Controlling the Uncontrollable

The title of Jim Manzi's new book, Uncontrolled, is a play on words. In its first sense, it's about the difficulty of controlling for confounding factors when you perform a statistical analysis. If you're comparing charter schools to public schools, for example, you need to control for things like income and prior test scores in order to make sure you're really looking at the effect of just the schools themselves, not being fooled by the fact that one might have had a better group of students to begin with. Unfortunately, as Manzi points out, no matter how many things you control for, it's almost impossible to know if you've really controlled for everything.

The problem is what Manzi calls "causal density." If you're studying the orbit of a planet, you can pretty much assume there's only one important cause of the planet's movement: gravity. Causal density is low. In medicine, there are more things to worry about, but a lot of problems are still tractable. Causal density is moderate. But in human affairs, there are lots of causes of everything, there are causes of the causes, and the causes often interact in complex ways. Causal density is very high, which means it's very hard to make sure you've accounted for everything. No matter how sophisticated your statistical tools are, it's always possible that something you haven't thought of is lurking in the background and throwing off your results.

Manzi's background is in business analysis, and he spent much of his career trying to use these tools to help businesses figure out how to improve their operations. Are you better off opening new stores in malls or on street corners? Will you pick up more new customers if you offer them a dollar off or if you offer them a 2-for-1 deal? Businesses, he writes, "have sunk vast resources into trying to develop useful, reliable predictions for behavior in the absence of experiments. In doing so, they have run into the same problems and hit the same dead ends. I know, because I spent years doing it."

So what's the solution? This is the second sense of uncontrolled. The answer, Manzi says, is allowing businesses to perform lots of random experiments. Go wild, because you never know what might work. But there's one catch: Although trial and error should be king, you should also insist that every experiment be performed rigorously enough that you can glean useful information from the results.

I'm enough of a geek that I found this the most arresting part of the book. Although Manzi's right about the difficulty of multivariable analysis in the social sciences, in the end I don't think it's quite as futile as he makes it out to be. Add to that the fact that the only studies he finds reliable all support conservative interventions (charter schools, broken-windows policing, work requirements for welfare), and you might feel like he's loading the dice a little too heavily in favor of his own ideological convictions. And this feeling only increases in the last chapter, where he recommends several policies that have nothing to do with random experimentation at all.

But put that aside. Manzi's a conservative, so it's only natural that he's put his thumb on the right side of the scale. I'd probably do the same if it were my book. Instead, focus on his most interesting idea: federalism with accountability. He wants the federal government to offer states more waivers to try out variations on social programs, but only as long as they conduct well-structured random field trials (RFTs) while they're doing it:

The federal government has certain specific roles under the waivers approach. It enforces whatever rules are in place at any given time. It guarantees consistent, reliable information: results of consistently executed RFTs when possible, and general statistical information when experiments are not feasible…The federal government would be metaphorically "making a market" in policy improvements.

There are plenty of things to argue about here, including the scope of the waivers and the rigorousness of the federal government's oversight. But it's unquestionably an interesting idea. Manzi claims that these kinds of experiments have become routine in the business world (a decade ago he founded a software company to help businesses do just that), and he recommends a goal of running 10,000 social policy RFTs per year. Why so many? Partly, he says, because the vast majority of experiments don't work, so you need lots of them to find any winners, and partly because you need to make rigorous experimentation ingrained in the culture of governing, and the only way to do that is to make it routine.

And if these experiments start showing that liberal ideas work, too, as they almost certainly will? I guess we'll cross that bridge when we come to it. In the meantime, it's a provocative proposal, and a reality-based challenge to Manzi's fellow conservatives.

The Problem With Supply Side Art-onomics

I was idling away a few minutes yesterday waiting for technical difficulties to be cleared up on a conference call, and by chance Michael O'Hare was on the same call, also idling away. "Did you read the Times forum about arts funding?" he asked while we waited. "Huh?" I responded brightly. Turns out the New York Times put up one of its "Room for Debate" roundtables a couple of days ago about how to fund the arts in America, and Mike was pretty unhappy about it. "It's all about supply side," he said, "and nothing about" — what? I never found out because at that moment our technical difficulties got cleared up and the conference call started.

