Thursday, March 26, 2015

We are living in a material world and I am a material girl?

By Robert H.

So we might live in a computer simulation.  The best advice on what this means for how to live your life comes from Robin Hanson, with one big exception: the dude wants you to be too interesting.  For example

If you can identify an especially interesting event around you, you might also try to prevent it from ending, as the simulation might end soon after the event does.
If our descendants prefer their simulations to be entertaining, all else equal, then you should want you and the events around you to be entertaining as well, all else equal. "All the world's a stage, and the people merely players." Of course what is regarded as entertaining does vary somewhat across time and cultures, and our distant descendants' tastes will likely vary from ours as well. So one should emphasize widely shared features of entertaining stories. Be funny, outrageous, violent, sexy, strange, pathetic, heroic, ... in a word "dramatic." Being a martyr might even be a good thing for you, if that makes your story so compelling that other descendants will also want to simulation you. 

I think Hanson pays insufficient consideration to his reader's starting conditions.  If you know famous people, if you are at a historical event, if you are a famous person, of if you are dramatic and entertaining, I think it's worth considering whether you are in a short term simulation designed to entertain or explore something interesting.  In that situation, yes, your goal should be to keep all that stuff going.  The moment it gets boring or the event passes you will be shut off.

But for most people most of the time, their lives are boring and they are boring.  That's certainly true of most people currently reading an essay by Robin Hanson.  And there are only three reasons to simulate that:

1. Best case scenario: it is a very deep and computationally heavy simulation.  Someone is investigating something like "what if humans had evolved in a galaxy without other intelligent life," and getting it right means simming millions or billions of insignificant lives from birth to death.  That means you live in something very like what you thought your world was, and can keep living very much like you have been.  And how you have been living life is boring.

2. Middle scenario: Someone is simulating an important event or person, but getting it right means simulating the back story.  You are going to be famous, or influence someone famous, or be somewhere famous, or etc.  The advice here is obvious: stay boring and try to keep those around you boring.  Avoid interesting things.  As soon as the famous or interesting event happens you will be shut down, so avoid it happening.  Note that this is the exact opposite of a lot of Hanson's advice.  He wants you to be interesting or around the interesting, I want you to be boring.

You  may want to take this a step further.  If your creators notice that you are intentionally avoiding what they want to simulate, they might write you off as a loss.  Practice self deception.  Aim to be the sort of person who promises themself that they will do big and exciting things -- tomorrow.  But mostly I think this is unnecessary.  If you are being simmed in this scenario you are probably either going to end up doing what your masters want you doing (in which case your goal is to do it later) or else your creators are very insensitive to wasting computational power (in which case you should be less worried about being shut-down). 

3. Worse case scenario: You are an extra.  Something or someone interesting is happening right now and your creators are taking the time to sim you while you happen to walk by or be near the important event.  You are going to be turned off in a few minutes.  You are doomed.

Add it all up and my main advice for people worried about living in a simulation is, assuming they are living relatively boring and unimportant lives, to remain boring and unimportant.  Avoid going near anything interesting and get your kicks from media consumed by the masses.

As an aside, there is a fourth scenario: your creators are running you for ineffable reasons, for no particular reason, or for reasons totally independent of your actions (maybe you *were* in an event they were interested in simulating but now they are just keeping you running because they think turning you off is unethical).  But those aren't very fun to think about.

Tuesday, March 17, 2015

Why Has the Low Income Nuclear Family Disapeared?

By Robert H.

David Putnam wrote a book about how the nuclear family no longer exists among low-income Americans.  Why has this happened?

It's about shifting norms and the breakdown of morals.  Or no, it is a socio-economic thing.  Or no, maybe it is a norms thing since the poor were always with us and were always poor, so why would their marriage patterns diverge unless social norms diverged?  Etc. etc.

This is really bad analysis, and I've been seeing a lot of it on the internet.  We're really asking two questions here, and both have superficially easy legal answers. Question 1, Why did poor people stop getting married as much, starting around the 60's?  Superficially easy answer: women got other ways to make a living, at least in part because of they made employment discrimination illegal and expanded welfare, so they stopped prioritizing marriage.  Further, since the 60's it has become easier and easier to have a child out of wedlock and still get the father to be forced to support it, as states have signed onto UIFSA , increased resources to the attorneys general, increased enforcement sanctions, etc., so that's another reason to not prefer marriage.  Question 2, Why did the poor start splitting up more?  Superficially easy answer: they made no fault divorce legal

Again, those answers are superficial and would not hold up under in depth analysis.  But they are still important.  They would be a factor.  We are, after all, talking about enormous changes in American family structure coterminous with or preceded by enormous changes in American family law.  Make the connection! 

