Tuesday, March 1, 2022

Information Problems in Autocracy

In my business-government relations class yesterday, we were originally scheduled to discuss empirical papers on general information problems in government and bureaucracy. We instead discussed the Russian-Ukrainian situations as best we could.

One interesting result, coming out of intelligence work, is that Putin and others seem generally surprised at how difficult this invasion has been. Why might that be? 

One problem in general is that leaders like to surround themselves with "yes people". Leaders don't enjoy having thorns in their sides who challenge their every move or tell them they are wrong. (That's one reason Georgetown's provost is unlikely to ask me to be vice provost.) They prefer to have people who tell them good news and give them optimistic assessments of their plans. That's true even of nice people, like your kindly dean or friendly boss. But this means that in general people in power in bureaucracies will often get lower quality information than they need, or get information presented to them in a rosy way. 

Dictators in particular have a tendency to shoot the messenger and so probably get worse information. For instance, consider a story about Stalin. He orders a census. Because he had murdered and forcibly starved so many people, the real census numbers are low. The census officials know they can't tell Stalin the truths they lie and inflate the numbers. They didn't inflate them enough, so Stalin has them shot. The message is clear: Tell Stalin what he wants to hear. And thus throughout his dictatorship, Stalin is fed false information because his subordinates want to survive.

You might think a perfectly rational dictator would want perfect information so they can make optimal choices. However, dictators--and indeed bosses in general in hierarchies--have significant leeway to externalize the costs of their own ignorance and irrationality onto others. For instance, suppose Putin did not know that the West would respond as it has. Who is suffering the most right now? I doubt Putin is eating any worse, but the average Russian citizen is suffering a lot. Accordingly, while dictators need sufficiently good information and sufficiently rational processing of that information to stay in power, they do not have an incentive structure which leads them to prioritize perfect information or perfect rationality. They, too, are partially rationally ignorant and rationally irrational. 

Thursday, February 17, 2022

Is Fairness in the Rules?

 I’ve found that I’m more skeptical than average about whether the deliverances of political institutions carry some sort of moral authority.  It no longer surprises me that much when people disagree with me about this.  It still surprises me a little when people take the side of the rules of institutions in which they have no stake at all.

The latest case is about Russian Olympic skater Kamila Valieva, who tested positive for a banned drug but is still being allowed to compete this week.  Lots of people find this egregious.  NBC’s announcers were so incensed that they declined to comment on her short program.

The general feeling seems to be that it’s unfair for Valieva to skate, given her failed drug test.  Letting her compete upends the “level playing field,” thereby compromising “the integrity of the sport.”

I’ve been exercised before about how I think rules are overrated.  Here I want to make two more points.  First, rules aren’t always connected to fairness.  Second, fairness isn’t always the most important thing.

Consider five ways that rules and fairness can come apart:

(1) There may not be good reasons supporting rules.  Just because there are rules, it doesn’t mean they’re the right rules.  I don’t work for the IOC, so why do I care about what rules they happen to have?  Remember the runner who got banned from the Olympics for using marijuana?  That doesn’t make sense, right?  It’s not as though smoking marijuana improves one’s sprinting.  There’s no issue of fairness here.  Some journalists have made the entirely fair point that “if you cared about the ‘rules’ with Sha’Carri Richardson and weed, make sure you still care about them with Kamila Valieva and trimetazidine.”  I would just add that we can put the point the other way, too.  If you were against the rules then, I invite you to side against them now, as well. 

(2) Even if there are good reasons for rules, those reasons may not be based on considerations of fairness.  Sometimes it does make sense to have a rule, but it’s not because violating the rule would give a competitor an unfair advantage.  It might be that there are other reasons to constitute the game in some ways rather than others.  I’ve heard differing explanations about why the drug in question – trimetazidine – is banned.  Some experts seemed puzzled why one would be motivated to take it at all in pursuit of an advantage.

(3) Even if there are good, fairness-based reasons supporting rules, the underlying value may not be implicated by the case at hand.  Maybe it’s true that in most cases, taking the drug would confer an advantage by enhancing heart performance and increasing endurance.  But it seems like there is reason to doubt this holds for Valieva, a 15 year old and a high level athlete.   As an NPR explainer observes, “certainly for an elite athlete, this may not make much of a difference.”

