All posts by Eric Posner

What Legal Authority Does the Fed Need During a Financial Crisis?

A new paper, at SSRN:

The financial crisis of 2007-2008 exposed gaps in the law that authorizes federal agencies to provide emergency liquidity support. This essay describes the ways in which legal constraints hampered response to the crisis, proposes reforms that would make possible a unified and strong response, and criticizes the Dodd-Frank Act for weakening, rather than strengthening, the crisis-response agencies.

Cruz’s eligibility for the presidency and the Naturalization Act of 1790

The Naturalization Act provides:

And the children of citizens of the United States that may be born beyond Sea, or out of the limits of the United States, shall be considered as natural born Citizens: Provided, that the right of citizenship shall not descend to persons whose fathers have never been resident in the United States: Provided also, that no person heretofore proscribed by any States, shall be admitted a citizen as aforesaid, except by an Act of the Legislature of the State in which such person was proscribed.

Michael Ramsey argues that this language “strongly supports” the view that Cruz is a natural born citizen. Congress’ approach is “consistent with the eighteenth-century English parliament’s view that it could define natural born subject as it pleased (and indeed could tinker with the definition for policy reasons).” Or, as he puts it, a “natural born citizen” means “entitled to citizenship by whatever statute was in effect at the time of the person’s birth.”

But this is circular. Under Ramsey’s definition, the Naturalization Act provides that “the children of citizens of the United States [born on foreign soil] … shall be considered as entitled to citizenship by whatever statute was in effect at the time of the person’s birth”—namely, the Naturalization Act itself! The definition swallows its own tail.

The Naturalization Act uses a legal fiction as so many statutes do. Consider the Dictionary Act, which provides that the word “person” includes a corporation unless the context indicates otherwise. The Dictionary Act presupposes that the ordinary meaning of person is someone with a physical body and a beating heart. If the ordinary meaning of the word person encompassed corporation, then the Dictionary Act definition would not be necessary. People would read “person” to mean “corporation” without having to be told to. The Dictionary Act in this way is consistent with, and at some future point could be taken as evidence for, the proposition that the ordinary meaning of person excludes corporations. (By the way, over time courts came to refer to a non-corporate person as a “natural person,” again reinforcing the idea that we put natural before a term to refer to the regular or normal, as opposed to technical, usage.)

Similarly, the Naturalization Act assumes that “natural born citizen” does not include foreign-born children of Americans. That’s why it’s necessary to redefine this term to include foreign-born children. Enacted shortly after the Constitution was ratified, the law assumes that ordinary readers of the Constitution would have interpreted “natural born citizen” to mean only those born on American soil.

Is Cruz eligible for the presidency?

Michael Ramsey, in his excellent article on the meaning of “natural born citizen,” argues that the founders likely used that term with British precedent in mind. In a series of statutes, Parliament redefined the common-law definition of “natural born subject”—which meant born on UK territory, with some minor exceptions—to encompass people born on foreign territory to a British father, a necessary elaboration in an era when large numbers of British subjects spent years abroad roaming around in service of foreign-born British monarchs while spawning offspring. Blackstone discusses this history; the founders read Blackstone; ergo the founders likely had Blackstone in mind when they agreed on the term “natural born citizen,” replacing subject with citizen but preserving the meaning of “natural born.”

But did the population at large? Most people weren’t lawyers, fewer still read Blackstone, and in the current version of originalism—of which Cruz is an enthusiastic adherent—the relevant question is the “public meaning” of the Constitution when it was ratified. I checked Samuel Johnson’s dictionary, which does not contain an entry for “natural” but does include “nature.” The most pertinent of the definitions is “the regular course of things.” Citizenship, in the regular course of things, came to those who were born on U.S. territory. The complex process known as “naturalization”—by which an alien becomes a citizen after jumping through various hoops—reinforces this idea. People who are not citizens in the regular course of things (by birth on U.S. territory) can become citizens through operation of law.