But he threatened to blog about it, and today he did. The problem, he says, is that these conversations always end up being solely about high art, something that "deepens a division between art for the educated and rich and art for everyone":

Identfying “the arts” as what highbrow institutions offer also makes the whole conversation about the supply side, being nice to artists and arts institutions. Most arts funding gives money to arts presenters and, indirectly, offers art to art consumers. But if you give a concert that a dozen university music professors, a few critics, a charitable funder, and some composers and musicians absolutely kvell over, what have you really done for art, or for that matter for society, if nobody else comes to hear it?

What the arts most need is a demanding, competent, large audience, and supply-side programs aren’t very good for this; in fact, they are quite liable to capture by élites who use the arts to maintain their status. The research on this is long-standing and solid: the most important correlate of consumption of highbrow art [i.e., the demand side] is parental introduction to museums, theater, and concerts in childhood. Not much government can do about that, but the second is introduction to the arts in schools, especially hands-on learning, and the history of the last twenty years has been to trash this entire enterprise as a frill we “can’t afford”, along with physical education and sports for everyone.  Why the things that make life worth living — art and health — are frills or optional in a sane, rich society, and why Venezuela can afford a national network of youth orchestras and we can’t, are mystifying, but here we are.

Partly this is due to budget constraints, and partly it's due to our current national obsession with spending every hour of every school day drilling kids in the two or three subjects that will show up on high-stakes tests throughout the year. And it's a shame. I don't want to romanticize the past — the music teacher who came into my sixth-grade class once a week with an autoharp didn't do much to inspire a love of music in me, and that was back in the supposed golden age — but who knows? She might have inspired some of the other kids. And there were probably plenty of other music teachers who were a lot better than she was.

I don't agree with Mike about everything arts related, and I don't kid myself that we're ever going back to the age where we all sat around and enjoyed live performances by our friends and neighbors. That happened in the past not because we appreciated art more, but because good performances were simply too hard to come by. Still, art is fundamental to human culture, and our schools should do a better job of teaching our kids about it, even if it means taking half an hour away from filling in bubbles on a test sheet. And they should teach art in a language that kids understand. Learning to appreciate high art is fine, but that's a lot more likely to happen naturally if you learn something about art you actually enjoy first. If that means comic books and Britney Spears, fine.

In any case, click the link for much more. Mike has spent a big chunk of his career studying arts policy, and he has a lot to say about it. He even has something to say about paying \$120 million for an Edvard Munch drawing: "It properly exposes the whole culture of fine arts to ridicule as a game of poseurs, ignorant speculators, and predators that has nothing whatever to do with what paintings are about, or what art does for us, and that it should be a front page story as a serious event does a little bit to damage the quality of everyone’s engagement with art."

The Mysterious Case of Chen Guangcheng

So what really happened in the case of the Chinese dissident lawyer Chen Guangcheng? A week ago he walked into the American embassy in Beijing asking for asylum, and then, following a week of tense negotiations, he departed after the Chinese government promised he could return home and continue his education. But the deal apparently fell apart almost instantly. "Only hours later," says Paul Richter in the LA Times, "he began to complain that he had been pressured to remain in China." U.S. diplomats, Richter reports, were "stunned at the turn of events," while Bob Fu, a friend of Chen's, said the United States "has abandoned Mr. Chen." For his part, Chen now says his "fervent hope" is to leave China on Hillary Clinton's plane. So what happened? Did the Americans sell him out?

This just doesn't seem very likely, even if you think the worst of the Obama administration. As Robert Wright says:

The Obama folks may be cynical, but they're smart enough to have known that if Chen walked into a bait-and-switch, that would be a big problem not just for him but for them. It doesn't make sense, even in Machiavellian terms, that they'd have wanted to seriously mislead him.

