Sunday, June 24, 2012

off-beat data for off-beat research

Or maybe on-beat. A huge trove (50 years worth) of daily police bulletins turned up recently in LA and are going to be scanned and posted for the public. I could imagine some researchers having some fun digging through these.


Saturday, June 23, 2012

Always click the link

Andrew Gelman has a really weird post on modern art that somehow manages to tie in Warhol, Karl Popper and Al Qaeda. It is built around a bizarre op-ed by some art history professor but, though Gelman quotes her at length, he doesn't bother to rebut statements like "Art has also failed miserably at its secondary goal of protecting us from terrorism.


I don't know how Gelman could let something this get past him, particularly given his background as an art historian.


p.s. Before commenting on this or on Gelman's piece, give some thought to the title of this post.



Yglesias on the Apple Store

I wish that I was able to quote the entire by Matt Yglesias, but this portion makes the most important point of the article:
More precisely, the Times says the average hourly base pay is $11.91 an hour, which if you work 40 hours a week for 50 weeks a year comes out to the slightly lower figure of $23,800 per year. That's not very much. But it's a lot more than many people make right now. Currently 15 percent of the population is living below the Federal Poverty Line including a shocking 22 percent of American children. And yet a single mom raising three kids working full time at the Apple Store at an average wage would be above the Federal Poverty Line.
That should tell us something about how dire the conditions facing poor Americans are. But, again, those are the circumstances in which 15 percent of the population and 22 percent of kids find themselves.
And yet at the Apple Store workers get "very good benefits for a retailer, including health care, 401(k) contributions and the chance to buy company stock, as well as Apple products, at a discount." So think about a world in which these kind of jobs were the absolute worst jobs around. You're thinking about a world in which everyone has health insurance, and essentially no full-time workers or children of full-time workers are living in poverty. That would hardly be a world with no problems, but it would be a tremendous achievement. And it seems to me to point to the fact that the really urgent question isn't why aren't Apple Store jobs better, but why are so many jobs worse than this? Why can't we live in that world where people who work hard and play by the rules aren't poor?

One thing that this article really does is answer the question of why do I care so much about inequality.  I write a lot about topics like CEO pay; the truth is that I would stop caring about these issues in a world in which this really was the case.  My issues with wealth redistribution are based on the dire poverty which people find themselves rather than class envy.  If people who worked hard and played by the rules could obtain these types of jobs with ease then I would care a lot less about the wealth of the upper classes.

The other thing that I like about this is that it represents a constructive social vision as to what we could do to actually improve the lives of working class Americans.  It may or may not be the best possible idea, but it at least gives a specific target to aim for.  And that is worth a lot.


Great post by Kaiser Fung

Okay, sometimes you miss a great article and feel really stupid when you do find it.  In this case the author is Kaiser Fung and he provides a fantastic example of how modeling decisions can completely undermine a research article's conclusions.  In this case the authors manage to completely reverse conventional wisdom with their analysis.

The only thing I disagreed with Kaiser on was using income as a proxy for the value of one additional year of life.  I worry that could end up in bad places if we were not careful.

Definitely worth a complete read.

Burn Notice is back

And, unlike Justified, you still watch this one for free.

 

Friday, June 22, 2012

The magic of markets

Paul Krugman has a very interesting piece about the privatization of prisons:
But if you think about it even for a minute, you realize that the one thing the companies that make up the prison-industrial complex — companies like Community Education or the private-prison giant Corrections Corporation of America — are definitely not doing is competing in a free market. They are, instead, living off government contracts. There isn't any market here, and there is, therefore, no reason to expect any magical gains in efficiency.
The whole piece is worth pondering.

Google has an interactive Turing tribute

You really ought to check it out.

