Category Archives: MISC.

The bin Laden raid: where was the OLC?

The Times article describing the legal deliberations prior to the bin Laden raid mentions four lawyers: the CIA’s general counsel; the NSC’s legal adviser; the Joint Chiefs’ legal adviser; and the Pentagon general counsel. All of them, of course, did what executive-branch lawyers do: identify the most convenient legal categories needed for permitting the executive action.

More interesting, the OLC–which would normally be called upon to render the final opinion–was not included. Not just the OLC, but the entire Justice Department was frozen out. Why? Could it be that the OLC was less than cooperative when the White House sought a legal rubber stamp for the Libya intervention in 2011? Has the OLC been demoted for its insubordination?

Partisanship and the supreme court: summary

Agrees with co-partisans Disagrees with co-partisans
Agrees with counter-partisans Bridge-builders: Kennedy, Roberts Traitors: ∅
Disagrees with counter-partisans Loyalists: Kagan, Ginsburg, Breyer, Sotomayor, Scalia, Alito Mavericks: Thomas

Something like this. Perhaps, Justice Stevens could be classified as a traitor (to his party–he was appointed by Nixon Ford but in later years voted with the Democratic appointees–which is not the same thing as saying that he was a bad justice, of course). Among the loyalists, Scalia and Alito vote in a less consistently partisan manner than the other four.

The table is based on my previous posts: here, here, and here.

Partisan independence on the supreme court

partisan independence

Define partisan independence as the ratio of cross-party agreement to cross-party disagreement. Partisan independence is highest for those who both frequently disagree with co-partisans and frequently agree with counter-partisans. Kennedy and Roberts lead the pack. In the second group are Scalia and  Alito. Thomas now clusters with the Democrats. While Thomas often disagrees with co-partisans, he also often disagrees with counter-partisans–the two tendencies cancel out.

Cross-party agreement in the supreme court

cross-party supeme courtHere is another angle on the relationship between partisan identity and agreement on the supreme court. Three clusters appear. Kennedy and Roberts are most likely to vote with counter-partisans–in their case, the Democratic justices. Thomas (a cluster of one) is least likely to vote with counter-partisans. The remaining justices–the four Democrats, Scalia, and Alito–fall in the middle. As before the 2014 term seems to magnify trends that are perhaps (with the benefit of hindsight) partly discernible in earlier years. Data from Scotusblog.

Partisan agreement in the supreme court

within party agreement sup ctThis graph shows that Republican-appointed justices disagree among themselves more often than Democrat-appointed justices, who march almost in lockstep. What accounts for the striking acceleration of the trend this year? While some of the most polarizing cases have not yet been decided, I’d be surprised if the final cases make much difference in the pattern. Data source: Scotusblog.

Has the Supreme Court become less ideological?

Or, maybe I should have asked: Have the Republicans on the court become less ideological?

Justice Same-Party Agreement, % Opposite-Party  Agreement, % Propensity to Same-Party Agreement
Roberts 74.5 82.3 0.91
Kennedy 73.8 84.8 0.87
Scalia 76.5 74.3 1.03
Alito 77.3 75.3 1.03
Thomas 73.5 62 1.19
Breyer 92.7 82.5 1.12
Ginsburg 91.3 75.2 1.21
Sotomayor 91.3 74 1.23
Kagan 90.7 77.3 1.17

The numbers in the second and third columns are averages derived from Scotusblog (scroll down to the bottom). The last column is the second column divided by the third, so >1 means greater agreement with co-partisans.

Roberts and Kennedy are more likely to agree with the Democrats than with the other Republicans. I don’t think this has happened before. Last year, Roberts agreed with his co-partisans 88.8% of the time, and his opposite-partisans 81% of the time. For Kennedy, it was 86.8% and 82.25%. The Democrats and Thomas follow the ideologically polarized pattern of prior years.

Martin Schmalz: How passive funds prevent competition

Guest post by Martin C. Schmalz, University of Michigan.

Last week, Nelson Peltz’s hedge fund Trian lost a proxy fight at DuPont. The outcome of the battle received much attention, among others because the “passive” investors Vanguard, BlackRock, and State Street were instrumental in making his bid fail – they voted against him. Commentators have ex-post rationalized the failure of the campaign with gaps in some of Peltz’s arguments, his personality, and other factors. Curiously, nobody seems to have taken a look at how Peltz’s and the “passive” funds’ economic incentives differed. This note takes a first look at those, and comes to rather interesting conclusions.

