the man entrusted with the world’s most fearsome arsenal.

Last week, recall, Trump threatened North Korea with nuclear annihilation: “North Korea best not make any more threats to the United States…. They will be met with fire, fury and frankly power the likes of which this world has never seen.” “Fire and fury” was ad-libbed, apparently, but on Thursday, he upped the ante: “if anything, that statement may not be tough enough.” (For a cooldown lap, on Friday, Trump warned he was “not going to rule out a military option” in Venezuela.)

When you’re faced with a president who has weekly meltdowns on Twitter and likes to “wing it” with nuclear threats, it tends to concentrate the mind painfully on the legal and practical restraints to presidential power. Does the president have the constitutional power to launch a nuclear first strike on a country for “mak[ing] threats”? If he decides to act on that impulse, is there anything Congress can do to stop him?

The first question’s the easy one: the answer is no. In the absence of an imminent attack, the president has no constitutional power to rain down “fire and fury” on North Korea. As Ilya Somin explains here, “the Constitution very clearly reserves to Congress the power to start a war.”

The president retains some independent power to act defensively: to “repel and not to commence war” or “repel sudden attacks,” as Madison’s notes from the Convention put it. We can argue about whether a second strike—launch under attack—is included within this power. But the constitutional power to “repel sudden attacks” doesn’t include the power to launch them.

The whole point, as James Wilson told the Pennsylvania ratifying convention in 1787, was to design a “system [that] will not hurry us into war…. It will not be in the power of a single man… to involve us in such distress; for the important power in declaring war is vested in the legislature at large.”

Is there anything Congress can do to prevent a trigger-happy president from hurrying us into nuclear war?  Congressman Ted Lieu has drafted a bill that he hopes will do just that.  HR 669, the “Restricting First Use of Nuclear Weapons Act,” provides that “The President may not use the Armed Forces of the United States to conduct a first-use nuclear strike unless such strike is conducted pursuant to a declaration of war by Congress that expressly authorizes such strike.”

Someone from the John Yoo school of constitutionalism might argue that the law encroaches on presidential prerogatives by “micromanaging” the means available to protect national security. But Lieu’s bill is clearly constitutional: if Congress can tell the president not to use ground combat troops in a particular war, it has the legal authority to bar him from launching an unauthorized nuclear first strike.

Would Lieu’s bill work, though? Here, I have my doubts.

In principle, the military is not supposed to obey an illegal order by the president, but in practice, they follow the chain of command. In all but the most extreme cases—e.g., orders to commit war crimes—it’s good that they do. Civilian control of the military is a core principle of free government, and it’s almost always better preserved by erring on the side of military obedience to the CINC, even where his authority may be in doubt.

As a practical matter, though, that means that simply passing a law won’t necessarily restrain presidential warmaking. Our experience with the War Powers Resolution makes that clear. Put aside the question of how the law applies to limited, short-term military operations: if the WPR does anything, it bars the continuation of hostilities beyond 60 days unless Congress has declared war or passed a specific statutory authorization for war. In fact, “the view that the 60-day clock is constitutional has remained consistent executive branch law throughout the Reagan, Bush-Quayle, Clinton, Bush-Cheney, and Obama administrations.” But twice in recent decades, presidents have blown by that limit: first in Kosovo in 1999; second in Libya in 2011. Each time, the clear language of the law didn’t matter: the president ordered the military to keep bombing, and they did.

Why would Lieu’s “no-first-use act” fare any differently? The nuclear launch process is built for speed—designed to give the president the power to launch a second strike in mere minutes, while enemy missiles are still in the air. That also makes it difficult to stop a president who decides to launch first. The guys in the bunkers who turn the keys have surely been vetted to weed out “question authority” types. In the ‘70s, there was actually an Air Force major who was training for the job, and when he asked, “how can I know that an order I receive to launch my missiles came from a sane president?”—that was the end of his career.

Trump can’t literally “push the button” and have Pyongyang obliterated. But once the order is authenticated with the codes on the “biscuit,” and it goes out from the Pentagon, it’s likely already too late.

The one effect Lieu’s bill might have, and this is speculative, is it might embolden the president’s secretary of defense or the chairman of the Joint Chiefs to put up roadblocks before the order is transmitted. (I say “or the Chairman of the Joint Chiefs” because it appears that the president has the option to issue the order without going through the Secretary of Defense).

There’s a story about Richard Nixon, whose behavior during the Final Days—frequently drunk and raving—made people worry about his access to nukes. In a meeting with congressmen, Nixon blurted out that: “I can go in my office and pick up a telephone, and in 25 minutes, millions of people will be dead.” One of the attendees, Senator Alan Cranston called then-Secretary of Defense James Schlesinger, worried about “the need for keeping a berserk president from plunging us into a holocaust.” Schlesinger apparently told the top military brass that if the president issued any unusual orders, they needed to check with him first.

The practical effect Lieu’s bill could have, in the unlikely event it passed, might be to buttress a future Schlesinger. By giving the Secretary of Defense or the Chairman added cover to get in the way of a decision he knows to be crazy, it could slow or stop the delivery of the order from the Pentagon to the launch officers. For that reason alone, it’s worth considering. But when you’re depending on a guy whose nickname is “Mad Dog” to preserve sanity within the executive branch, you’re already in a pretty dangerous place.