Class stratification

Marketplace has a good story on (increasingly rare) marriages across class divides. The whole thing is worth a listen but this part in particular caught my ear.
That doesn’t come as a surprise to Susan Fiske, a professor of psychology and public policy at Princeton. She looks at how social class shapes our daily interactions. “It’s rude to talk about somebody being more privileged than somebody else,” Fiske says. “Social class is such a taboo topic in American society.”  
Fiske’s research sounds like a warning label for couples like Rider and Heavener. Consider the experiment where Fiske put volunteers in MRIs, and showed them pictures of people from different classes. When they got to the photos of really poor people, the part of the brain that normally lights up when we see another human being just didn’t light up. “What we find is that people say it’s hard to think about this person as a three-dimensional human being,” Fiske says.
I've been becoming more and more convinced that class identification has been growing more powerful for the past four decades (with the caveat that the identification is asymmetric; it extends up but not down). The experience of the Great depression and the war had, I suspect, greatly reduced group identification in economic classes. The war had entailed remarkable shared sacrifice and collective action while the Depression had deeply impressed on most Americans the possibility that they too could be among destitute someday. When LBJ was pushing through the Great Society, all of the electorate were born before 1945 (the 26th amendment was still years away)

People.born after the war had almost the opposite experience. They grew up in a period of great economic security and growth. The notion of "there but for the grace of God go I" was almost entirely foreign to them. Under these circumstances it would be difficult to imagine empathy for the poor not dropping.

What amazes me though is how far we've gone and how completely we've internalized these somewhat Randian class attitudes. Mass incarceration is seen as a way of maintaining social order. The same journalists who rail against overpaid teachers will be genuinely moved by the hardships of people scraping by on 300K. A political party in an election year loudly complains that the majority of Americans should pay more taxes so we can make the system less progressive.

Of course, all these examples involve a fairly rarefied group of leaders and opinion makers. Perhaps there's some hope for the rest of us.

p.s. Should have worked in a link to this previous post on the subject.

Improving work-life balance

Dana Goldstein:
1. Universal early childcare and pre-K
2. Extended learning time at school, not just for more test-prep, but for art, music, sports, and other enrichment and supervision affluent kids get as a matter of course. This would help the school day better conform to parents' work day, which helps women (and men) work and parent (Slaughter also points this out)
3. A higher minimum wage and workplace representation in the service sector (especially helpful for single moms)
4. Paid family leave
5. More enlightened men - men who do chores! According to the Bureau of Labor Statistics' latest time-use study, men still do only about 25 percent of housework, 29 percent of food preparation and clean-up, and 33 percent of childcare.*
 I found this list interesting because it seemed to really focus on three elements.  One is a stronger social saftey net (points 1 and 2).  Two is less income inequality ( point 3, and to some extent 4).  Three is the culture of the people involved (point 5), which seems to be a hard point to attack as a matter of government policy. 

The question that I have is what is the relative priority of each item on this list?  What are the trade-offs?  And what else could you do to reach these goals indirectly?  For example, something like the Canadian "baby bonus" program could indirectly address the concerns in point 3, without increasing wages for an entire class of workers.  We already do this indirectly, via tax policy, so why not be explicit? 

What do others think

A good story on structural unemployment and an excuse to bring up intra/inter-group dynamics again

From Marketplace:
Basically, [Management professor Peter Cappelli at the Wharton School] thinks employers are just being cheap. He says this is something they learned from the downsizing of the 1980s. It was so easy to snap up laid-off workers somebody else had already trained. So then companies downsized their own training programs to save money. 
Meanwhile, they’ve steadily raised the bar on job applicants -- demanding ever-more credentials and work experience -- then complain they can’t find good help.
...
And which employees do companies provide with the most skills-enhancement? According to the training organization’s data, it’s not production workers, or customer service reps or new employees. It’s supervisors, managers and executives.
For a variety of reasons (demographic, economic, political, educational), I suspect that the group cohesion in management and the associated social classes has gotten stronger over the past few decades. If this is true, we would expect executives to hold an increasingly high perceived value for executive work while seeing non-executive work as less valuable and more interchangeable.

You can make the case that much of the training given to executives is of limited value. By comparison a strong and unambiguous case can be made for training that, for example, increases workers' relevant computer skills. If we're spending more on the first than on the second, there's something wrong. My theory is intra/inter-group dynamics. Any other suggestions?