According to Trian’s presentation filed with the SEC, the primary goal of Peltz’s campaign was to help DuPont achieve “best in class revenue growth.” He points out that DuPont’s performance in recent years was satisfactory only because of a positive industry-wide trend, but not if measured relative to DuPont’s competitors. Note for later that the first peer mentioned for comparison in the presentation is Monsanto. A second point of critique concerned DuPont’s lack of aggressive investment in R&D and other measures to gain market share. Third, Trian criticizes DuPont’s CEO for selling a large fraction of her shares in DuPont under the tenure of the index funds, a move that weakened her incentives to make DuPont perform well as an individual firm and strengthen the firm’s relative competitive position. Fourth, Peltz criticizes that DuPont willingly violated a Monsanto patent, then chose to pay $750m more than required in a settlement, and entered a licensing agreement with Monsanto until 2023, effectively pre-committing future cash flows to the competitor. Peltz also criticizes DuPont for “paying competitors” for such licenses more generally.

Peltz’s arguments make perfect sense according to the conceived wisdom reflected in corporate finance textbooks, which assume that all shareholders are undiversified: Peltz’s motion for an increased use of relative performance evaluation, for steeper CEO incentives and against wealth transfers to competitors at the expense of DuPont’s shareholders – textbooks would consider all of these measures value-enhancing improvements of DuPont’s corporate governance. Also, Institutional Shareholder Services (ISS), a proxy advisory firm, supported Peltz’s campaign.

As we know since last week, these arguments left DuPont’s largest shareholders – the diversified investors Vanguard, BlackRock, and State Street – unimpressed. They voted against Peltz, leading to a rejection of his bid, and a multi-billion dollar drop in DuPont’s stock price (indicating the market also thought DuPont would have been more valuable with Peltz on the board). So why did the mutual funds not share Trian’s goals?

To answer that question, it is instructive to see who DuPont’s competitors are. A quick browse gives a first indication why Peltz considers Monsanto to be DuPont’s primary competitor:  Monsanto and DuPont are the two firms dominating the seeds market, and Monsanto is DuPont’s next-largest competitor in the fertilizer and pesticides market.  Conversely, these two markets generate almost all of Monsanto’s revenue.[1]

schmalz tables 1 and 2

Peltz believes competing harder would increase DuPont’s value. For example, DuPont could decrease prices in the seeds market, and thus increase its market share. The stock market, as indicated by the stock price reaction amid the news of Peltz’s failed bid, appears to agree. The lower market prices for seeds would also lead to greater output and ultimately lower product prices for consumers – for short, greater economic efficiency. However, it would hurt Monsanto’s shareholders if DuPont were to compete more aggressively: DuPont’s increase in market share would come at the expense of Monsanto’s. So who are Monsanto’s shareholders?

It turns out that the same “passive” funds that helped reject Peltz’s bid at DuPont are – in the same order – also the dominant shareholders of Monsanto. In fact, with the exception of Peltz’s Trian Fund, the two firms’ top shareholders are almost identical.

schmalz table 3

The “passive” funds have no reason to object against cash transfers from DuPont to Monsanto – it’s just a transfer from one pocket to the other. Of course, Peltz (and everyone else who has a sufficiently steep interest in DuPont’s value) should object. That is the first source of imperfectly aligned incentives between the passive funds and Trian.

The more important insight, however, is that the common shareholders of the two firms would suffer from increased competition. Because prices would be lower, so would be the combined revenue and profits of DuPont and Monsanto. That outcome is in strict discord with the economic interests of Vanguard, BlackRock, and State Street. That is the second – and socially important – source of disagreement between the economic interests of Trian and the mighty mutual funds.

Here is one last nugget. Guess which company among DuPont’s competitors experienced a major change in stock price while DuPont’s price dropped in response to the news that Peltz’s bid failed. From market close on Tuesday to opening on Thursday, Monsanto’s shares gained 3.5%.

It appears that a dispassionate look at different shareholders’ economic incentives supplies a rather simple rationale for why the passive funds did not themselves enforce relative performance evaluation, protest the weakening of DuPont’s CEO’s incentives, encourage more R&D and gains in market share, and so forth.[2] Doing so simply isn’t in their economic interest. Peltz’s campaign, by contrast, aimed at increasing DuPont’s value in isolation, by strengthening DuPont’s relative competitive position. Predictably, the mutual funds voted against him.[3]

Before we conclude, pay attention to the dog that didn’t bark: Peltz’s failed campaign sends a strong signal to activists with similar goals as those Peltz tried to advance. If not before, then now they know: the combination of the index funds’ economic interests and voting power makes it unlikely that a campaign aimed at tougher competition will pass the ballot – so it might not be worth it to target a firm with these goals in mind in the first place.[4] That is how common ownership by “passive” funds can cause anti-competitive outcomes. If we want lower product prices, and higher output and efficiency, then taking a close look at the power and industrial organization of the asset management industry might be a good place to start.