I am pleased to announce that this week the Cato Institute published the Encyclopedia of Libertarianism online, unabridged, and unpaywalled

The Encyclopedia was my first project at Cato, and if you ask me its current format is exactly where it belongs. Some of the articles are a bit dated by now, and there are some regrettable gaps in the first edition. We will be working on those in the coming months to create a new and improved Encyclopedia. In the meantime, though, feel free to browse. I still think the original is a pretty good statement of what libertarianism is all about.

When the Department of Labor (DoL) rolled out its fiduciary duty rule last year, I (and others) noted that its likely effect would be to harm the very people it purports to protect.  Unfortunately, it seems I was right.

The rule is intended to help individuals make good choices when saving for retirement.  Under the rule, brokers who sell retirement investments are to be held to a “fiduciary duty” standard.  This is often expressed in the legal world as the care a prudent person would take in managing his or her own affairs.  Those who hold positions of great trust, such as those who are given authority to act for another, are often designated fiduciaries under the law.  For example, corporate board members are fiduciaries of the company they serve, and lawyers owe a fiduciary duty to their clients. 

As I’ve discussed previously, the legal reality of being subject to a fiduciary duty standard is much more than simply deciding to offer good customer service.  And “more” in this case means more liability and more cost.  The risk is therefore that brokers may find it’s just not worth the risk or the cost to serve clients with only moderate amounts of money to invest.

It seems a recent poll of the industry shows my predictions may be correct.  According to a letter submitted to the DoL by the Financial Services Roundtable, its members have reported the following trends as a result of the rule:

 (1) less guidance and support to IRA owners and small plans; (2) increases in minimum account size; (3) limited product shelf; (4) shift to fee-based accounts; (5) moving clients with smaller accounts to self-service or robo-advice; (6) orphaning of smaller, less profitable accounts due to heightened risks; (7) reduced willingness to discuss or consider unmanaged assets with clients due to risks; (8) poor client service due to the time required to perform comparative analysis on the proposed account to the existing account; (9) disinclination to sell annuity products because of uncertainty surrounding the Rule and inability to launch new products because resources are tied up with Rule implementation; (10) additional disclosure documents and other changes to sales process make the sales process markedly longer in each client appointment; (11) less discretion on small accounts and compensation changes make working with small accounts more challenging and less cost-effective for financial professionals;  (12) higher manufacturing and distribution costs; and (13) new liability concerns.

Specifically, the poll found that 68 percent of respondents would be taking on fewer small accounts due to increased compliance costs and legal risks.  It also found that 63 percent expected that the new rule would limit the investment options or products the firms could provide to their clients.  And that 52 percent expected that higher compliance costs would be passed on to clients in the form of additional fees. Only 12 percent of respondents said the rule “is helping me to serve my clients’ best interests.”  Most notably, the poll report highlights the following: “Even advisors who say that the rule is ‘helping me to serve my clients’ best interests’ or has had ‘no impact on my ability’ say that there will be more complicated paperwork and fewer small accounts.”

An anonymous source sent an advanced copy of S.1757, otherwise known as the “Building America’s Trust Act,” to Ars Technica. If passed as written, the bill would dramatically expanded surveillance at the border and ports of entry, putting the privacy of immigrants and citizens alike at risk.

The bill, sponsored by Sen. John Cornyn (R-TX) and co-sponsored by six of his Republican colleagues, mandates increased border drone surveillance and the collection of more biometric information, including but not limited to voice prints and facial scans.

The drone provisions of the bill are consistent with President Trump’s campaign rhetoric. During his campaign, he said drones should patrol the border 24/7. Cornyn’s bill doesn’t quite go that far, requiring Customs and Border Protection (CBP) to fly drones at least 24/5.

Drone surveillance at the border isn’t new, nor is it effective. In 2014 the Department of Homeland Security’s Office of Inspector General (OIG) released a report on CBP’s drone operations. It found that the drone program, which includes Predator B drones originally designed for military use, did not achieve expected results and contributed to very few apprehensions of illegal border crossers and marijuana. The report also found that the drone program cost $12,255 per flight hour. In FY2013, CBP’s drones flew for 5,102 hours for a combined cost of around $62.5 million. 

This was a large expense for an inefficient border security tool. Aside from the fiscal costs, the increased use of drones on the border will worsen the militarization of the border, with American citizens being under the ever-snooping eye of border patrol surveillance equipment. In 2013, Americans on the border were already regularly seeing military-grade surveillance tools in the air. From a 2013 New York Times report:

The United States-Mexico border has become a war zone. It is also a transfer station for sophisticated American military technology and weapons. As our country’s foreign wars have begun to wind down, defense contractors look here, on the southern border, to make money.

Lately it has become entirely normal to look up into the Arizona sky and to see Blackhawk helicopters and fixed-wing jets flying by. On a clear day, you can sometimes hear Predator B drones buzzing over the Sonoran border. These drones are equipped with the same kind of “man-hunting” Vehicle and Dismount Exploitation Radar (Vader) that flew over the Dashti Margo desert region in Afghanistan.