But how can legal changes lead to big changes in low-income family structures but not high income family structures?  Doesn't the law affect everyone?  Well, use your imagination!  You could argue, for example, that lower income people, unlike higher income people, always had a preference for more diffuse family groups, then legal changes starting the 60's finally let them express it.  Norms and economic pressures didn't change so much as they were revealed.   Alternately, perhaps some of these legal changes affected poor people more than affluent people.  For example, no-property divorces can be orders of magnitude cheaper and easier than divorces between people with significant assets.  Perhaps the shift to no-fault divorce made divorce a cheap and readily available option to poor people, but for the rich divorce is still often a long, grinding, economically damaging prospect, so the advent of no-fault divorce had much less of an effect.  Perhaps the creation of streamlined processes for adjudicating support lowered the cost of having an out-of-wedlock baby for low income people, but not high income people who still prefer the old, more expensive, arguably-worth-it rout of going to district court.  Etc. Etc.

Again, I'm not wedded to the argument that legal changes changed society.  The arrows normally run both ways there.  But arguing about this stuff without even addressing the massive legal changes is crazy!

Thursday, February 26, 2015

Thinking Like a Lawyer: Wrong Triangles

By Robert H.

What are lawyers good for?  Well here is one thing: Economists are better than you at noticing economic trade offs.  Engineers are better than you at noticing design trade offs.  We lawyers may just be better than you at noticing the trade-offs inherent to rule making. 

This may be oversimplified, but I think the following is a good model for how lawyers think about rules.  Imagine you've got n triangle.  The vertices are labeled "simple" "clear" and "good outcomes in the case(s) at hand."  When contemplating a rule change, pretend the current rule puts you in the middle of the triangle.  The problem with making and updating legal rules is that any move towards one vertex is liable to move you away from another. 

For example, Alex Tabarrok just blogged about a decision in which the Supreme Court ruled that some unsupervised state regulatory boards aren't exempt from the anti-trust act.  In this case, the Court ruled that a state regulatory board made up of dentists can't try to create a monopoly in tooth whitening for their own industry.  Tabarrok, not a lawyer, thinks that Kennedy was balancing federalist concerns (ie, letting states have the authority to regulate without federal interference) with concerns about regulatory capture.

Nope!  Or at least, not entirely.  He was also balancing along the triangle, as Alito's dissent makes clear.  "Dentists can't use regulatory capture to make us all pay more for tooth whitening" is CLEARLY the right result (good outcomes), even if you love federalism.  The problem is that getting there creates either ambiguity (moves away from "clear") or complexity (moves away from "simple").  That's because the new rule -- regulatory bodies like this need state oversight to be exempt from anti-trust law -- begs the question of what counts as sufficient state oversight.  That question can either be left unanswered for district judges to work out on a case-by-case basis, which creates an ambiguity, or it can be worked out in exacting detail now or at a later date, which will create complexity.  Complexity and ambiguity are bad, but Kennedy thought it was worth it in this case to get good results.

Alito didn't.  He wanted a simple rule -- if a state says a regulatory body is backed by the state then that's good enough, and the regulator is exempt from anti-trust.  That would have moved us away from good outcomes (now dentists get to screw us) but towards "simple" and "clear" (we now know exactly what sort of regulatory bodies are exempt, and the rule for telling us that that is relatively simple). 

Obviously there is more to deciding a case than imagining this triangle, there was certainly more going on here than imagining a triangle, and some laws are so awful that you can improve them along all three of these dimensions.  But there is definitely a truth here lawyers are sensitive to and most people are not: the world is really complex, our moral judgments are really complex, and if you want real world legal outcomes to match your judgment of right and wrong you are going to get law that is either very complex (which has costs) or which puts a lot of the burden on individuals applying ambiguous directives to the facts at hand (which has costs).   The result is universally disappointing law: if the law gets an unfair result it is bad; if it is so impenetrable you need to hire an expert to explain it to you it is bad; and if the best your lawyer can say is "well, it could go either way, the judge gets a lot of wiggle room here," it is bad. 

Good laws are less bad; no laws are good.

Wednesday, February 11, 2015

Anti-War Movies Don't Have to be Anti-The-War-They-Are-About

by Robert H.

The non-controversy about whether American Snipper is pro-war trickles on.  Spoiler: it isn't, except in the sense that all war movies are pro-war.

But one argument for why it is pro-war strikes me as particularly bad: "You can't set an anti-war movie in Iraq and not show the horrors of the Iraq war," the argument goes.  "Since this movie glosses over civilian casualties, the bad reasons for going to war, etc., it is pro-war."

The problem with this is that most of the best anti-war movies throughout history have not been anti-the-war-they-depict movies.  They have been anti-war movies.  This makes sense.  A movie that criticizes war generally shouldn't focus on why one war particularly is bad.