The Constitution withholds the presidency from those who are “naturalized,” reserving the prize only for those whose citizenship is natural. The vast majority of Americans at the time who were not naturalized aliens were citizens of the various states because they were born in the states. Citizenship in “the regular course of things” would mean citizenship by virtue of birth on U.S. territory.

My argument would be strengthened if the practice in the states in the eighteenth century was to require people born abroad whose parents were U.S. (and/or state) citizens apply for naturalization. I am not aware of evidence in either direction.

What do the presidential candidates think are the limits on presidential power?

The major presidential candidates refused to answer Savage’s survey of their views about executive power. Why not? They did in 2011 and 2007. The survey was designed from the beginning to force candidates to acknowledge limits on their power if elected. Savage, like most commentators, has persuaded himself that Americans think that presidents have too much power. He made his name by documenting Bush-era executive-power abuses, and he has just published a book about Obama’s abuses as well. But it seems to be dawning on him that Americans want limits on executive power only when they don’t like the president in office. Today, Republicans complain about presidential power; eight years ago, Democrats complained. The opposition is tactical, skin-deep, which is why serious limitations on executive power—either self-imposed in response to public opinion, or imposed by the other branches—will not take place anytime soon. Americans want a strong president to solve their problems, and the candidates know this.

Bank fees and antitrust

bank ghhiIn Slate, I discuss a new paper by José Azar, Sahil Raina, and Martin Schmalz, which argues that the institutional investors that own banks in common do not want banks to compete over the interest rates they offer to, and fees they charge, depositors. The figure above (from their paper) shows that while bank competition looks fine under the traditional HHI index, their alternative measure–the GHHI, which takes account of common ownership–shows that concentration has increased over the last decade. Like their earlier paper on airline competition, this paper offers evidence of the monopolizing tendencies of capitalism, which offers a higher return to investors who are able to circumvent the incentives to compete.

A number of people have expressed skepticism about the mechanism. How exactly do institutional investors ensure that CEOs do not compete? There are many hypotheses. Their pay structure could reward them; subtle pressures can be brought to bear through communication; there are all kinds of interlocking relationships such as board memberships; and so on.

Below is the trend in bank fees (a similar trend exists for the threshold at which fees are charged). Correlation or causation? The paper uses an instrumental variable approach to claim causation. As always, showing causation is tricky. But more research is justified.

bank fees

Neil Irwin on The Big Short

Irwin praises the film while correctly pointing out that the movie implies that only its heroes foresaw the collapse of housing prices, when in fact many people did. But the movie makes a more serious error: It implies that the heroes foresaw the financial crisis as well as the popping of the bubble. They didn’t. The movie bumbles this important distinction, implying wrongly that the bad guys either also knew that they would cause a financial crisis but didn’t care because they would be bailed out, or were too stupid to anticipate the financial crisis. Both of these claims are wrong. No one anticipated the financial crisis so no one could have expected bailouts; and if no one anticipate the financial crisis, it is a misuse of the word “stupidity” to imply that it encompasses the failure to predict what no one predicted.

However, the distinction does play a small role in the movie. In 2007, the short bets were supposed to pay off. Burry, Lippman, and the others correctly saw that mortgage defaults would increase as housing prices leveled off and ARMs reset to higher interest rates; this should have pushed down the value of mortgage-related assets. But it didn’t, not initially, causing some much-needed drama that the film makers skilfully exploit. The film implies that the “marks” did not move because of fraud on the part of the banks that set them. The banks did not adjust the marks downward because if they did, the banks would have to declare themselves insolvent. They preferred to engage in fraud.

But here things get complex. Alone among the banks, Goldman does reduce its marks. So was Goldman a good guy? Not in the movie. The explanation is that Goldman was acting in its “self-interest,” having figured out what the heroes figured out and shorted the market–in fact, acting identically to them in all relevant respects except in expressing anguish. Everyone acted in his (perceived) self-interest. Someone was right and someone was wrong–who?

