Alex Tabarrok, a libetarian economist who writes for Marginal Revolution, has a new paper out: Is regulation to blame for the decline in American entrepreneurship?  No, he finds.  I admire his intellectual honesty, but I question his definition of “economic dynamism.”

There’s a book, by Scott A. Shane, called The Illusions of Entrepreneurship.  Shane explains that there is a huge gap between what we imagine when we think of “entrepreneurship” and the reality of what the go-to statistics describe.  Things like the “rate of new startup formation” are dominated overwhelmingly by the churn of things like laundromats, taco trucks, and lawnmowing services, not by firms driving the adoption of new technology or economic growth.  Think of people you know who own small businesses – would any of them be excited to hear that the rate of entry and exit into their market has gone up?  And would you, as a consumer, be happier if the rate of laundromats in your neighborhood opening up and closing down were high or low?

Countries that do have high rates of entrepreneurship tend to be developing countries where many ambitious people are excluded from the formal job market – and often, small business in developed countries are run by immigrants or others whose access to domestic employment networks is limited.  These are not signs of economic health, so if we want to measure “economic dynamism” in the sense we imagine it – the adoption of new and better technologies and business techniques – we need different metrics.


A thought about depression and placebos.

According to two separate lines of research:

1) The placebo effect for antidepressants has gotten stronger over time.

2) The placebo effect does not exist for depression – the apparent placebo effect is actually just regression to the mean.

If both those things are true, it implies that people spontaneously recover from depression more quickly than they used to.  I can think of several explanations for that:

1) (Most likely, I think) We are more willing to diagnose depression than we used to be.  In the past, people wouldn’t get diagnosed unless their depression was long-lasting and intractable, but today, people often get diagnosed for relatiely short-lived depression.

2) Something about our society makes it so people recover from depression more quickly than they used to.  It could be as simple as “some of the people who enroll in antidepressant trials are also taking additional antidepressants that they don’t tell the researchers about.”

3) Something about the way we run clinical trials now is different than it was in the past, so the apparent trend is actually just a measurement error.

Lobster and grits with grilled green tomatoes

My girlfriend’s family just moved from New England to the South, so I tried “fusion” for Christmas brunch – a variation on the traditional shrimp and grits.  It’s not trivial to find lobster in upcountry South Carolina; we believe the enormous creature below is a “rock lobster”, also known as a “spiny lobster”, which has similar taste but slightly different appearance and texture.

This was my first time cooking grits, and they turned out a little clumpy, but the addition of Gruyère and gouda cheese kept anyone from complaining.  I have never been a big fan of fried, battered vegetables, so we grilled the green tomatoes, which turned out to be delicious.

Bellini cocktails are not Southern, per se, but peaches are, so they seemed like a reasonable substitute for brunch mimosas.


Rouladen with spaetzle

The first of two holiday meals I cooked for my girlfriend’s family: rouladen with spaetzle.  I’m an outspoken fan of the cheaper cuts of beef, which are often more flavorful than the more expensive cuts; thin-slicing and pounding is one of many ways to tenderize them.  Spaetzle can be made with a specialized tool, but fortunately we found a colander with large holes that made a fine substitute.


Parallel proceedings

Did you know that many jurisdictions in the United States already have “affirmative consent” standards for sexual assault? You could be forgiven for not knowing that, given that it’s almost never mentioned in the current controversies. And I don’t mean like the California law, that was passed recently and only applies to disciplinary hearings on college campuses; I mean long-standing laws where you can be fined or sent to jail if you touch someone sexually without their affirmative consent.

This should tell us two things:

1) Given that you probably have no idea whether you live in one of these jurisdictions or not, affirmative consent laws clearly do not lead to an atmosphere of fear surrounding sex, a sense that the government is digging too deeply into our sex lives, or what have you.

2) By the same token, affirmative consent laws don’t make it much easier to get convictions. These things really do come down to “he said, she said” – “he said” almost always describes behavior that does not meet the state’s definition of sexual assault, and “she said” almost always describes behavior that does meet the definition…because both parties have lawyers who generally want to win their cases and aren’t idiots.

Basically, anyone who says “affirmative consent shifts the burden of proof from the victim to the accuser” is wrong, whether or not they’re saying that’s a good thing or a bad thing. Burden of proof isn’t going anywhere; international human rights law says you can’t impose criminal penalties on someone unless you prove them guilty beyond a reasonable doubt. Let’s say we believe that only 2-10% of accusations are false…but reasonable doubt is often said to be roughly a 95% threshold; it’s still iffy whether you can clear that on a victim’s testimony alone, which is often all there is.