Thursday, June 21, 2012

Thought provoking idea of the day

Maybe meritocracy is more complicated than I assumed:

What this means paradoxically, is that to encourage an honest “meritocracy”, one may have to discard equality of opportunity. To come from a dishonored or disgraced family would be a hurdle to overcome. To come from a family with an unblemished record for integrity, would be boost. Not because this is fair to the child. It is demonstrably unfair. But, because it may be the only enforcement mechanism that will work against ambitious parents.
The context for this post is Karl Smith thinking about why cheaters will rise to the top of a meritocracy and the implications that this will have.  I often disagree with Karl Smith but, on this point, I think that he might be on to something that is hard to deal with in a social context like that of the United States.

But I would be delighted to be proven wrong.

Wednesday, June 20, 2012

One more point about that "On the Media" story

There's something I should have spelled out explicitly in my previous rant. "Sociopathic, genocidal dictators plant propaganda in respected news outlets through think tanks and paid experts" is bad enough. These commissioned op-eds can do real damage, muddying the debate and preventing or delaying sanctions and other actions against these regimes, but the more frightening headline is "Even sociopathic, genocidal dictators can plant propaganda in respected news outlets through think tanks and paid experts."


We are literally talking about some of the worst people in the world arguing some of the most indefensible positions imaginable. If they can, with enough money, buy not only a platform but also a veneer of journalistic respectability then there really is no standard. No speaker or message is too repugnant or dishonest to be whitewashed by the press.

If editors and publishers won't bother to vet proxies for Assad and Charles Taylor, what kind of scrutiny can we expect for a political party or a Fortune 500 company? When we read an editorial about global warming or see a study on tax increases cited, can we assume that the journalistic standards are any higher?

Education Reform: back from the grave?

It has been a while since we did a post on education reform.  But Karl Smith Adam Ozimek reported on a recent paper by Jesse Rothstein on teacher firings. Since this was similar to a discussion here on Observational Epidemiology. I thought that I would quote it in full:
But they assume that replacement teachers can be hired without limit, with no increase in compensation, from the same quality distribution from which current teachers are drawn. This is implausible. The elasticity of labor supply to teaching is unlikely to be infinite. Moreover, one would expect reductions in job security to require a compensating differential even to maintain the current level of supply. Thus, if salaries are not adjusted, filling positions vacated by fired teachers would likely require reducing hiring standards and thus, presumably, quality.
And then quote a line from Karl Adam Ozimek:
As a theoretical exercise I wouldn’t draw too strong conclusions from the study, but it does tell an important story that I think is often ignored in the education debate: many reforms increase risks to the teaching profession, and absent some offsetting benefits, this will decrease labor supply.
What I find interesting here is the need to hedge on very simple economics when they have a non-conservative conclusion.  Of course labor is not infinitely elastic.  But we take economists deadly seriously when they discuss things like dead-weight loss in taxation.  In these cases we are often willing to accept the theoretical conclusions of high taxes = lower growth despite inconvenient empirical examples (compare growth under Eisenhower to Kennedy, or Clinton to George W Bush).

That being said, to give credit where credit is due, Karl Adam Ozimek does then go on to talk about the policy implications of these changes.  What comes out is the story of a tough trade-off in policy and the need to cleverly design a system.  That is a possible path to improvement.  Careful rethinking and redesign may work, as might seriously questioning ideas like credentialism.

But these ideas are not a clear path forward with risks or costs.  Thoughtful reform can work and does work.  Cutting resources because you dislike "government workers" is a very strong belief in the absence of public goods.

EDIT: Name error noted in comments is corrected

Tuesday, June 19, 2012

A rant about On The Media and journalistic criticism

I found myself listening to NPR's On the Media in the car this afternoon. It's not something I do often. I have a huge problem with the current state of journalistic criticism, a field that now seems to consist entirely of  self-serving, self-satisfied trivialists and tribalists united by a blood oath not to acknowledge any of the serious problems facing journalism. OMD is often among the worst offenders.

I got lured into this story partially because the topic (brutal regimes hiring PR firms) interested me and mainly because there was nothing else interesting on the radio.