CBP drones do not have to be large, military-grade predator drones. Earlier this year I noted that The Department of Homeland Security (DHS), CBP’s parent agency, is interested in small, portable drones outfitted with facial recognition technology.

Facial recognition is also mentioned in S.1757 as part of the bill’s passport screening section, requiring that CBP “utilize facial recognition technology or other biometric technology” to “inspect travelers at United States airports of entry.”

DHS stated earlier this year in a privacy impact assessment concerning facial recognition, “the only way for an individual to ensure he or she is not subject to collection of biometric information when traveling internationally is to refrain from traveling.” There are already facial recognition trials at airports in six American cities, but the technology will become more prevalent if S.1757 is passed.

Unlike other biometrics such as fingerprints and DNA, facial images can be obtained from individuals who haven’t come into contact with the criminal justice system, the primary source of most biometrics. About half of American adults are already in a facial recognition network thanks to these networks’ access to DMV and passport data. If American citizens’ passport and driver’s license facial images are going to be included in airport facial recognition systems there should be strict limits on sharing these images with other law enforcement agencies and how long facial scans are stored. Although CBP states that facial scans at airports are deleted after 14 days that commitment is policy, not law, and could change.

S.1757 mandates that a biometric exit program will be implemented in America’s 15 busiest airports, seaports, and land ports of entry within two years of enactment. One section of S.1757 requires DHS to upgrade or build a biometric system that enables DHS to store alien iris scans and voice prints that federal, state, and local law enforcement can access.

As if the surveillance technology outlined in S.1757 wasn’t concerning enough, the legislation also includes an increase in the number of border patrol agents (at least 26,370 from around 20,000) and more federal judges in southern border states. Under the Border Patrol Strategic Plan included in S.1757 DHS will consider using military technology.

More drones and biometric technology tools on the border and at ports of entry should concern anyone who values privacy. CBP has statutory authority to stop vehicles within 100 miles of the border, so even those people who don’t live near the border risk being exposed to CBP surveillance tools. In addition, the policies outlined in S.1757 will require DHS to collect more information about American citizens and foreigners alike, information that could be used for reasons other than border security.

Writing at the Niskanen Center, Samuel Hammond has some harsh words for libertarians. It’s a short step, he says, from anti-statism to some particularly ugly forms of nationalism:

The appeal of white nationalism to libertarian anti-statists should not be surprising. After all, nationalist and revanchist movements have historically represented powerful tools for mobilizing secession and other forms of political resistance to “the state.” Their common cause is all the stronger in multicultural, liberal democracies where ethnic grievances can be called upon to portray “the state” less as a political compact between competing groups, and more as tyrannical sovereign infringing on some sub-group’s right to self-determination.

To the extent that he’s right about this, that’s pretty embarrassing. Hammond cites AnCap YouTube to argue that there have been all too many who took this path. I’m not sure that it’s fair to judge anyone else by AnCap YouTube, although his judgment on some of them is certainly correct.

Other parts of his essay I think are quite wrong: It’s not necessarily crazy or evil to think that the state should be at least somewhat congruent to the nation. That proposition does not necessarily entail ethnonationalism, and certainly doesn’t when I assert it. A nation, as an imagined community, need not be ethnic at all. A pluralist nation may include people of many different ethnicities, religions, and other affiliations. The American nation has always been pluralist in its aspirations. Throughout our history we have increasingly delivered on the promise of pluralism, not just to favored groups, but to all. That work should continue, and if saying “you too are a part of this nation” can help with the task, then we should say it loudly and often.

Hammond also claims that “liberty needs the state.” On this point I am sure that the Niskanen Center will get the usual howls of protest from exactly the people who should be the least surprised. Of course the Niskanen Center would say something like this. But is it true?

It’s clearly correct to say, with Hammond, that in many cases “state sabotage automatically empowers the most dominant and dominating subgroups in our otherwise open society,” it’s much less clear that this must always be the case.

The way forward for radical libertarians and others who dream of a stateless (or just a less state-dominated) society consists of figuring out how to manage these tendencies toward domination, so that when the state does retreat, it is individual that liberty advances, rather than some other form of unjust domination.

I don’t know quite to what extent the project can succeed. But I think it’s reasonable to expect that we can enjoy a much smaller state than the one we have right now. Reasonably as well, this development could leave the vast majority of citizens, and particularly the least well off, better off by a range of widely acceptable criteria. What seems in order is not a broad declaration for or against the state, but a constant and relentless tinkering on the margins, with the aim of delivering less arbitrary domination of one person or group by another. Racial groups most certainly included.

To be blunt, Republicans are heading in the wrong direction on fiscal policy. They have full control of the executive and legislative branches, but instead of using their power to promote Reaganomics, it looks like we’re getting a reincarnation of the big-government Bush years.

As Yogi Berra might have said, “it’s déjà vu all over again.”

Let’s look at the evidence. According to The Hill, the Keynesian virus has infected GOP thinking on tax cuts.

Republicans are debating whether parts of their tax-reform package should be retroactive in order to boost the economy by quickly putting more money in people’s wallets.

That is nonsense. Just as giving people a check and calling it “stimulus” didn’t help the economy under Obama, giving people a check and calling it a tax cut won’t help the economy under Trump.