[1] I haven’t gotten around to finding data on DuPont’s other product markets, which is why the following analysis is limited in scope and the external validity of its conclusions.

[2] Of course, this analysis does not prove that the passive funds voted against Peltz because he wanted DuPont to outperform the peers held by the passive funds, or because he criticized the voluntary wealth transfers to Monsanto and the lack of steep CEO incentives. Yet, whatever the reasons why the passive investors voted against Peltz, be it their economic incentives or other considerations, it is undisputable that their vote did prevent a campaign aimed at tougher competition. They thus caused less competition, compared to what it otherwise would have been.

[3] This outcome was indeed predicted. In this paper, my coauthors José Azar, Isabel Tecu, and I wrote: “owners generally need to push their firms to aggressively compete, because managers will otherwise enjoy a “quiet life” with little competition and high margins. Only shareholders with undiversified portfolios have an incentive to engage to that effect, while only large shareholders have the clout to do so. However, the largest shareholders of most firms tend to have diversified portfolios and therefore reduced incentives to push for more competition, whereas smaller undiversified investors don’t have the power to change firm policy without the support of their larger peers.”

[4] Opportunities abound for activist campaigns that didn’t or will never happen: very many U.S. firms are commonly owned by a similar set of diversified mutual funds as those owning DuPont and Monsanto. Here are a few examples.

Citizenship for sale

In Slate, I criticize the US citizenship-for-investment (effectively, citizenship-for-sale) program, while defending other countries that have dabbled with this approach.

I wasn’t able to find much research on the EB-5 visa program. It’s a good topic for someone to write about.

Why you should write for the New Rambler Review

The New Rambler Review takes its name from Samuel Johnson's The RamblerMany academics consider it a professional obligation to write reviews of new books in their field. The reviews are published in academic journals which are hidden behind firewalls, so that they can’t be read by the public. And because most academics don’t read journals in other disciplines, an interesting book with cross-disciplinary appeal can easily be overlooked.

This is a wasteful legacy from the dead-tree era, and the NRR aims to fix it. The NRR is accessible to everyone with a browser. More to the point, if you read a book and are curious what experts think about it, you can find an NRR review just by googling the book. Not only that, the review is likely to appear on the invaluable first page of the Google search results.

Take, for example, Michael Glennon’s new book, National Security and Double Government. This is a book that anyone might read. If you type “Glennon National Security Double Government” into Google, you’ll find the NRR review of it by Clifford Bob on the first page of the search results. You won’t find this review, which appeared in the academic journal Public Choice. Even if you do, you won’t be able to read it unless you belong to an academic institution with a site license or are willing to shell out $39.95 (!).

So if you see a book that you think the public should know about, would like to review it, and want people to be able to find and read your review, contact us.

Shaming is back!

Law professors of a certain age will remember that back in the 1990s, there was a debate about whether courts should impose shaming sanctions on offenders. This debate was caught up in larger discussions about the relationship between social norms, nonlegal sanctions of all sorts, and the law. But there was always an academic quality to this debate. It seemed at the time that shaming could not be an effective tool of social control in a huge, mobile, mostly anonymous society. So we all imagined shame and other social sanctions in confined settings: neighborhoods and communities, merchant and professional groups, and so on.

That was then. One Internet later, everything has changed. I discuss on Slate.

The Most Good You Can Do

In Slate, I discuss Peter Singer’s new book, The Most Good You Can Do. I like Singer’s utilitarian outlook, and I like the way he follows its logic into all kinds of dark corners, though I like less his attempt to prettify it in order to make it seem appealing to ordinary people.