Here was the introduction:

"The New York Times reported this week that the Assad family employs Western PR firms to polish its image for the rest of the world. A few years ago, Harper’s contributing editor Ken Silverstein went undercover and approached PR firms as a fake representative of a tyrant who needed to improve his image. He talks to Bob about what he learned."

The interview was carried out with the usual mixture of smugness and breathlessness. The superiority was understandable if annoying (they were, after all, talking about apologists for dictators), but the surprise was truly difficult to fathom. Silverstein told about firms competing for these huge contracts as if the behavior was unprecedented. Not only was what he described pretty much what you'd expect; it was also perfectly consistent with what we see from law firms, lobbyists and consultants. In other words, there wasn't really a story there.

Silverstein then described what he called the "dirty little secret" of the industry, namely that the work (according to him) wasn't that effective, an assertion he supported with an anecdote about a firm that had bragged about getting a client off a ten-worst list by getting it moved down to the eleventh spot. He didn't seem to realize that this point undercut the importance of much of what he had said before (why should we care about dictators hiring PR firms if those firms are ineffective?). Nor did he seem to realize that in this context, moving past threshold values like 1, 5, and 10 really is a big deal.

Buried in the middle of this, Silverstein quickly and casually threw out the following: the firms promised clients that the firms would "write and place op-eds in American newspapers [and] recruit academics or think tanks to put their name on it." I at that point assumed that even Garfield (who is truly terrible at his job) would jump in and and ask for details, but the subject held no more interest for him than it did for Silverstein.

In any sane world, "Sociopathic, genocidal dictators plant propaganda in respected news outlets through think tanks and paid experts" would be the lede on a program about the media while "PR costs more when you're caught doing horrible things but it may not be worth the money" would be, at best, a side note. Unfortunately, OMD tends to follow the Jack Shafer school of journalistic criticism: sins are weighted by the sinner's membership and position in the journalistic tribe (multiple examples available on request).

The PR firms described here are certainly sleazy but not in a surprising or perhaps even unethical way. I would certainly have more respect for a PR firm or law firm that refused to have dealings with these regimes but I'm not sure any rules are being broken by taking them on as clients. As for the suggestion that these monsters are not getting their money's worth, I'm not entirely sure I buy the argument and even if it's true, I don't find it particularly newsworthy or worrisome. Lots of goods and services aren't worth the price and if someone is going to get screwed on a deal, I'd actually prefer it to be a murderous dictator.

By comparison, there are clear and serious ethical breeches by the editors who print unvetted op-eds by experts with undisclosed conflicts of interest and by the reporters who cite commissioned studies as if they represented reliable, impartial research, but under the Shafer system, the offenses of the PR firms are more heavily weighted because they are committed by outsiders.

I know this inter/intra-group explanation may seem like a bit of a jump, but there's clearly a problem with journalism and accountability. If you have an alternate theory, I'm open to suggestions.

Monday, June 18, 2012

Hoisted from the WCI comments

Comment from Shangwen on Worthwhile Canadian Initiative:

I think a corollary of your title could be, "liberalization matters, not privatization". When most people argue for or against the "privatization" of health care, all they are really talking about is changing the payment stream so that individuals (out of pocket or privately insured) pay entirely for the same system for which the government now pays a majority share. But that does not involve changing occupational monopolies, disencumbering drug development, or making it possible for innovation to force prices down. That is flipping only one switch in a very large machine.


I think that this is a really good point. As a private individual I cannot purchase most medicines (even at the marginal cost) without a prescription from a medical doctor. But there are some conditions (e.g. hypertension) where I could probably do a better job of monitoring on my own using a snazzy blood pressure machine and being able to look at factors like time of day or day of week.

In the same sense, if I get an infection at 5 pm on a Friday afternoon, my options are essentially the emergency room or, just maybe, urgent care. There are no circumstances in which I could call a nurse with my symptoms and try a first line antibiotic before heading into the emergency room.

It is really unclear that market forces will work under these restricted conditions. And this ignores other factors like information asymmetry. Nor am I convinced that all of these conditions are necessary to force innovation forward -- restricted resources can also be an effective spur to innovation and American (and Canadian) medicine has mostly avoided that scenario so far.