(4) Even if there are good, fairness-based reasons supporting rules that are implicated in the case at hand, some penalties might still make things less rather than more fair.  The IOC’s take, which has really been getting worked over, strikes me as completely reasonable.  It might both be true that there is a reason for the rule, that the reason is based on fairness, and also that enforcing the rule by banning Valieva would be more unfair.  Set aside all this debate about how she’s a protected person and how best to protect her.  Suppose we are running a race and you step slightly out-of-bounds, shortening the distance between you and the finish line by – say – 3 inches.  That's unfair.  But suppose also that you beat me by twenty yards.  I guess I could tsk tsk you about the misstep, but that seems like poor sportsmanship.  My point here is that these issues will depend on local facts to the activity in question.  Admittedly, I know nothing about professional skating.  I’m just saying that we would need to know more about how the rules work – not just the facts about what the rules say.

(5) Fairness might just be outweighed by other considerations.  Remember when Michael Jordan, in the last seconds of his last game playing for Chicago, pushed off of Byron Russell to hit the game winning shot for a sixth championship?  In Utah, people still remember.  Anyway, my understanding of the basketball consensus on this point is that Jordan did in fact foul Russell, but also that no NBA referee would call the foul.  Why not?  Well, to have the greatest player in the world in the moment of his greatest achievement – and then to intervene to settle things with a rules violation – that doesn’t seem right.  In other words, it might not have been fair that Jordan got the call, but fairness wasn’t the most important value in the moment.  I’m not saying that morality doesn’t trump; rather, maybe morality isn’t really the issue in these cases.  Russell himself was philosophical in his reflections on the moment.  Why have somebody that works that hard against you?  Why not have him with you?” Russell asks.  “It doesn’t matter,” he says.


Monday, February 7, 2022

National Self-Defense and Just War Theory

Suppose a powerful nation unjustly invades its weaker neighbor. International law is clear: this is aggression, and the victim nation has the right to individual and collective self-defense. That is, the victim of aggression has the right to defend itself and to seek help from other nations. Article 51 of the UN Charter makes this right virtually unconditional and absolute: “Nothing in the present Charter shall impair the inherent right of individual or collective self-defense if an armed attack occurs against a Member of the United Nations.” 

Philosophers have in turn developed a body of rules to evaluate war: Just War Theory (JWT). JWT is richer and more complex than the bare rules of international law. It proposes a list of conditions that any war must satisfy to be justified. Writers vary in the formulation, but most of them would endorse the following list:


1) The belligerent must have a just cause.

2) The belligerent must abide by ius in bello (the laws of war).

3) The war must have reasonable chances of success.

4) The war must be proportionate, that is, the expected damage should not outweigh the realization of the just cause.


In our example, a defensive war by the victim of the aggression has a just cause, so it complies with condition 1. Let’s assume that it can also comply with ius in bello, condition 2.

But the defensive war cannot possibly comply with conditions 3 and 4. The defensive war has no reasonable chances of success. Can it be proportionate? No, if we compare the decision to fight with the decision to submit to the aggressor. Say that fighting will predictably cost 10000 innocent lives for naught (since the war is doomed to failure), while submitting to the aggressor will save those lives. Or worse: suppose the victim, exercising its right under article 51, asks its powerful allies to assist it in countering the aggression. This will predictably escalate the conflict, kill hundreds of thousands, and cause devastation of all sorts. Even if the allies' help will increase chances of victory, under JWT the defensive war will be likely disproportionate. 

It seems then that under Just War Theory the victim nation must submit to the aggressor. But that sounds bad, to say the least. We instinctively feel that the victim of the aggression has the right to resist no matter what, as international law says. Maybe the attacked nation reasonably expects to impose a heavy price on the aggressor. This, of course, is not enough to satisfy JWT’s success condition: imposing a heavy cost is not the same as winning. And choosing to bring about the deaths of 10000 innocents, or worse, triggering a regional conflict, when the attacked nation had the option to submit and avoid those dire consequences, cannot possibly satisfy proportionality.

Who is right? The UN Charter, which allows nations to resist aggression regardless of cost? Or Just War Theory, which indicates the opposite?