Despite offering a surprising paean to capitalism (see p. 50 if you don’t believe me), Singer doesn’t take institutions very seriously, which I think is a problem in much of his writing (above all, in One World). Institutions coordinate people’s behavior for the common good; the sort of uncoordinated giving through philanthropic organizations for the benefit of impoverished foreigners won’t work, at least not at a large enough scale to make a difference, or that is at any rate the lesson I take from the foreign aid literature. Foreign countries have their own cultures, institutions, practices, and values. Agency costs exist in charitable organizations just like in for-profit organizations. All of these things spell trouble for “effective altruism” if understood to be committed to searching out those with the highest marginal utility per dollar. That said, by all means give your dollars to GiveDirectly or the other charities recommended by GiveWell if you want to maximize aggregate well-being, conditional on not too many other people doing the same.

There is a tension in the book between Singer’s relentless utilitarianism and human psychology. If you take Singer-the-philosopher seriously, then basically anything you do kills someone in the developing world. X number of ice cream cones means so much money less for malaria nets that will save the lives of children in Africa. Here is a philosopher who finally takes opportunity costs seriously! Singer simultaneously thinks that you should forgo the ice cream cones and somehow absolves people who don’t go this far, recognizing that the psychological burden of effective altruism is immense if taken to the extreme. Everyone has limits, he admits.

It’s clear why he does. Singer is afraid to scare off people who are willing to donate 10 or 20 percent of their income by telling them that they are not acting ethically unless they donate 80 or 90 percent of it. Singer’s style of utilitarianism may be philosophically impeccable, but it is a loser when it comes to motivating people. He tries to get around this by saying that the 10-percenter is more ethical than a purely selfish person, so one can take comfort in that. But people really want to know whether they are behaving ethically or not–yes or no–not where they fall on a scale, and Singer can’t answer that question to their satisfaction.

Debate about the right to be forgotten

You can watch a debate about the right to be forgotten between me and Paul Nemitz (pro) and Jonathan Zittrian and Andrew McLaughlin (con) here. Nemitz is a top EU privacy official with extraordinarily deep knowledge of privacy matters, while Zittrain is an internet law expert and McLaughlin is the CEO of Digg with extensive government and NGO experience.

I like to think our side landed some blows, but measured by audience reaction, our clock was thoroughly cleaned (is that the right expression?). Nemitz emphasized the political dangers of a world in which information about everyone is available on the Web, and hence available to the government, which can use it to monitor and control the public. I emphasized the personal costs in a world in which one’s identity is defined by search results that reflect a slip-up from decades ago.

I suspect that Nemitz’s argument made little headway with the New York audience because government repression based on surveillance is just not a part of historical memory in America, unlike in Europe. And my argument was probably too abstract (despite my uncharacteristic effort to pluck heartstrings). Although there are famous examples of people who lose jobs and suffer other harms because of some indiscretion that makes its way on the web, I think this worry seems remote to most people, at least so far, and there is a tendency to blame people for their indiscretions, however minor and whatever the consequences.

On the other side, McLaughlin and Zittrain warned of the dangers of censorship, and the risk that the right to be forgotten would be enforced in an arbitrary fashion. They also skilfully painted a dynamic and optimistic portrait of the Web as self-correcting; the harms that the right to be forgotten would address in blunderbuss fashion will eventually be addressed by the Web itself, as search engines and other institutions respond to public demand for more nuanced and fairer search results. Regulation at this point would short-circuit these developments.

The bottom line is that in America (unlike in Europe), even in the upper west side of New York, people trust corporations more than they trust the government.

King v. Burwell: my prediction

Okay, I’ll say it. I predict that the plaintiffs will win  by a vote of 5 to 4. Not that they should. Why?

1. While I agree with the government that Chevron deference is warranted, this doctrine is too squishy to constrain a majority of the Court. Academic research shows that Supreme Court justices don’t take Chevron very seriously.

2. And while I think the government’s interpretation is better than the plaintiffs’, I don’t think the plaintiffs’ interpretation is crazy. This sort of thing–where the relevant statutory language taken in isolation seems clear and is allowed to trump context even if context provides powerful evidence that the language was not intended–happens all the time in the courts. Whatever you think of textualism and how it should be done, the conservative justices won’t have to engage in embarrassing linguistic gymnastics to find for the plaintiffs.

3. And, finally, as I explain in Slate, justices who hate Obamacare and see it as an extension of the hated New Deal administrative state will, even if they try to be conscientious, find the plaintiff’s interpretation more persuasive. This is the well-known power of motivated reasoning. Not that I’m subject to motivated reasoning, or am I?

New Rambler Review

Announcing The New Rambler, an online review of books. Its mission is to publish high-quality reviews of intellectually ambitious books, in the spirit of The New York Review of Books, The Times Literary Supplement, and the back half of the (old) New Republic.