Tax cuts boost growth when they reduce the marginal tax rate on productive behavior such as work, saving, investment, or entrepreneurship. When that happens, people have an incentive to generate more income. And that leads to more national income, a.k.a., economic growth.

Borrowing money from the economy’s left pocket and then stuffing checks (oops, I mean retroactive tax cuts) in the economy’s right pocket, by contrast, simply reallocates national income.

Indeed, this is one of the reasons why the economy didn’t get much benefit from the 2001 Bush tax cut, especially when compared to the growth-oriented 2003 tax cut. Unfortunately, Republicans haven’t learned that lesson.

Republicans have taken steps in the past to ensure that taxpayers directly felt the benefits of tax cuts. As part of the 2001 tax cuts enacted by President George W. Bush, taxpayers received rebate checks.

The article does include some analysis from people who understand that retroactive tax cuts aren’t economically beneficial.

…there are also drawbacks to making tax changes retroactive. …such changes would add to the cost of the bill, but would not be an effective way to encourage new spending and investments. “It has all of the costs of the tax cuts but none of the economic benefits,” said Committee for a Responsible Federal Budget President Maya MacGuineas, who added that “you don’t make investments in the rear-view mirror.”

I’m not always on the same side as Maya, but she’s right on this issue. You can’t encourage people to generate more income in the past. If you want more growth, you have to reduce marginal tax rates on future activity.

By the way, I’m not arguing that there is no political benefit to retroactive tax cuts. If Republicans simply stated that they were going to send rebate checks to curry favor with voters, I’d roll my eyes and shrug my shoulders.

But when they make Keynesian arguments to justify such a policy, I can’t help but get upset about the economic illiteracy.

Speaking of bad economic policy, GOPers also are pursuing bad spending policy.

Politico has a report on a potential budget deal where everyone wins…except taxpayers.

The White House is pushing a deal on Capitol Hill to head off a government shutdown that would lift strict spending caps long opposed by Democrats in exchange for money for President Donald Trump’s border wall with Mexico, multiple sources said.

So much for Trump’s promise to get tough on the budget, even if it meant a shutdown.

Instead, the back-room negotiations are leading to more spending for all interest groups.

Marc Short, the White House’s director of legislative affairs, …also lobbied for a big budget increase for the Pentagon, another priority for Trump. …The White House is offering Democrats more funding for their own pet projects.

The only good news is that Democrats are so upset about the symbolism of the fence that they may not go for the deal.

Democrats show no sign of yielding on the issue. They have already blocked the project once.

Unfortunately, I expect this is just posturing. When the dust settles, I expect the desire for more spending (from both parties) will produce a deal that is bad news. At least for those of us who don’t want America to become Greece (any faster than already scheduled).

Republican and Democratic congressional aides have predicted for months that both sides will come together on a spending agreement to raise spending caps for the Pentagon as well as for nondefense domestic programs.

So let’s check our scorecard. On the tax side of the equation, we’ll hopefully still get some good policy, such as a lower corporate tax rate, but it probably will be accompanied by some gimmicky Keynesian policy.

On the spending side of the equation, it appears my fears about Trump may have been correct and he’s going to be a typical big-government Republican.

It’s possible, of course, that I’m being needlessly pessimistic and we’ll get the kinds of policies I fantasized about in early 2016. But I wouldn’t bet money on a positive outcome.

The annual Education Next gauge of public opinion on numerous education issues is out, and as always it offers lots to contemplate, including special questions this year on the “Trump effect.” I won’t hit everything, just what I see as the highlights.

School Choice

The poll’s headline grabber is a big drop in support for charter schools, public schools run by ostensibly private entities but subject to many public school controls, especially state standards and testing. When people with neutral opinions were removed, 52 percent of respondents approved of “formation” of charters—that word likely made some difference—down from a peak of 73 percent in 2012. With neutral answers included, only 39 percent of the general public supported charters.

The good news is that support for private school choice programs—superior to charters because they offer access to far wider options, including religious schools—saw upticks. Scholarship tax credits remain the choice champ, with support (absent neutral respondents) rising from 65 percent to 69 percent. With neutrals, support stood at 55 percent of the general public. For vouchers, a lot depends on question wording, but without a loaded emphasis on “government funds,” support (minus neutrals) stood at 55 percent, up from 50 percent the previous year. With neutrals, support was at 45 percent, with 37 percent opposing. Education savings accounts—basically, money parents can use not just for tuition, but other education expenses like tutoring or buying standalone courses—garnered only 37 support from the general public, but the concept is pretty new and people may just not have wrapped their heads around it yet.

Why the big drop in charter support but improved backing of private school choice? As always, wording, question order, and other artifacts of the poll itself matter, but assuming those aren’t the major causes of the results, perhaps the answer is that charters, as a compromise between empowering parents and maintaining government control, have traditionally tended to have the highest profile bipartisan support of the various choice mechanisms. As a result of Trump-driven polarization, perhaps they have also had the most visible schisms, maybe casting a more negative light on them. Or maybe people have started to perceive, as Education Secretary Betsy DeVos borrowed from Rick Hess to warn, charters are becoming “the Man” they were supposed to replace.