It seems hard to resist the logic of Just War Theory. If resisting leads to a catastrophe, perhaps submission is mandatory. On the other hand, perhaps sometimes fighting is the right thing to do, even against the odds (think of Churchill circa 1940).  And at any rate, the success and proportionality conditions in many cases favor aggressors.  This suggests that the theory of justified national self-defense must be revised. One way to start is to relax the success condition and add qualitative considerations to the calculus of proportionality.


Friday, January 21, 2022

School Choice as Occupational Choice

Many people have noted that the public school model is obviously unattractive when applied to other goods and services. For instance, few would support a shift to residentially-assigned, government-run grocery stores. It's much better to have a choice of grocery stores and to address issues of access with income supplementation.

But here’s another way of getting at the same idea. Think of attending school as a kind of occupation. Suppose that your place of work was dictated by the government based on where your house was located. Your boss might do a bad job. You might have serious conflicts with your co-workers. Your day-to-day work might be tedious and unproductive. But if that's the place of employment assigned to you based on your location, that's where you'll work.

Now suppose there's a much better workplace down the street. You'd prefer to work there. You're permitted to switch jobs, but there's a catch--your former employer will garnish your wages forever even though you no longer work there. You can't afford this loss of income, so you decide to stay.

I assume few would support a shift to this system of employment. Yet this is all too similar to our current system of public schooling. 

Bernie Sanders says that police departments are socialist institutions. Is he right?

Sophisticated socialists and anti-socialists rightly dunk on “socialism is when government does thing” takes. But sometimes these takes contain a grain of truth. For instance, socialists were embarrassed when Bernie Sanders called police departments socialist institutions and tried to explain why this is a mistake. But it’s not obvious to me that it is. Consider that policing is publicly provided and financed, centrally planned, subject to (indirect) democratic control, and accessible to all citizens (in principle). This strikes me as a lot closer to a socialist institution than a capitalist one by the standards of socialists themselves.

One objection alleges that socialism requires worker control and police departments don’t have that. Fair enough, but this objection deprives socialists of one of their favorite lines of defense—namely, pointing to benevolent public institutions as examples of socialism in action. For instance, Chris Maisano writes in Jacobin, “It’s one thing to identify public libraries with socialism. They operate according to democratic principles of access and distribution, providing services to all regardless of one’s ability to pay. They would be one of the most important institutions in any socialist society worthy of the name.” But public libraries don’t function as democratically-controlled, worker-run co-ops, so they shouldn’t count as socialist according to this objection.  

Another objection is that police departments don’t exemplify real socialism because they fail to operate in accordance with principles of egalitarian justice—for instance, they are often corrupt, self-serving, and benefit the few at the expense of the many. But this all-too-familiar reply is unconvincing for many reasons. I’ll just mention one: if we are judging an institutional arrangement in light of the outcomes we simply stipulate that it ought to promote, then socialist objections to capitalism instantly evaporate. The United State doesn’t exemplify real capitalism because employers often fail to treat their employees fairly, the rich obtain political favors, and so on.

Maisano writes, “If the forces responsible for killing Sandra Bland, Eric Garner, and Rekia Boyd exemplify socialism in action, then no person who wants freedom and justice should be a socialist.” Yet these forces do exemplify socialism in action more than Maisano wants to admit, which is why no person who wants freedom and justice should be a socialist.

Thursday, December 2, 2021

Gifts of the Magi

 “The Gift of the Magi” is an old O. Henry short story about a poor, young couple planning for Christmas.  Della cuts and sells her long hair in order to afford a gold chain for Jim’s heirloom pocket watch.  Meanwhile, Jim sells the watch in order to afford the fancy combs Della had wanted.

The story wants you think this is all great.  It ends this way:

I have told you the story of two children who were not wise.  Each sold the most valuable thing he owned in order to buy a gift for the other.  But let me speak a last word to the wise of these days: Of all who give gifts, these two were the most wise.

I’m not buying it.  I think we feel inspired by the story only when we view the characters as mere allegorical place holders.  Sure, we don’t care about Della’s precious hair or Jim’s old watch, so it seems nice enough to us that they ditched their respective vanities in a well-intentioned mix-up.  But we’re told that they really did care about those things.  If we believed that they really had given up their sole prized possessions, we would have to think of that as at least a medium-sized personal tragedy.  Worse still, neither gets anything out of the other’s heroism.  If we take Della and Jim seriously, this is just plain bad.