Our first few reviews are up. We’ll be adding new reviews every day or two for the next few weeks.

Check out the site, and let one of us know if there is a book that you’d like to review or see reviewed.

How Judge Hanen was able to rule against President Obama’s immigration program

He redefined non-enforcement of law (which is generally discretionary and non-reviewable) as the conferral of a benefit, namely, “three years of immunity from [the] law” and “legal presence status” (p. 87). While the president can underenforce statutes based on his constitutional authority, he can’t confer these “things,” these benefits, on unauthorized aliens without statutory authority.

But there are no such things. The beneficiaries of the program do not receive “immunity” in a legal sense because the president can change his mind and prosecute or deport them. “Legal presence status” similarly just means non-enforcement.

The grain of truth in Judge Hanen’s opinion is that “legal presence status” does typically trigger a right to a driver’s license under state law. That is a benefit. The government replies that states could refuse to issue driver’s licenses to aliens who are legally present under the program. The judge is (I think, rightly) skeptical of this argument. But he is rightly skeptical only because the Supreme Court has held that the president’s non-enforcement decisions in the area of immigration take precedence over state law. Which gets us back to where we started: legally speaking, the program is a non-enforcement program, unreviewable for that reason.

A Framework for Bailout Regulation

A new paper, written with my colleague Tony Casey. The abstract is below.

During the height of the financial crisis in 2008 and 2009, the government bailed out numerous corporations, including banks, investment banks, and automobile manufacturers. While the bailouts helped end the financial crisis, they were intensely controversial at the time, and were marred by the ad hoc, politicized quality of the government intervention. We examine the bailouts from the financial crisis as well as earlier bailouts to determine what policy considerations best justify them, and how they are best designed. The major considerations in bailing out and structuring the bailout of a firm are the macroeconomic impact of failure; the moral hazard effect of the bailout; the discriminatory effect of the bailout; and procedural fairness. Future bailouts should be guided by principles that ensure that the decisionmaker properly takes into account these factors.

King v. Burwell and ideological voting

Adrian Vermeule has argued that under the Chevron rule, the Supreme Court should defer to the government’s interpretation of the relevant portions of the ACA because the disagreement among lower court judges indicates that the statute must be ambiguous. Of the nine judges (six appellate, three district court) who have voted or ruled on the merits of the challenge, six agree with the government and three agree with the plaintiffs. If we believe that the judges acted in good faith, then their disagreement about the meaning of the statute is itself evidence that the statute is ambiguous, triggering Chevron deference to the government’s interpretation if it is reasonable.

But what if we don’t believe that they acted in good faith? The judicial behavior literature sometimes gives the impression that judges are algorithms through which presidents ensure ideological outcomes that they prefer. If judges just vote their ideology, then disagreement does not mean that the statute is ambiguous; it just means that the judges have different ideologies. (N.B.: the literature does not come to such an extreme conclusion, but it gives credence to the widespread view that ideological leanings of judges matter, particularly for high-profile, ideologically charged cases.)

So what do the data show? I compile the relevant information in the table below.

Judge Winner Nominating Pres. Court
Spencer Government Reagan E.D. Va
Friedman Government Clinton E.D.C.
White Plaintiffs Bush E.D.Ok
Griffith Plaintiffs Bush D.C. Cir.
Randolph Plaintiffs GHW Bush D.C. Cir.
Edwards Government Carter D.C. Cir.
Gregory Government Clinton/Bush 4th Cir.
Thacker Government Obama 4th Cir.
Davis Government Obama 4th Cir.


Surprise! All the judges voted consistently with their ideological priors, as measured by the president who nominated them–with one quasi-exception and one real exception.

The quasi-exception is Judge Gregory. Clinton appointed him during a recess, and then Bush nominated him. He voted in favor of the government. The other is Judge Spencer, who was nominated by Reagan and ruled in favor of the government.

One ought to consider the argument that, taking into account ideology, Judge Spencer’s ruling is the only one that is credible because his opinion was the only one that was clearly counter-ideological. Judge Gregory–as someone who was acceptable to presidents from both parties–might also be considered credible for just that reason.

On my version of the Vermeulean vote-counting approach, then, the government also wins. Of course, supporters of the ACA probably shouldn’t take much comfort in this analysis. If judges normally vote their ideology, then more likely than not the Supreme Court will ignore both my and Vermeule’s analysis and vote 5-4 in favor of the plaintiffs.