The problem was that no one knew how to value the assets. Because they had stopped trading, it was impossible to “mark them to market.” Accounting rules allow banks to depart from market  values in just such  circumstances. Otherwise, panic-driven fire-sale prices get transmitted onto the balance sheets of firms that could otherwise survive a liquidity crisis and imply that the banks are insolvent when they remain sound. However, in the absence of a market, no one really knows how to  mark the assets. It was this uncertainty about market values–not the housing bust itself, which everyone expected to be self-contained–that led to the financial crisis. Goldman was subsequently criticized for marking down values too aggressively, which could have exacerbated the crisis and forced healthy firms into bankruptcy, but enriched our friends, Burry et al.

Meanwhile, we know that our heroes did not anticipate the financial crisis because if they had, the last thing they would have done is to make deals with investment banks. A brilliant bet against housing is worth nothing at all if the counterparty is Lehman. Burry and others were supremely lucky that the federal government bailed out Bear and provided life support for the other investment banks. Their investors–depicted in the movie as dumb suits–had good reason to worry.

Stone on ISIS and freedom of speech

In the Huffington Post, Geof Stone criticizes my argument that Congress should consider a law that prohibits people from accessing ISIS websites and ISIS-related recruitment social media posts. Geof argues that history shows that the government overreacts to security threats, either succumbing to or exploiting public panic. He makes this argument in greater length in his fine book, Perilous Times. While I have read his book and others like it, I remain unconvinced, for the following reasons:

  1. It ‘s too easy when looking at historical examples to succumb to hindsight bias. While today it is obvious with the benefit of hindsight that the Red Scare prosecutions and deportations were wildly excessive, it was hardly obvious at the time. In our own memory, we have 9/11, and most of us remember very clearly that it was almost impossible to evaluate the enormous range of possible countermeasures that could have been implemented in the aftermath of the attacks.
  2. The basic logic of emergency response points toward what will seem excessive later on. The reason is that an inadequate response will cost lives, whereas an excessive response will result in infringements of civil liberties like speech, which–as Geof points out–are almost always temporary. There is a view among many civil libertarians that every time the government introduces a new security measure, civil liberties are irretrievably lost. The ratchet is invoked as a metaphor. Yet the First Amendment is far stronger today than it was ever in the past; today, it is a wrecking ball used not only against security measures, but against campaign finance rules, consumer protection laws, labor laws, anti-pornography laws, and many other rules that in the past were considered constitutionally sound.
  3. Geof’s best example of an overreaction is the internment of Japanese Americans, which government officials apparently believed was not necessary even at the time. Conceded. But I don’t think it’s fair to argue that Lincoln overreacted in the midst of a civil war, or even that the Alien and Sedition Acts were–at the time, 1798–unjustified given the standards of the time, the existence of a quasi-war, and the extreme political instability as well as vulnerability of the early Republic. It’s simply too hard to put ourselves back into that historical setting.
  4. It’s also too easy to blame “fear” or “panic.” When the government enhances security measures after an attack, critics almost always say that the government is “panicking” or exploiting public panic. The problem with this argument is that enhanced security measures are also a logical response to an attack. The mere fact that the government enhances security measures therefore does not tell us whether the response was panic-driven or reasonable.
  5. In fact, it is quite common for the government to consider and then reject harsh measures. A recent example is the proposal to ban all Syrian refugees. The House passed a bill to this effect but the bill was rejected by the Senate, which has instead supported a more reasonable-sounding enhanced inspection regime for visa applicants. So did the government panic or not?
  6. Finally, the government cannot just ignore public fears; it must address them. Critics who complain about “security theater” in airports just don’t understand that the government cannot wave a magic wand and make fear vanish; it must respond. When it doesn’t, people legitimately complain that the government doesn’t take their interests seriously.
  7. I don’t think we can do better than to consider the costs and benefits of a proposed security measure on the merits. As history recedes, it can provide as much misguidance as guidance.