Despite its misguided moralizing, the story does offer a concept I’ve found helpful.  It describes a case in which people mutually give up something they care about for each other in the false belief it will provide value to the other, when in reality it won’t.  I now refer (not just in this post, but all the time, for anyone who will listen) to such cases as gift of the magi situations.  

Let me give you a couple of other examples.

Years ago, I helped a good friend move apartments.  I picked some doughnuts on the way in an effort to make the best of a necessary evil.  I worked all morning, but I’ve never been an especially adept mover.  My helpfulness was modest, but my surly antipathy for the whole business was extensive.  When I left her later that day, my friend said, “Thanks for your help, and doughnuts!” But then couldn’t stop from adding, in a moment of honesty, “Mostly the doughnuts.”

I live in Utah, where I enjoy taking people fly fishing.  But I’ve experienced confusions in which another person and I will accidentally end up thinking we're doing each other a favor.  So – in the past – I’ve found myself trying to teach someone who would rather not be there at all.  “What a hero I am,” I think, “for using this precious fishing time to help another person!”  Meanwhile, that person is thinking how much they’d rather be sitting on a lawn chair on the bank than getting badgered over how to mend the line. 

These are bad times!  In the first, I gave up something I cared about quite a bit in the misguided belief that I was making my friend’s life better.  With the clarity of hindsight, I think my friend felt correspondingly obligated to suffer my inept assistance, which was a net bad.  Because we are real persons rather than parable characters, it would have been much better for us to have both come clean about our individual interests.  Life is short!

The fly fishing case works the same way.  Politeness dictates we act as though we are benefiting from each other’s sacrifices.  In reality, fishing by myself would have been all upside for me, and not having to fish with me would have been all upside for the other person. 

How much of what we do for others turns out to be gift-of-the-magi style miscommunication?  My conjecture is: An awful lot.  Psychologists find that pathological altruism – when one person’s altruistic behavior harms rather than benefits its recipient – comes up all over the place.  In personal, familial, community, and even national settings, a lot of what passes for self-sacrifice turns out better described as “self-addiction,” wherein we misperceive the consequences of our putative well-doing out of an implicit effort to shore up our moral self-image. 

Grant me my conjecture for just a moment.  What might explain the pervasiveness of gift of the magi exchanges?  My guess is that it’s very tempting to think that if I am giving something up for someone else, they must be getting something out of it.  Even stronger: The thing they’re getting out of it must be comparable in magnitude to the thing I’m giving up.  Let’s call this the preservation of value hypothesis.

Consider the Tim McGraw song, “Just to See You Smile”.  It’s told from the perspective of a down-on-his-luck cowboy.  He spends his meager earnings on his beloved’s whims, quits his job to accommodate her career, and finally accedes to her breaking up with him and moving on with another guy.  The singer has no regrets:

                Just to see you smile

                I’d do anything, that you wanted me to

                When all is said and done

    I’d never count the cost, it’s worth all that’s lost

                Just to see you smile.

He thinks his sacrifices were worth it for the sake of her smile.  But why is she smiling?  I take it that the singer thinks all the stuff he’s doing must be good for her values.  After all, it’s bad for his, and nobody else is involved.  When I hear this song, I imagine how the story might go if told from the other person’s point of view.  “Mopey, self-indulgent cowboy can’t take a hint!”  He thinks he’s the hero of the story because he insists on sacrificing.  But his losses ensure transfer to her as gains only if the preservation of value hypothesis is correct.  The trouble is, it’s not.  You can do things that are both bad for you and bad for the other person.  With each successive tragedy, the speaker seems increasingly unable to brook this possibility.  No wonder she breaks up with him.

Here are my guesses.  At least for personal relationships among grown-ups, most of the things we do for others that are not net-valuable to us, by our own lights, are also not good for those for whom we sacrifice.  The problem with glorifying gift of the magi cases is they mislead us into thinking that we should be in the realm of losses for the other person to be in the realm of gains, when really the fact that we are in the realm of losses should tip us off in the other direction.  If what I’m doing is bad for me, it might not be great for those around me. 