And so this leads to very few sexual assault convictions (actually, clearance and conviction rates for sexual assault are just as high as for other serious crimes; it’s the reporting rate that’s low, and I think we can safely assume that most of the unreported assaults are the ones that would be difficult to prove beyond reasonable doubt.)

So what can we do? I think we know, honestly. Why have activists focused so much energy on Title IX? Because it does what we need – it provides non-criminal penalties for sexual assault along with a lower burden of proof. Which would be great, except that it also replaces the institutional competence of the court system with university bureaucrats who have serious conflicts of interest and no idea what they’re doing, and it discriminates egregiously on social class by serving only college students, who are less likely to be sexually assaulted than women of the same age who are not in college.

There’s a solution, however imperfect. Remember Taylor Swift’s trial? Remember that she won, even though we couldn’t see absolutely-for-sure in that photograph whether the guy’s hand was on her ass? That’s because she sued him for civil damages. She sued him for only one dollar, which was fine from her point of view but probably not what most people would want; in general, the best bet would be “parallel proceedings” involving simultaneous criminal prosecution and civil damages. One fairly likely outcomes of such proceedings would be that the perp is found “not guilty” in the criminal trial but has to pay civil damages; his likelier-than-50% guilt is now a matter of public record. And though he’s not in prison, he may have nevertheless gotten the fear ‘a jesus put in ’em, much to the benefit of future, potential victims.

Parallel proceedings for sexual assault are possible now, and maybe even kind of common, for cases that are brought to court. But they’re not normative – when someone who has been sexually assaulted considers their options, they don’t immediately think of parallel proceedings as the way to go. What if that changed?

Extremely late-breaking thoughts on the Lewinsky scandal.

  1. Committing sexual harassment, under its legal definition, should be an impeachable offense. Conduct that falls just short of the legal definition by not being “severe” or “pervasive” enough should often be an impeachable offense, especially if there’s a pattern of it.
  2. Having an exploitative-but-consensual relationship with someone should definitely not be an impeachable offense, and I do not believe it’s possible to argue otherwise while taking consent seriously and without infantilizing forty-four-year-old Monica Lewinsky, who has been telling us a consistent story for twenty years. Sexual harassment is by definition “unwanted,” so anyone who says this was “textbook sexual harassment” is using the wrong textbook.
  3. Creating the appearance or risk of sexual harassment, by having a relationship that’s unethical and makes it difficult (at the time) for outsiders to know whether consent is genuine, should probably not be an impeachable offense; however this is a closer call, and I don’t think it’s illiberal or infantilizing to argue otherwise.
  4. The actual crime Bill Clinton was impeached for, and that no one seems to be talking about currently, was lying under oath in a sexual harassment case. That case was initially dismissed because the judge felt Clinton’s alleged behavior was not severe enough to count as sexual harassment, a decision that looks dubious in retrospect (at very least, there should have been no summary judgement.) That decision was appealed and the case was later settled. In retrospect, I think lying under oath in a sexual harassment case should be an impeachable offense, given that the behavior he was lying about – a pattern of having affairs with subordinates – was clearly relevant to the case.
  5. I don’t think most people should beat themselves too much over not realizing this at the time, because the credible allegations of sexual harassment and assault were mixed in with a deluge of blatantly false accusations and conspiracy theories about other things; it simply wasn’t possible, back then, for the average person to pull up a bunch of documents about the accusations on the newfangled interwebs and realize they credibly pointed to a pattern of abuse. I feel differently about journalists who covered the case, activists who made public statements, or politicians who voted on impeachment; they had a greater responsibility to weigh all the available evidence.

Against “The Toxoplasma of Rage.”

“The Toxoplasma of Rage” is regularly cited as one of Scott Alexander’s best essays, and I think it’s fundamentally mistaken, at least about the high-profile incidents it showcases.

He calls it “toxoplasma” because – like toxoplasmosis in rats and cats – there are supposedly two stages to how a story goes viral:

  1. Activists promote dubious stories because that demonstrates their zeal to their fellow activists.
  2. Counteractivists promote the same dubious story to demonstrate how dumb the original activists are.


I have a different theory: Continue reading