It is always instructive to see how much people think we spend on public schools, how much we actually spend, and whether the truth sets people free of the assumption that we need to spend more. As in the past, the poll finds that people greatly underestimate how much their districts spend per-pupil, with respondents guessing only about 69 percent of the actual expenditure. The average respondent thought their local district spent $8,877 per student, when in fact their district shelled out $12,899. When actual spending is not provided, 54 percent of respondents think their districts should increase outlays. When it is furnished, that drops to 39 percent. Of course, none of this actually says what the “right” amount of spending is, because we do not know what that is—we don’t have competitive markets that let us see how efficiently education can be provided.

Common Core

With the passage of the Every Student Succeeds Act, which at least in spirit puts states back in charge of their own standards and tests, and with many states likely loathe to quickly revisit the process of overhauling their systems, this once burning issue has dwindled to smoldering. Perhaps that is why after Core support cliff-dove between 2012 and 2016—with neutral responses removed, it plummeted from 90 percent to 50 percent—it caught a bit of an updraft this year, with support lifting to 52 percent. Add the neutral answers, though, and support sits at just 41 percent, and that’s with a loaded question that states the Core will be used “to hold public schools accountable for their performance.” (Who’s against “accountability”?) Much more disturbing, asked generically about uniform standards, federal tests, and letting parents opt their kids out of testing, the public overwhelmingly supports imposition. For educational freedom fans, perhaps the angel is in the details; once you move to real standards with concrete content, and real imposition, generic support turns to real world disapproval.

Agency Fees

From depressing news about imposition, we move to somewhat gratifying news. Agency fees are charges that teachers have to pay a union for bargaining on their behalf, whether they like that bargaining and union or not. These fees likely would have lost in the US Supreme Court in 2016 had it not been for the death of Justice Antonin Scalia. (A new case is on its way that might still strike down these enemies of free speech and association.) The public, it seems, might recognize the injustice of compelled support of hyper-politicized teacher unions, with 44 percent opposing agency fees, 19 percent neutral, and only 37 percent in support.

Affirmative Action

This became a hot topic a couple of weeks ago when it appeared that the Trump Justice Department would undertake investigations of colleges to see if they were discriminating against white applicants. The Trump effort appears now to be specifically connected to one suit by Asian Americans—not a broad hunt for discrimination against whites—but that could always change. Education Next only asked about minority preferences for hiring professors, not admitting students, and as I wrote recently I would encourage private colleges to undertake some affirmative action. That said, I still find the overwhelming result here—people want professors hired on merit, not goosed by considerations of race, gender, or political opinions—encouraging. It suggests that the public shares a crucial ideal: that people be treated as unique individuals, not members of a group.


As with many polls there is good news and bad news aplenty in this one, and lots to debate about question wording, order, and the why of the results. But the news is not bad for the best form of choice—private school choice—and some other important matters, so I’ll take it. Now go ahead and pore over the whole thing yourself!

Kenneth S. Rogoff stands out as the advocate of restricting hand-to-hand currency who has argued the case most comprehensively and probably the most cautiously. I critically reviewed his recent book, The Curse of Cash, in the July 2017 issue of Econ Journal Watch. Here I summarize highlights from my review, taking some passages verbatim. I encourage anyone who is interested to read the review in full. But in this piece I also comment on Rogoff’s response to my review that appeared in the same issue of EJW. When I provide page numbers for quotations, they reference Rogoff’s book; otherwise Rogoff quotations are from his response to my review.

A Case Against Cash

Rogoff does not propose eliminating all cash. In developed countries, he would phase out over a decade or more only large-denomination notes: in the United States, for instance, first $100 and $50 bills and then $20 bills and perhaps $10 bills. For small transactions, he would leave in circulation smaller-denomination notes, although he considers eventually replacing even these with “equivalent-denomination coins of substantial weight” to make them “burdensome to carry around and conceal large amounts” (p. 96). To put this in perspective, $1, $2, and $5 notes comprise less than 2 percent of the value of U.S. notes, or a little over 3 percent if we add in $10 bills. For less developed countries, Rogoff concedes that it is “far too soon” to “contemplate phasing out their own currencies” (p. 205).

Rogoff hedges his case with caveats and tries to address obvious objections. Yet like most proponents of phasing out cash, his argument is two pronged. Because cash is widely used in the underground economy, he believes that the elimination of large-denomination notes would help to significantly diminish such criminal activities as tax evasion, the drug trade, illegal immigration, money laundering, human trafficking, bribery of government officials, and even possibly terrorism. He contends that suppressing such activities would have the additional advantage of increasing tax revenue. The second prong relates to monetary policy. Rogoff believes that future macroeconomic stability requires that central banks have the ability to impose negative interest rates not only on bank reserves but on the public’s money holdings as well.

The Underground Economy

With respect to the underground economy, Rogoff’s Curse of Cash offers no genuine welfare analysis, considering the benefits as well as costs of the underground sector.[1] He gives little attention to the potential deadweight loss from forcing what is productive but unreported activity from a marginal tax rate of zero into marginal rates as high as 30 to 40 percent. Indeed, he disregards any gains to consumers of illegal drugs (except for an offhanded admission that legalization of marijuana may be a simpler approach for at least that part of the illegal drug trade) and exhibits scant concern for the welfare of illegal immigrants (despite his favoring increased legal immigration). The one transition cost that Rogoff tries to quantify is the impact on low-income individuals, recommending that the government provide at the very least about 80 million free, basic electronic-currency accounts, with a total bill of $32 billion per year.