The Big Short

The Big Short teaser poster.jpgI was wondering whether this film would really be able to explain collateralized debt obligations, cdo-squareds, credit default swaps, and all the rest, as the reviews suggested. When I explain these things to students, I need to write out diagrams showing cash flows, and then spend more time explaining why people created them, how they worked, and what went wrong. It takes a long time and is pretty boring. The movie’s approach was to put a woman in a bubble bath who spends 10 seconds explaining what one of these instruments was (I forget which one–the woman in the bubble bath was distracting), and to use a Jenga set as a prop for a CDO. Occasionally, celebrities were brought out to explain things. Anthony Bourdain says that a CDO-squared has something to do with reusing old Halibut in a fish stew. Even Dick Thaler, who makes an appearance, let me down. He says that the hot hand fallacy explains the financial crisis (it doesn’t–and apparently it has been debunked anyway). I suspect Thaler was happy enough to read the script, whatever it happened to say, so he could appear in a movie and meet Selena Gomez.

The movie’s creators faced a problem. How do you make a compelling story about the financial crisis for moviegoers who wouldn’t know a CDO from a cabbage. The brilliant solution was to pretend to explain what these concepts were rather than to explain them, so that the viewer would not be distracted from the glittering images and sharp dialogue by the nagging sense that he has no idea what is going on. The violation of the normal customs of movie making–including the breaking of the fourth wall by the actors–would enhance the viewers’ sense that what they were seeing was really true, and not only true but important enough to risk a collapse in verisimilitude which is normally maintained by keeping the fourth wall intact.

The relationship between the movie and reality is bewildering. The movie is based on Lewis’ nonfiction book, but only loosely–with most (but not all) of the characters’ names changed so as, I suspect, to allow the film makers to invent colorful details about them (various humanizing traumas, etc.). Lewis’ book itself, while apparently accurate in the details, gives a deeply misleading interpretation of the financial crisis that fits his crowd-pleasing template of eccentric outsiders versus complacent suits. If Lewis’ book was misleading, the movie is a CDO-squared version. Short it.

More on anti-propaganda laws

My piece in Slate advocating an anti-propaganda law garnered little enthusiasm. A few themes emerged. Some critics deny that ISIS poses a serious threat. That may well be true, as I acknowledged. One wouldn’t want to pass laws that counter a threat that does not exist. Others warned that it is easy to overreact, as I also acknowledged. However, my major concern was not ISIS but the new technology environment in which we live. The question, which no one seems to want to address, is whether social media and other forms of Internet-based communication create opportunities for radicalization that pose a threat to public safety. People have finally gotten around to admitting that the Internet poses a threat to privacy. I think in a few years the radicalization risks posed by the Internet will be too obvious to ignore. (Incidentally, law enforcement does not ignore the risks, nor do people who think about security for a living.)

Otherwise, the critics fall back on three generic arguments. One is that people who are sufficiently determined will be able to evade the anti-propaganda law that I proposed. The generic version of this argument is that no law that can be evaded is worth enacting. If the argument were accepted, then no laws would be enacted. While it’s true that some laws are ineffectual because they contain loopholes or rely on vague definitions, that’s hardly a reason for abandoning the effort altogether. It is in the nature of radicalization that at an early stage a person does not yet believe the doctrines to which he later commits himself. The purpose of an anti-propaganda law is to catch people at this early stage, before they find it worthwhile to take steps to conceal their identities. For that reason, evasion is not as serious a problem as it is in most other areas of the law.

The second is that it is simply impossible to draw a line between propaganda and legitimate political speech. That’s why I argued that the law should ban relatively concrete actions like linking to websites that are sponsored by or otherwise closely associated with ISIS. Line-drawing is a ubiquitous problem in legal regulation. The unavoidable arbitrariness of drawing lines has never been an argument against doing so, not even in the area of freedom of expression, where line-drawing is ubiquitous.