That sounds right to me. Even if you look at this in the most cynical political way, Obama would have known perfectly well that he couldn't sell out Chen without the entire world knowing about it immediately. And that, of course, would open him up to exactly the kind of opportunistic jeering that Mitt Romney so drearily delivered on cue today ("This is a dark day for freedom and it’s a day of shame for the Obama administration").

But if that's not it, what did happen? Wright points us to some speculation from Walter Russell Mead

At some point one or more internal police officials either got to his wife or got to Chen after he'd left the embassy and told him in the most bloodcurdling and alarming way that he was under threat, that they would be watching and waiting, and that his wife and family would meet very unpleasant fates once the security forces got him back out of Beijing. And they would have told him in a very chilling way that he was not to tell anyone about this little conversation....After that kind of talk, a weary and blind man, much more worried about the safety of his family than about anything that would happen to him, might well change his mind about staying in China -- and might also need to give a good reason for the change of mind without mentioning any recent encounters with the security forces. This is a completely speculative theory with no evidence, and other explanations are possible. But it fits the known facts.

Maybe. In any case, it's the most plausible story I've heard so far. Stay tuned.

Donald Rumsfeld is Still an Insufferable Twit

You make the call: is Donald Rumsfeld the worst Secretary of Defense ever? It's a tough one, since there have been a few other famous losers too. But you know, at least Robert McNamara eventually came to his senses and later admitted that he had been wrong. That's something. But Rumsfeld just continues to be a jackass.

Digging Into the Pay Gap

Are women paid less than men? You betcha. Are they paid 77 cents for every dollar that men make? That's more contested, and Bob Somerby is pretty scathing about Rachel Maddow's insistence on sticking with that number even when her own guests tell her it's not quite right.

But this argument sort of misses the point. It's true that some of the gap goes away when you account for the fact that women tend to work in different jobs than men and take more time off to have children. But that's all part of the story. When all's said and done, women are punished financially in three different ways: because "women's jobs" have historically paid less than jobs dominated by men; because women are expected to take time off when they have children, which reduces their seniority; and because even when they're in the same job with the same amount of experience, they get paid less than men. All of these things are part of the pay gap. Whether you call all three of them "discrimination" is more a matter of taste than anything else.

And where is the pay gap most pronounced? That might surprise you. (Or might not.) The Institute for Women's Policy Research took a look at pay for the most common men's professions, and the biggest gap came in the very area that's supposedly the most meritocratic job title in the country: CEO. How about them apples?

Is Income Inequality Driven by Credit Booms?

In his new book — which I think I have a copy of but haven't read yet — economist James Galbraith takes a deep look at income inequality and concludes that it's been rising everywhere in the world. This suggests it's primarily been driven by macro changes in the global economy, not by specific changes in individual countries. Brad Plumer asks him what happened:

The current rise in U.S. economic inequality really only gets underway starting in about 1980. What changed then?

Between the end of World War II and 1980, economic growth in the United States is mostly an equalizing force, and job creation isn’t dependent on rising economic inequality. But after 1980, economic booms and rising inequality go hand in hand. So what’s going on? In 1980, we really went through a fundamental transformation. We stopped being a wage-led economy with a growing public sector that was providing new services. Programs like Medicare and Medicaid were major drivers of growth in the 1970s.

Instead, we became a credit-driven economy. What the evidence in the U.S. shows is that the rise in inequality is associated with credit booms, which are often periods of great prosperity. We had one in the late 1990s with information technology and one in the 2000s with housing, before everything fell apart. But this is also a sign of instability — the crash that follows is very ugly business. If we’re going to go forward with growth on a more sustainable basis, then controlling inequality and controlling instability are the same issue. One is an expression of the other.

....This view runs counter to what a lot of economists argue, which is that economic inequality has been driven by a “skills bias,” where there are growing returns to being well-educated.