Examples: MASH heavily sanitizes and lightens what it was like to work at a MASH, and totally ignores any atrocities or strategic blunders the UN forces committed more broadly.  The Deer Hunter isn't about the massacre at Mai Lai or operation rolling thunder or the invasion of Cambodia or the US's bad strategic decisions or etc.  Same with Full Metal Jacket.  All Quiet on the Western Front (the film) doesn't have much that is explicitly "this is why the war shouldn't have been fought and these are the atrocities committed by the Germans."  Etc.

No one has a problem, now, reading these movies as anti-war films that just happen to be set in certain eras, not as anti-war films with a moral responsibility to correctly show the horrors of that particular war in that particular era.  Only in partisan politics mode does reading the movies that way makes sense.  For example, when the Deer Hunter was released, lots of pro-peace or pro-communist types complained about how ridiculous the movie's presentation of the NVA was.  "Right, uh huh.  They all forced their prisoners to play Russian Roulette or face torture.  Sure.  Way to apologize for American imperialism, Hollywood."

Forty years later, no one gives a shit.  The movie is obviously a broad story about innocence lost in war, and it could just as well be set in the Sepoy Mutiny with vicious Indians (or British) bad guys, or set in Iraq with vicious American (or insurgent) bad guys, or etc.  Who cares?  The point of the movie is the toll war takes on combatants, not that the Vietnam War was justified because the North Vietnamese were monsters.

The same thing happened with Army of Shadows.  The movie is about a resistance fighter who works for and meets De Gaulle.  Unfortunately for it, it was released in the last months of the De Gaulle presidency, when he was an extremely controversial figure.  So the movie gets panned as Gaullist propaganda, something it clearly, to modern eyes is not.

All quiet on the Western Front was perceived as anti-German specifically, not just anti-war, and banned in Axis countries.

And now it's American Snipper's turn.

Don't fall for the same trick all those Fascist and Communist and French (but I repeat myself) film critics did!  American Snipper may only be a mediocre movie, but it's a mediocre anti-war movie.  The fact that it isn't an endless procession of Americans killing civilians and not finding WMD does not change that fact.

Tuesday, January 6, 2015

We All Need a Fifth

By Robert H.

I'm flipping through old The Big Question posts on legal issues, because I'm always interested in smart non-lawyers asking questions about the law, and I came across one I'm surprised none of the readers there were able to answer.

A guest blogger wanted to know why we have a fifth amendment.  Why a total right to remain silent?  Why not just safeguards that prevent coercive confessions, ie only allowing people to be questioned about their alleged crimes on the stand in a public trial where a judge can oversee the process and the public can oversee the judge? 

The main reasons given in the comments over there are 1. Because under outlandish hypotheticals this rule works better, and 2. Because the rule is overprotective in order to account for police and prosecutorial abuse.  Point 2 is important and probably explains the historical basis for the amendment.

But in the modern world, there are two other strong arguments for the fifth amendment: 1. A witness's behavior and mannerisms have very large effects on how finders of fact interpret the case.  Some people look trustworthy on the stand, some people look untrustworthy on the stand, and this doesn't necessarily correlate with who actually is trustworthy or untrustworthy.  This raises the problem of a defendant being forced to give exculpatory testimony and the jury saying "He seems like a liar, I bet he's lying about that testimony.  Why would he lie unless he has something to hide?  It's now more likely that he's guilty."  Worse, they might subconsciously think "That guy's testimony made me dislike him, so now I am more inclined to think he is a murderer because f*** him." 

Normally we just have to live with this when it comes to witnesses, but in the case of criminal law, where we have a policy preference against false convictions, we don't want people being found guilty of crimes because their mannerisms or lack of coaching make them look like a shifty witness.  So we don't make them give testimony.  This is a big reason why the defendant choosing to testify opens the door to the prosecution now being able to question him -- his mannerisms and character as a witness is already out there.  It's also a big reason why the fifth amendment is less robust in a civil setting -- the witness can't be convicted by anyone observing his mannerisms at that trial.

2.  Humans overrate, both in their conscious reasoning and in their subconscious impressions, the accuracy of human memory.  If you force someone to testify it is likely to be relatively easy to poke holes in their story, regardless of whether they are lying, simply because they misremember stuff.  This can make them look like a liar and leads to the same problems as above.  Again, we just live with the fact that lawyers can dismantle most stories from most witnesses in most cases, but here we are more worried about false positives.

Why do I think these are the big issues?  Because these are the two main issues criminal defense lawyers consider when deciding whether or not to put their witness on the stand.  It's 1. Can he tell a coherent story? 2. Will he come across as a sympathetic witness?  Defense attorneys do not ask how the testimony will look to dispassionate observers who ignore the witnesses mannerisms and account for the fact that he could be misremembering details through no fault of his own.  They also don't really worry about exposing their client to the judge forcing a confession at trial, even if that is the historical root of the right.