The third generic argument is that once one makes an exception to broad protections for freedom of speech, the camel’s nose is under the tent, we have stepped onto a slippery slope, etc. These clichés are as dry as dust and not even true. Courts have constructed countless exceptions to the First Amendment’s apparent unconditional protection for speech, including exceptions for defamation, child pornography, copying, fraud, and more—and yet none of these exceptions have expanded to swallow up the rule. Time, place, and manner restrictions could easily be expanded into a regime of effective censorship, but never are. To the contrary, protections for freedom of speech have never been stronger. If there is a slippery slope, it has gone in the other direction, with a conservative Supreme Court strengthening freedom of expression in areas like campaign finance and commercial activity in the last several decades, building on protections for political and artistic expression that were created by a liberal Supreme Court in the 1960s. If we could ban slippery slope arguments, the quality of public debate would be vastly greater.

Has Obama paved the way for President Trump’s anti-Muslim immigration order?

I was asked this question by a reporter, who wrote this story. Although the answer is “no” in any direct sense, the question reflects a low-level uneasiness I have noticed among Obama supporters who have praised (or rationalized) his aggressive executive unilateralism (especially as reflected in his immigration orders) but worry that it may set a precedent that future Republican presidents will take advantage of.

But it’s not clear whether presidents set precedents or, if they do, how it works, and when it is good or bad. Consider some possibilities.

  1. When a president engages in an action that is legally questionable, future presidents will find it easier to engage in the same action, but it must be the same action. For example, President Obama’s deferral order sets a precedent for just that—deferral orders in the immigration context. President Trump can do the same thing, but why would he?
  1. Future presidents will be able to engage in roughly comparable actions in different settings. President Trump will be able to defer action against corporate tax law violators, for example. But how would this work? If Trump said that he can defer action against corporate tax law violators because Obama deferred action against immigration law violators, would anyone really say, “well, then, okay”?
  1. Future presidents will be able to engage in executive actions that are unrelated. President Trump will be able to impose the death penalty on cop killers, as he has proposed, even though this is not deferral of prosecution or any other executive action that is supported by precedent, but an affirmative act that would violate deeply rooted constitutional norms. This is even less plausible than #2.
  1. Or—future presidents will not obtain any additional power whatsoever, or will even lose power. Some commentators claim that George Bush’s national security-related executive actions (surveillance, torture) provoked a backlash, making it more difficult rather than easier for future presidents to stretch the law in the area of national security. If a Republican president says he can defer action because Obama did, won’t people say—but you said that Obama acted illegally?

No one understands how unwritten constitutional norms develop. A common idea is that norms of reciprocity prevail throughout the political system. If Democratic Senators obstruct a Republican president’s appointments, then Republican Senators will obstruct appointments when a Democratic president comes to power—and vice versa. But are these norms of reciprocity powerful or weak?

Another theory is that it may be easier, as a political matter, to defend an action—like a military intervention without congressional support—if the president can point to a like action of an earlier president, especially an earlier president of the opposite party. But are these arguments really effective? Hard to know.

By the way, Democrats should consider one other possibility—that the next president (or the next several presidents) will be a Democrat. If that is the case, then Obama’s actions that expand executive power will benefit Democrats, not Republicans, in the near future. And a last thought: you might think that a Congress of either party or a divided party might cause mischief relative to even a Republican president. To evaluate Obama’s precedent-setting actions, you need to take all of these factors into account.

Legal realism and the driverless car

From Bloomberg View:

Turns out, though, their accident rates are twice as high [for driverless cars] as for regular cars, according to a study by the University of Michigan’s Transportation Research Institute in Ann Arbor, Michigan. Driverless vehicles have never been at fault, the study found: They’re usually hit from behind in slow-speed crashes by inattentive or aggressive humans unaccustomed to machine motorists that always follow the rules and proceed with caution.