When Rogoff gets to bona fide predatory acts within the underground economy, such as extortion, human trafficking, and violence associated with the drug trade, he descends primarily into lurid anecdotes. He fails to give even crude quantitative estimates to buttress his claim that eliminating cash would curtail these activities. As for corruption and bribery, Rogoff admits that they are really serious only in poorer countries—precisely where he also concedes that a premature elimination of cash would have dire economic consequences. In his discussion of terrorism, he admits that eliminating cash would have at best trivial impacts.

Approximately 50 percent of United States currency is held abroad. Yet Rogoff simply ignores negative effects on the nearly dozen countries that have completely dollarized (including Panama, Ecuador, El Salvador, several island countries in the Caribbean and Pacific) or another dozen partially dollarized (including Uruguay, Costa Rica, Honduras, Bermuda, the Bahamas, Iraq, Lebanon, Liberia, Cambodia, and Somalia). This is just another instance of Rogoff’s avoiding a complete cost-benefit analysis. As Pierre Lemieux, in his review of The Curse of Cash, succinctly puts it: “the economist venturing into normative matters would normally attach the same weight to a foreigner’s welfare as to a national’s.”[2]

Nor can Rogoff demonstrate any increased revenue for the U.S. government from phasing out large denomination notes. Relying on IRS estimates of the legally earned but unreported taxes in 2006 and extrapolating forward to 2015, he puts the potential gains to the national government at $50 billion annually (or less than 0.3 percent of GDP), along with approximately another $20 billion gain for state and local taxation. Yet his most comprehensive estimate of the seigniorage the government would lose from phasing out cash is $98 billion, or over 0.5 percent of GDP. Add to that the $32 billion annual cost of free electronic accounts for the poor, and Rogoff has failed to make a credible case that his proposal would create a net gain for the U.S. government, much less a net benefit for society overall.

True, other developed countries without a foreign demand for their currency have much lower rates of seigniorage than the U.S., and therefore government losses if those countries eliminated most cash would be less severe. The relative size of the underground economy in other countries, whether rich or poor, is also almost universally larger than in the United States. Indeed his cited estimates of the underground sector as a percentage of reported GDP for some of these countries—including developed countries such as Greece (25 percent), Italy (22.3 percent), Spain (19.6 percent), and Portugal (19.5 percent)—suggest that this is where a large fraction of their ordinary citizens live and survive. Rogoff states that the GDP “share of Europe’s shadow economy is more than double” (p. 63) that of the U.S., and he admits that this probably stems from higher tax levels and more burdensome regulation in Europe. But rather than reaching the obvious economic conclusion that the deadweight loss in Europe from inhibiting these activities would therefore be considerably larger than in the U.S., Rogoff merely touts “the benefits of phasing out paper currency” in Europe “in terms of higher tax revenues” (p. 89).

Negative Interest Rates

The second prong of Rogoff’s argument is that it would facilitate imposition of negative interest rates. The reasoning is as follows. When an economy sinks into a depression, the central bank should stimulate aggregate demand by lowering interest rates. But if interest rates are already extremely low, in what is alternately termed the ‘zero lower bound’ or a ‘liquidity trap,’ central banks are constrained in their ability to do this. Central banks can charge a negative interest rate on the reserves that commercial banks and other financial institutions hold as deposits at the central bank, and some are already doing so. If the monetary authorities push negative rates too far, however, the public can just flee into cash with its zero nominal return. Banks can also do the same by replacing their deposits at the central bank with vault cash. Elimination of cash would close off this way of avoiding negative rates, making negative rates truly comprehensive and effective.

The term ‘negative interest rates’ actually obscures somewhat the nature of what Rogoff contemplates. If negative rates can be extended to the general public, they in effect represent a direct tax on the public’s monetary balances, since most cash would be gone. Negative rates thus reverse the causal chain of traditional monetary theory, which focuses on the money stock. To the extent monetary expansion increases spending, it causes higher inflation with its implicit tax on money. Negative interest rates, in contrast, would explicitly tax money in order to cause increased spending with higher inflation.

I will not repeat here my extended discussion of why I believe that a policy of negative rates is not needed and would not work. Readers can find that in my review. Nor will I delve into one of the best parts of Rogoff’s book: his penetrating criticisms of fashionable alternatives for dealing with the zero lower bound, including a higher inflation target, forward guidance, fiscal policy, and dual-currency schemes. I simply point out that the problem vanishes once one thinks about monetary policy in terms of money rather than interest rates. Milton Friedman’s well-known thought experiment about a helicopter drop of money demonstrates this, as does Ben Bernanke’s writings on Japan’s experience with low interest rates, before he became chairman of the Federal Reserve.[3] If the economy needs monetary stimulation, the central bank through merely buying assets that genuinely increase the base of outside money can ultimately end up owning everything in the entire economy—except that sometime before it has done so, people will certainly start spending and drive up inflation.