Yes, I think the skills bias argument — the notion that inequality is being driven by technological change and education and the supply of skills — is comprehensively rebutted by the evidence. What you had was a concentration of income in very particular sectors. In the 1990s it was tech and in 2000s it was in real estate and the military, and those were closely related to the credit cycle. That’s what’s driving the data.

And this distinction has important implications for policy, for how one thinks about what should be done about this. If skills bias is the explanation, then that at most supports a case for more investment in education and training. It’s essentially saying that inequality is the result of technological change, and since technological change is wonderful, we should just accept it. But that’s not what the evidence shows.

I don't know if Galbraith is claiming that credit cycles are literally responsible for all of the increase in income inequality, but I certainly wouldn't be surprised if it were responsible for a very large part of it. The 1970s are an underappreciated decade in this regard. That was when the old Bretton Woods system broke down and we flailed our way into a new era of looser financial rules, the globalization of finance, and systemically low interest rates. There was no high-level conspiracy that drove the new system we eventually adopted, but by hook or by crook, pretty much every political and economic response to the 70s ended up benefiting Wall Street and the very rich.

For more, I recommend Greta Krippner's Capitalizing on Crisis: The Political Origins of the Rise of Finance. It's a very good introduction to how we got where we are today.

Local Government Idiocy Watch, Football Stadium Edition

In the great scheme of things this is trivial, but if you want a good example of the inertia and idiocy of local government, here you go:

The state appeared to be ready to put the brakes on the Los Angeles Memorial Coliseum Commission's proposed lease deal with USC on Wednesday, the same day the panel posted losses of more than \$7 million since 2009.

...."The state won't be rubber-stamping ... what the commission has negotiated or agreed to. We've identified some concerns," such as control of the property's revenue-rich parking lots, said Anna M. Caballero, Brown's secretary of the State and Consumer Services Agency and the governor's point person on the Coliseum issue....Later in the day, the Coliseum Commission released financial statements showing losses of \$2.4 million in fiscal 2009-10 and \$4.8 million the next year.

Yeah. The revenue-rich parking lots. Because obviously the state of California should be intimately involved in the lucrative parking lot business in South Central LA, which is apparently losing them several million dollars a year.

Watching this stuff from afar is just maddening. The Coliseum was built for the 1932 Olympics and is jointly run by the city, county and state. The NFL has made it crystal clear that no professional team will ever play there, which makes it completely useless to all three entities aside from the fact that it provides a cozy sinecure for members of the Coliseum Commission, who get free tickets to USC games and the occasional X Games event, as well as the warm glow of being important people. Except they don't even get free tickets anymore. They had to give them up after a series of LA Times articles exposed endemic corruption and mismanagement within the commission.

The answer to all this is obvious: just sell the damn thing to USC. They'd pay a fortune for it, they'd renovate it to within an inch of its life, and they'd turn it back into a jewel. That's because USC is really rich and has lots of alumni who really like football. And the city, county, and state would be out of the football stadium business, which they never needed to be in in the first place.

But that will never happen because — because shut up, that's why. Running a stadium is just too much fun and provides too many perks to think about giving it up. Idiots.

UPDATE: In comments, nominal says that Coliseum Commission members aren't paid salaries. That's true, and I've corrected the text to reflect this. Most of the recent scandal has involved Coliseum executives, not commission board members directly. As the LA Times says, "The groundwork for the scandal lay in years of clumsy stewardship, marked by inattentive commission members and a cozy relationship between [former GM Patrick Lynch] and his bosses."

Can We Please Stop Saying the Construction Sector is Bloated? It's Not

At the beginning of a long rant aimed at Raghuram Rajan, Karl Smith takes aim at Rajan's contention that "bloated finance, residential construction, and government sectors need to shrink." That may or may not be true of finance and government:

But there is no evidence that the residential construction sector is bloated. A cursory look at actual production levels would tell you that the United States is producing far too few homes. If you don’t trust that then you can look at prices. Rents — the price of the service flow from housing — are rising rapidly against a backdrop of depressed demand and high unemployment.