Are supreme court justices loyal to the president who appoints them?

justice loyalty

Yes. Lee Epstein and I report statistical results in our new paper. A justice is more likely to vote for the government when the president who appointed him is in office, than when subsequent presidents are in office–even when those subsequent presidents are of the same party (and controlling for a bunch of other things).

The figure above suggests an interesting U-shaped time trend as well. The x-axis shows the initials of justices in chronological order. The y-axis shows a measure of preference for the government under the appointing president over subsequent governments. Note that the figure excludes Obama and Bush appointees because they have not yet served under a same-party president subsequent to the one who appointed them.

But President Trump couldn’t exclude Muslims by himself, could he?

Wouldn’t he need to ask Congress to pass a new statute that authorized him to block Muslims? Nope.

Immigration law gives the president more than enough authority to deny entry to classes of people who would otherwise be allowed into the country. From 8 U.S.C. 1182, the following aliens are inadmissable:

  • “An alien whose entry or proposed activities in the United States the Secretary of State has reasonable ground to believe would have potentially serious adverse foreign policy consequences for the United States is inadmissible.”
  • “Any alien who the Secretary of State, after consultation with the Attorney General, or the Attorney General, after consultation with the Secretary of State, determines has been associated with a terrorist organization and intends while in the United States to engage solely, principally, or incidentally in activities that could endanger the welfare, safety, or security of the United States is inadmissible.”
  • “Whenever the President finds that the entry of any aliens or of any class of aliens into the United States would be detrimental to the interests of the United States, he may by proclamation, and for such period as he shall deem necessary, suspend the entry of all aliens or any class of aliens as immigrants or nonimmigrants, or impose on the entry of aliens any restrictions he may deem to be appropriate.”

The provisions leave it to the president and his subordinates to determine who is a threat and who isn’t. But doesn’t the Constitution limit the president’s power to use these sections? The relevant case is Kleindienst v. Mandel from 1972, in which the Supreme Court ruled that the Attorney General acted lawfully by denying entry to a Marxist journalist because of his political beliefs. The Court said:

In summary, plenary congressional power to make policies and rules for exclusion of aliens has long been firmly established. In the case of an alien excludable under s 212(a)(28), Congress has delegated conditional exercise of this power to the Executive. We hold that when the Executive exercises this power negatively on the basis of a facially legitimate and bona fide reason, the courts will neither look behind the exercise of that discretion, nor test it by balancing its justification against the First Amendment interests of those who seek personal communication with the applicant. What First Amendment or other grounds may be available for attacking exercise of discretion for which no justification whatsoever is advanced is a question we neither address or decide in this case.

Courts and commentators have disagreed about the precise meaning of this language. And under some interpretations, Trump would lose. But his argument is not a bad one. President Trump could certainly argue that the “Muslims-are-a-threat” reason is “facially legitimate and bona fide.” Indeed, the post-9/11 sweeps of Muslim as well as Arab men relied on just such an assumption. That Trump singles out a religion rather than a set of political expressions could conceivably be a distinguishing factor, but just how much of a threat is a Marxist intellectual, after all?

The bottom line: if you (like me) don’t want Trump to block Muslims from entering the United States, then stop him from getting elected president. Don’t depend on the Constitution, Congress, or the courts.

Is an immigration ban on Muslims unconstitutional?

Probably not. The Supreme Court has held consistently, for more than a century, that constitutional protections that normally benefit Americans and people on American territory do not apply when Congress decides who to admit and who to exclude as immigrants or other entrants. This is called the plenary power doctrine. The Court has repeatedly turned away challenges to immigration statutes and executive actions on grounds that they discriminate on the basis of race, national origin, and political belief, and that they deprive foreign nationals of due process protections. While the Court has not ruled on religious discrimination, it has also never given the slightest indication that religion would be exempt from the general rule.