Rogoff rejects this solution to the zero lower bound because he assumes it requires coordination with fiscal policy. But this assumption is simply wrong. Although we can imagine circumstances in which a desired monetary expansion would exceed the supply of Treasury securities available for open-market purchases, central banks can purchase and have purchased other financial assets or made other types of loans. The Fed has already purchased mortgage-backed securities, and other central banks have extended their acquisitions still more broadly, some even purchasing equities. Though far from ideal, such rare, limited, and temporary expansion of central-bank involvement in credit markets, if needed, would be less invasive than an untested, all-embracing money tax.[4]

The Public-Choice Dimension of a War on Cash

Not only are the positive arguments that Rogoff makes for confining currency to small denominations extremely weak, but his proposal also raises serious political-economy concerns that he hardly addresses and seems largely oblivious to. Even if the phasing out of all but small-denomination notes would accomplish what Rogoff has failed to convincingly substantiate, a marked reduction in crime, would it still be desirable? Not necessarily. Even when gains appear to be greater than losses, we should still hesitate about policies that punish or severely inconvenience the perfectly innocent. Lemieux has most trenchantly made this point:

Criminals are probably more likely than blameless citizens to invoke the Fifth Amendment against self-incrimination, or the Fourth Amendment against ‘unreasonable search and seizures.’ … But that is not a valid reason to abolish these constitutional rights.

These are necessary institutional constraints on State power, not just protecting the innocent but proscribing barriers that protect a free society more generally.

Indeed, one could argue that the underground economy is often a more effective check on government abuses than voting itself. Would alcohol prohibition in the U.S. have been repealed without widespread evasion by countless Americans? Would the U.S. be belatedly moving to marijuana legalization without the escape mechanism of the underground economy? Obviously none of these considerations excuse human trafficking and other forms of violence or brutality that are also within the underground economy. But one should be very cautious about drastic government impositions that indiscriminately impinge on almost the entire population, no matter how deplorable the outrages they are intended to diminish.

Rogoff’s faith in government is so strong that he evinces no discernible unease about possible abuse of his proposed interventions. Consider the battery of ancillary coercive regulations that he thinks might be vital to ensure the success of his proposal:

  • aggressive inducements to get people to turn in their cash (expiration date on large notes, restrictions on the maximum size of cash payments, and charges for very large deposits of small bills);
  • strict regulation of cryptocurrencies, such as Bitcoin and Etherium;
  • “more forceful steps…to pull the plug on money market funds” (p. 86);
  • lowering cash limits on anti-money-laundering regulations;
  • redoubling of “efforts to discourage” prepaid cards “as an alternative for moving large sums anonymously” (p. 97); and
  • banning “large-scale currency storage” or imposing “a tax on storage over a certain amount” (pp.160–161).

He goes on to predict that “[t]o the extent that new approaches to financial transactions are developed to evade government efforts to root out their sources, they will be met with a stiff hand” (p. 214). After all, “it is hard to stay on top of the government indefinitely in a game where the latter can keep adjusting the rules until it wins” (208). Rather than considering this government capacity a chilling concern, Rogoff enthusiastically embraces it.

Rogoff’s Response to My Review

Rogoff’s response to my review is quite respectful. He clearly wishes to encourage a civil dialogue on this question. Indeed, much of his response consists of amplifying details of his proposal. He does accuse me of “polemic exaggeration” because I titled my review “The War on Cash,” but that hardly seems unwarranted given that the title of his book is The Curse of Cash. More important, Rogoff’s response exhibits a shift in emphases in order to make his proposal appear still more tentative than in his book. Thus, he includes “many years of discussion and analysis” before any “advanced democracy is likely to start down the less cash-road.” And he pushes the “ultimate move to coins only (which I throw out as a very long-run idea …)” to “a time frame on the order of half a century or more.” He also shifts his geographical emphasis by conceding that

the case for pushing back on wholesale cash use is weaker for the United States than for most other countries, first because perhaps 40 to 50 percent of all U.S. dollars bills are held abroad, and second because the U.S. is a relatively high tax-compliance economy thanks to its reliance on income taxes for government revenue.

However these shifts introduce some additional tensions into Rogoff’s case. By admitting that phasing out cash is less of a priority for the U.S. than for other countries, especially those with high levels of tax evasion, he in essence is saying that his scheme is least needed where it is least onerous to implement and most needed where it is premature or dangerous to impose. After all, the most serious levels of tax evasion occur in less developed countries, such as Brazil and India. To be fair, Rogoff’s response still confines his focus to relatively advanced economies. He specifically mentions Greece and Italy, where he reports the underground economy as equaling about 25 percent of GDP. But phasing out large denominations in an economy in which unreported cash transactions lift the economy’s total output by as much as one-fourth strikes me as obviously drastic, even if the transition is slow.

By adding emphasis to how slowly he is willing to implement his proposal, Rogoff also undercuts the urgency he has attached to overcoming the zero lower bound, which in his book he characterized as having “essentially crippled monetary policy across the advanced world for much of the past 8 years” (p. 4). Indeed, if he is really willing to wait “at least a couple decades” for the phasing out of large denomination notes, why not just rely on market processes and technological innovations already in play to achieve a less coerced transition? Rogoff even predicts in his response that “the use of cash in the U.S. in legal tax-compliant transactions will be well under 5 percent ten years from now and probably only 1-2 percent twenty years from now, and that is assuming no change in government policy on cash [emphasis mine].”