O. Henry, then, was wrong twice over.  He thought the gift of the magi was both rare and wonderful.  I think gift of the magi cases are routine and terrible.  Which is happy news.  Our own values probably fit pretty well with the values of those around us.  Want to help someone?  Don't forget what’s good for you.

Monday, November 8, 2021

Why I'm Skeptical about the University of Austin

See the announcement here: https://bariweiss.substack.com/p/we-cant-wait-for-universities-to

Their expressed intentions include a strong commitment to liberal free speech and research policies and to increased rigor in the classroom. I hope they succeed.

But I am skeptical that they will, though--based on this statement--I am rooting for them. (I reserve the right to stop rooting if subsequent statements show bad intent or behavior.)

The reason is that the problems that plague universities seem to be built-in, and I'm not sure how this new university will overcome them. Let me explain the basics.

Every major organization faces a problem in that the goals of the organization as a whole and the goals of individual people within the organization may conflict. The organization might be dedicated to, say, furthering research, but the individual secretary in cubicle C just wants a job. The organization might be dedicated to finding the truth no matter where it lies, but the dean wants the angry mob to go away and the students who make up the angry mob want to ensure their truth and their feelings are protected. 

What's good for the organization and what's good for any individual within it can conflict. You need to align the utility functions of the individuals with the organization's goals. 

Indeed, it's worse than that because there are only individual people. The organization does not really have goals. The organization itself has no heart to break or dreams to realize. There is no Harvard per se that aims for truth, or social justice, or status. There are only individual people at Harvard that have their individual aims. The organization aims at any particular goal only in the reductive sense that many individuals might, in light of their shared values or incentives, be directed toward that goal. When we say Harvard aims for status and money, what we mean is that most people there aim for that. When we say Harvard aims for social justice, what we mean is that most people there pretend to aim for it. (I presume we all know no one sincerely means it.)

If this university gets off the ground, it will probably at first attract faculty and students committed to a certain ideology of the university. It's a good ideology, as far as I can tell! (Universities should seek truth and provide a rigorous, demanding, mind-expanding education. Most people disagree with this today, but they're wrong!) Because at first they will likely attract people who share this mission, there's a good chance that at first they will adhere to it, for the most part.

But then there will still be the perverse incentives. Individual professors push their brand over the university's brand. Individual students might want easy grades in their case, if not other cases. Individual staff members might want to squash just this one speaker they find obnoxious. Deans don't want to deal with bad press. Competition from other schools requires compromise. Getting good press means playing along with prevalent ideas. Etc. On the margins, individuals will face these temptations and some will succumb. 

However, the real issue is what happens in the long-term. The start-up tech firm is founded by three dorks with a vision and a passion, but 30 years later it's run by career manager types who are just as happy to sell this tech as they are to sell shoes. The people there don't have any passion for that initial vision. It's just a job.

So it goes with universities. Suppose the University of Austin becomes successful. It attracts an elite student body which then gets good jobs and grad school admissions. It attracts elite faculty. It has a high ranking. 

What will happen then is that people will go the University of Austin as students or faculty for the same reasons they try to go to other elite schools. They won't care so much about the original mission. They'll care about the status and financial rewards attendance brings. Just as today's Harvard students are mostly there for self-promotion, not veritas, so their future students will likely not be there for the mission. 

That's what I'd expect if they succeed and become a big deal.

The other alternative, I suspect, is to stay small, like St. John's College, which the incoming president used to lead. St. John's works, as far as it does, because it only admits a small number of weird students attracted to its mission. (N.B., I was one such weirdo. I applied to St. John's myself and was admitted, but I couldn't afford to go.) It's probably possible to maintain a tiny student body and tiny faculty that believes in that mission over the long term. 

At any rate, conservative critiques of the academy often read to me as follows: The USSR, China, Cambodia, and Cuba were humanitarian disasters because their Communist Party leaders have evil intent. What we need to do is ensure the new central planners have the right motives. But the real problem is the central planning and the incentive structure it creates. There is no long-term version of "making sure the right people run the system". The same goes mutatis mutandis for universities.