There is even precedent for Trump’s plan. In 1891, Congress passed a statute that made inadmissable people who practice polygamy (directed, at the time, at Mormons), and in 1907 extended this ban to people who “who admit their belief in the practice of polygamy.” While Congress later repealed the latter provision (the former seems to be still on the books), no court–as far I know–ruled it unconstitutional.

The plenary power doctrine is universally loathed by scholars and some have argued that it is effectively a dead letter. But any honest answer to a journalist’s question about whether Trump’s plan to ban Muslim immigration is unconstitutional should start with the plenary powers doctrine, and observe that it would be an uphill battle to persuade the Supreme Court to abandon a century of precedent. Unfortunately, that is not what scholars–who certainly know better–are telling journalists. They are likely being abetted by journalists and headline writers who don’t like the idea that Trump’s ban would be lawful. Not everything that is stupid or offensive is unconstitutional.

(N.B.: blocking American Muslims overseas from entering the country would certainly be unconstitutional, and blocking immigration by Muslims would raise complicated international-law questions.)

Walter Laqueur on Putinism

putinism

Putinism is apparently a thing. Why? And what thing is it? Putin is enormously popular among Russians because (1) they think he is responsible for the increase in their living standards when in fact they have benefited from the spike in oil prices; (2) he controls the media, which presents him as a savior; (3) he has reintroduced public order after the chaos of the transition; (4) he has appealed to national pride by making Russia a superpower again (or so they think: really a Potemkin-power); and (5) Russians don’t like democracy or liberalism but prefer the iron fist of an authoritarian ruler.

Both legitimate scholars (in the West) and ideologists, opportunists, cranks, and crackpots (in Russia) have tried to identify the distinctive features of Putinism. Laqueur, who sees a combination of luck and tactical shrewdness, is persuasively skeptical.

[N.B.: google hits on the y-axis in the figure above. “Maoism” was left off because sometimes google yielded >4 million hits and sometimes around 600,000.]

Are we at war with Russia?

Not yet. The NATO treaty provides:

The Parties agree that an armed attack against one or more of them in Europe or North America shall be considered an attack against them all and consequently they agree that, if such an armed attack occurs … will assist the Party or Parties so attacked by taking forthwith, individually and in concert with the other Parties, such action as it deems necessary, including the use of armed force, to restore and maintain the security of the North Atlantic area.

Article 5. But what is an “armed attack”? The protocol admitting Turkey to the pact says:

For the purpose of Article 5, an armed attack on one or more of the Parties is deemed to include an armed attack … [1] on the territory of Turkey, … [2] on the forces, vessels, or aircraft of any of the Parties, when in or over these territories…

But neither the protocol nor the treaty define “armed attack.” The Russian warplane did not not fire on Turkish forces but it did violate Turkish airspace and it did so despite many earlier protests by Turkey at similar incursions. The question is then whether the violation of airspace is an “armed attack.”

Like nearly every term in international law, the term “armed attack” has no agreed-upon definition. Not every use of force constitutes an armed attack. As the International Court of Justice has noted,  “As regards certain particular aspects of the principle in question, it will be necessary to distinguish the most grave forms of the use of force (those constituting an armed attack) from other less grave forms.” Yet it continues “Every State has the duty to refrain from the threat or use of force to violate the existing international boundaries of another State.” Isn’t sending a military jet a “use of force” that violates Turkey’s boundaries? On the other hand, the jet did not drop bombs on Turkey or try to, as far as we know. For this reason, I suspect that nearly everyone would regard the incursion as a violation of international law but not as an armed attack per se. Meanwhile, Turkey’s response would be deemed a lawful countermeasure and not itself an act of war.

So we are not at war with Russia, but what if Russia retaliates by bombing Turkey? Article 5 supplies wiggle room. Each of the other NATO members is required only to take “such action as it deems necessary” which may, but need not, include the use of armed force. That is the law. Politically, the story may be different.