Rogoff attempts to answer my charge that his welfare analysis fails to consider the economic benefits of productive underground activity by reiterating his speculation that “if the government is able to collect more revenue from tax evaders, it will … collect less taxes from everyone else” (p. 217). As he explained in his book, “if taxes can be avoided more easily in cash-intensive businesses, then too much investment will go to them, compared to other business that have higher pre-tax returns” (p. 59). This is of course correct as far as it goes. But it depends entirely on Rogoff’s expectation that any tax changes from eliminating large-denomination notes will be absolutely revenue neutral. This is a strong assumption seemingly at odds with the politics of taxation. Does he really believe that if phasing out cash brings in more tax revenue, governments are going to graciously reduce tax rates in order to ensure that the total bite out of the economy remains constant? This represents yet another instance of Rogoff assuming the perspective of a central planner and naively ignoring any public-choice considerations.

The simplistic assumption of revenue neutrality ignores a host of other complications as well. Recall that phasing out cash will reduce government seigniorage, so it is not clear how large the tax gains would be, if any at all, even for countries less reliant on that source of revenue. As for the Eurozone, The Curse of Cash (p. 84) cites estimate of seigniorage as high as for the U.S. Moreover, seigniorage, arising from an implicit tax on cash balances, already bears more heavily on underground, cash-intensive businesses. Phasing out cash not only changes both the level and type of taxation that these unreported, productive activities would pay, but also could subject them to burdensome regulation that imposes costs without generating revenue. This concern is particularly acute for countries like Greece and Italy. A genuine welfare analysis should carefully weigh all of these complications.

Regarding the 50 percent of U.S. currency held abroad, Rogoff repeats an assertion that he made in his book:

while there are many reasonable uses of the $100 bill abroad, it is indisputably popular with Russian oligarchs, Mexican drug lords, illegal arms dealers, Latin American rebels, corrupt officials, human traffickers, etc., and of course North Korean counterfeiters. In the book, I argue (conservatively) that foreign welfare should be thought of as a wash.

My review points out that Rogoff offers no quantitative evidence for this bold claim, and even if it were remotely close to accurate, it would still ignore the poor outside the United States who rely on dollars.

Given that we have only guesses based on anecdotes about alternative uses of dollars abroad, it certainly is appropriate for me to quote a contrasting view from a friend who read both my review and Rogoff’s response:

Based on my experience with overseas relatives $100 bills are also favored by ordinary citizens seeking a refuge from their own country’s unstable currency. They have no use for smaller bills, as they don’t use dollars for ordinary transactions. Dollars, for them, are a way of protecting their savings from the vagaries of the local currency. They aren’t familiar with all the denominations of US currency and would not be confident that smaller bills were genuine but they know what a $100 bill looks like and are comfortable with it.

Rogoff, however, remains willing to overlook welfare impacts on foreigners, whether they be potential illegal immigrants or overseas users of dollars. He declares:

The Federal Reserve and U.S. Treasury, not to mention Congressional decisionmakers, certainly do not directly take into account foreign welfare. The long-established approach to studying international trade and finance issues has always assumed that national authorities take into account national welfare, and that coordination and cooperation are needed to achieve a global social optimum. This is the right way to think about the problem, and my discussion is completely consistent with it.

Maybe for a politician pandering to voters but for an economist? This nationalistic bias is one of the critical flaws in Rogoff’s overall approach.


Rogoff raises many other interesting issues in his response, and trying to cover them all would make this article much too lengthy. His arguments are generally sophisticated and sometimes challenging, even when I disagree with him or believe he hasn’t adequately addressed my concerns. Our most fundamental difference remains our analysis of the State. Rogoff unreflectively adopts what Harold Demsetz characterizes as the “nirvana” approach to public policy. This makes him far more optimistic than is justified about the overall benevolence and competence of governments, particularly in developed countries. He thus oversells any advantages from his scheme and ignores or understates the myriad disadvantages. And it is he who bears the burden of proof for such an extensive reshaping of monetary systems, no matter how cautiously or slowly implemented.


[1] Larry White made this point with respect to advocates of eliminating cash generally in a previous Alt-M article.

[2] Pierre Lemieux, “Banning Cash: This Time is Not Different,” Regulation, 39 (Winter 2016-2017): 51

[3] Milton Friedman, “The Optimum Quantity of Money” in The Optimum Quantity of Money and Other Essays (Chicago: Aldine, 1970), pp. 1-67; Ben S. Bernanke, “Japanese Monetary Policy: A Case of Self-Induced Paralysis,” in Japan’s Financial Crisis and Its Parallels to U.S. Experience, ed. by Ryoichi Mikitani and Adam S. Posen, (Washington, D.C.: Institute for International Economics, 2000), pp. 149-66; and “Deflation: Making Sure ‘It’ Doesn’t Happen Here,” remarks before the National Economists Club, Washington, D.C., November 21, 2002; George Selgin anticipated Bernanke’s argument in a 1999 unpublished paper, “Japan: The Way Out,” reprinted at Alt-M.

[4] George Selgin has written on ways to make open-market operations more flexible and efficient using competitive auctions, see here and here.

[Cross-posted from Alt-M.org]