So, to be clear, here is my hypothetical where the lack of the rule would lead to a worse outcome:  A guy is innocent of murder.  He is called to the stand and forced to testify.  Over the seven hours he is questioned, he sweats profusely, has difficulty answering some questions, sometimes contradicts the evidence and himself, and frequently gets angry.

The defense alerts the jury to possible reasons for this -- he is nervous, this is a very emotional issue for him, human memory is faulty, etc.  Nevertheless, on a gut level the jurors find him untrustworthy. The facts that, in their mind, he 1. is a liar, 2. strikes them as an angry person, and . is lying about this particular incident, as if he has something to fear from the truth, pushes them over the edge to conviction. 

Monday, December 15, 2014

Level, Slope and Curve Factor Model - What's this all about?

John Cochrane recently blogged about my paper.  My co-blogger asked me to explain it a little more, and I realized that a lot is lost if you aren't deep into the finance jargon.  Additionally, what's the deeper significance of the whole endeavor.  Here is an attempt to explain my goal and interpretation in a broader context.

First, let me define a "factor." A factor is just meant to describe common movements across stocks. All technology stocks go up together when there is some good news about the tech industry. All energy stocks go down together on some bad energy news (maybe lower world energy demand). The goal is to stop describing a stock as "Ford" and start describing it as a combination of factors.

The factors we really care about aren't things like industries. We want to understand the "priced" risk factors. These are the factors that represent risk to investors. I can diversify my exposure to ups and downs in tech or industry or automotives by just buying lots stocks. So we don't expect any higher returns to holding tech stocks.

It turns out there are lots of things that predict stock returns. Those are characteristics of companies: small companies have higher returns than large companies, "cheap" companies have higher returns than expensive companies. Each of these are associated with factors. Small stocks go up and down together and to some degree move opposite large stocks. Cheap stocks go up and down and to some degree move opposite expensive stocks.

The literature now is postulating more and more possible risk factors. I am arguing a lot of those factors are manifestations of three main risk factors level, slope and curve. Here's what I do. I use all of these characteristics and I sort stocks into "high return" stocks and "low return" stocks. I sort into 25 portfolios. The 25th portfolio we expect to have really high returns. The 1st we expect to have really low returns. The factors describe how those portfolios move relative to each other.

Level describes that stocks go up and down together. This has been known for 50 years or more. On good economic news all stocks go up, on bad economic news all stocks go down. The most common model of risk, the CAPM, basically describes all stocks risk as how much they go up and down with the overall market.

The "slope factor" describes that high return stocks and low return stocks often move opposite of one another. (I think) It means that whatever risk high return stocks are exposed to changes over time. Sometimes investors need to be compensated a lot to hold risky high return stocks (with high expected returns), sometimes not so much.

The "curve factor" describes that sometimes high return stocks and low return stocks go up or down together, while medium return stocks don't even move. The curve factor is the hardest to interpret. It may be related to the volatility of the slope factor above, but that's kind of speculative.

So now I can describe every stock as some manifestation of level, slope and curve. Yesterday, I needed theories to describe hundreds of priced factors. Now I only need theories to describe THREE!!! That's pretty good. And since we've been explaining level for years, we really just have two. So, I don't know why high return stocks and low return stocks move opposite each other, maybe some risk factor, as in these companies will do badly in bad times, maybe these stocks are susceptible to fits of irrational exuberance. But if I can explain level, slope and curve, I can explain why some stocks have higher returns than others.

Thursday, November 20, 2014

Executive Action on Immigration

By Robert H.

Ezra Klein has an excellent and fair take on Obama's immigration reform.  I disagree with his conclusions, though.

Congress has told the President to export over 10 million people, but they've given him the resources to export less than 500,000 people a year.  So that's a problem.  Fortunately, they've also given him broad discretion to choose which sort of cases to go after.  Given all that, you have two choices for how to manage your resources if you are the President: create clear rules for whom you will and won't deport, announce those rules, and follow them.  Alternately, just wing whom you deport in a jumble of local discretion, happenstance, and random chance.

So which of these supports rule of law?  Which supports good governance?  To me it is obviously the former. Laws should be as transparent, clear, and predictable as possible.  If our officials are choosing to prosecute some deportation cases and not others -- and they are by necessity -- we should want to know what criteria they are using.  This move gets us closer to that.

So I am not worried that this clearly legal action will somehow undermine the rule of law in America.  It will do the opposite.  But isn't it violating political norms?  Maybe.  But if there is a norm in American politics that the president should deport people in a haphazard, chaotic, and unpredictable manner rather than using his legal authority to prioritize deportations, that norm is dumb.  Thank God a president is ignoring it.