If you wanted to argue that a structural problem facing the United States was the inability of the residential construction sector to jumpstart itself, that would make sense. However, the idea that a problem facing the US is that it is spending too much capital and labor on residential construction is directly at odds with the facts.

There are lots of ways of looking at the construction market — and Karl has a couple of them here — but the easiest way is to simply look at the size of the housing boom of the aughts and the housing bust of 2008-12. That's the chart on the right, and its story is pretty obvious: the housing bust has been way bigger than the boom. For a year or two after 2008, you could plausibly argue that we had too much housing, which meant construction workers needed to find jobs elsewhere, but after that it just became a lazy mental tic that got repeated endlessly by people who hadn't taken a look at the actual numbers lately. If you do, it's obvious that we now need more housing, not less.

And yet, despite the fact that there's a lot of pent-up demand for housing, we haven't seen a real recovery in the housing market. That's a sign not of structural problems, but of the fact that people don't have enough money and bank credit is tight. That's the problem we need to address.

The Battle Over Your Endocrine System

The endocrine system is a collection of glands in the human body, including the thyroid, pituitary, and adrenal glands. Endocrine hormones regulate growth, blood pressure, serum glucose levels, and a whole host of other things. In other words: they're important! Naturally, then, we're doing our best to screw up our endocrine systems. Here's Nick Kristof:

Endocrine disruptors are everywhere. They’re in thermal receipts that come out of gas pumps and A.T.M.’s. They’re in canned foods, cosmetics, plastics and food packaging. Test your blood or urine, and you’ll surely find them there, as well as in human breast milk and in cord blood of newborn babies.

In this campaign year, we are bound to hear endless complaints about excessive government regulation. But here’s an area where scientists are increasingly critical of our government for its failure to tackle Big Chem and regulate endocrine disruptors adequately.

....Scientists have long known the tiniest variations in hormone levels influence fetal development....Now experts worry that endocrine disruptors have similar effects, acting as hormones and swamping the delicate balance for fetuses in particular. The latest initiative by scholars is a landmark 78-page analysis to be published next month in Endocrine Reviews, the leading publication in the field.

....Big Chem says all this is sensationalist science. So far, it has blocked strict regulation in the United States, even as Europe and Canada have adopted tighter controls on endocrine disruptors.

Yes, there are uncertainties. But the scientists who know endocrine disruptors best overwhelmingly are already taking steps to protect their families. John Peterson Myers, chief scientist at Environmental Health Sciences and a co-author of the new analysis, said that his family had stopped buying canned food.

“We don’t microwave in plastic,” he added. “We don’t use pesticides in our house. I refuse receipts whenever I can. My default request at the A.T.M., known to my bank, is ‘no receipt.’ I never ask for a receipt from a gas station.”

In the same way that we used to poison ourselves with lead, we're now poisoning ourselves with endocrine disruptors. Why? Because corporations who make lots of money from poisoning us say everything is OK. Just like they said lead was OK. And cigarettes. And benzidine. And chromium 6. And asbestos. And PCBs. And beryllium. And Vioxx. You'd think that even our venal political class would eventually learn not to listen to these guys, if only for the purely selfish reason of protecting their own children, but there's just too much money involved. So they go on scratching their chins and pretending that the "studies" produced by industry should be taken seriously. And we keep on poisoning ourselves.

For more, check out Judith Shulevitz's piece in the New Republic last year.

Barack Obama's Composite Girlfriends

A bizarre right-wing contretemps has blown up over Barack Obama's admission that some of the characters in his memoir, Dreams From My Father, are composites. Why bizarre? Because right in the introduction to the book, he says, "some of the characters that appear are composites."

But here's the really bizarre part. It's easy to see how you might have missed this. Or maybe not remember it. But these days, it's not like you have to reread the book cover to cover to see if Obama ever mentioned this. You just go to Amazon, click "Look Inside," and search for composite. That's it. I sort of assume that conservatives don't appreciate looking too much like idiots, so what explains not bothering with even this minimal level of scrutiny? Leave your best guess in comments.