Monday, October 10, 2011

Testing teachers

It has been a while since we talked about education reform but Arnold Kling has a great perspective:

Simple formulas can be “gamed.” That is, employees learn to achieve the objectives in the formula while failing to work toward the longer-term goals of the firm. On Wall Street, we have seen how bonus formulas proved dysfunctional. The older partnership form of organization appears to have provided better incentives.

A government-run system of teacher compensation, based on test scores, would in some ways be the worst of all worlds. It would create incentives for teachers to “game” the system. It would give too much weight to a noisy indicator of performance. As a result, it would do little or nothing to improve accountability or to reward better teachers.


This classic insight is shown in this dilbert comic.

Are we sure that test score based measures are the way to go? Most information technology jobs have the same sort of issues, often solved by comparative rankings and broad evaluations. Even worse, a bad metric drives out the good (meaning it could actually be counter-productive).

The whole post is worth reading.

Sunday, October 9, 2011

More on Executive Compensation

I do not agree with the amount of vitriol in the linked post, but Joshua Brown has a very strong set of feeling on the decision to pay two low level executives at Bank of America an $11 million dollar severance package:

You pay fired executives more in severance than the average American worker will earn in a lifetime. For most people on the outside looking in, this seems like it's from outer space, another world entirely. These numbers just do not exist to regular human beings, they cannot be fathomed.


He also points out the bad timing:

It's not that this isn't your prerogative as a private company - it is. But seriously, numbers like these at a time when you're instituting added fees on customer accounts just sound farcical, almost like you're making these payments to get a reaction out people.


I have been interested in this dynamic for a while. Mostly because I am beginning to see executive compensation as an intriguing form of market failure. After all, let us consider the example of these two executives. What are they being compensated for?

Do executives at banks really add so much value that $5.5 million dollar severance packages are just a way of saying thank you? Is there really no competitive pressure on salaries? Is the supply of potential bankers really this low? Supply side issues seem to be dubious. Are these skills really so rare (and, if they are, how do banks really select for them because the rest of us want to know).

Or is it due to the risk of taking a corporate job instead of being a school teacher? Well, these two executives are not really taking any real risk. Even if this is the last job either one ever holds. they are already well above the typical lifetime earnings curve based on this severance package alone. Debts required to reach this position (like School debt) are simply dwarfed by the size of the payout.

It is a very interesting problem.

Schlock Mercenary

It is certainly one of the best science fiction themed web comics out there. It is a new addition to the blogroll.

Enjoy!

Friday, October 7, 2011

Free TV blogging -- Why Weigel Broadcasting may be the best business story that no one's covering -- part I

[I should start with the disclaimer that all of the information I have about Weigel comes from two sources: Wikipedia and way too many hours of watching television. It's entirely possible that a competent journalist could discover that the truth here is something entirely different, but if competent journalists were paying attention I wouldn't be writing these posts.]

Though the improvement in picture and sound got most of the attention, another aspect of the transition to terrestrial digital was arguably more important, particularly for broadcasters: under the new technology, each station could broadcast multiple subchannels. The situation was analogous to the TV landscape thirty years earlier when cable and satellite stations were exploding on the scene. It's not surprising that someone would try to create the broadcast equivalent of superstations like TBS. What is surprising is who was able to get a channel up and running before any of the competitors were out of the gate.

The name of the channel was ThisTV. It was produced by a regional broadcasting called Weigel, best known for operating the last independent station in Chicago and being the home of the cult favorite Svengoolie -- last of old time horror hosts. Weigel had a content deal with MGM which was not nearly as impressive as it sounds -- Turner had bought out the classic MGM library years earlier -- but MGM still had a lot of films including the catalog of American International, the studio responsible for virtually every drive in movie you can think of from the late Fifties through the early Seventies.

Access to all those AIP films probably had a lot to do with the unique ThisTV brand. Here's how I summed it up earlier:
Weigel are the people behind ThisTV and the exceptionally good retro station MeTV (more on that later). ThisTV is basically a poor man's TCM. It can't compete with Turner's movie channel in terms of library and budget -- no one can (if my cable company hadn't bumped TCM to a more expensive tier I never would have dropped the service), but it manages to do a lot with limited resources using imagination and personality. As a movie channel, it consistently beats the hell out of AMC.

ThisTV has caught on to the fact that the most interesting films are often on the far ends of the spectrum and has responded with a wonderful mixture of art house and grind house. Among the former, you can see films like Persona, the Music Lovers and Paths of Glory. Among the latter you'll find American International quickies and action pictures with titles like Pray for Death. You can even find films that fit into both categories like Corman's Poe films or Milius' Dillinger.

If I ran a TV station, I would definitely combine Bergman and ninjas. I would not, however, run Mario Bava's feature length pulp magazine cover, Planet of the Vampires from twelve till two. Some of us have to get up in the morning.

This mix was in place from the very beginning. The station officially debuted on November 1, 2008 with Spike Lee's She's Gotta Have It but many stations started carrying it a day earlier to take advantage of a day of cheesy Halloween horror films. It was a formula that made a virtue out of cheapness (rarely seen auteur films and drive-in movies both have the advantage of not costing much) and it produced a format that's been running smoothly with remarkably few adjustments for almost three years.

For a small player to identify a new market, develop a concept, negotiate the necessary deals with a content provider (MGM), line up affiliates, make the countless other arrangements that accompany a major launch and to be up and running with a quality product when the support technology first comes online is an impressive accomplishment. But it gets better.

So far we have a solid business story -- small yet nimble company with some good ideas beats big, well-established competitors into a new market. Not exactly the most original piece of journalism but certainly good enough for the front page of the business section. However the story doesn't stop there. Weigel didn't just beat its big and well-financed competitors; it lapped them. Before the next entrant, Tribune/WGN, was able to get its station, AntennaTV on the air, Weigel managed to launch a second channel, the ambitious classic television station, METV. If this weren't enough, AntennaTV is the only one of the three to look slapped together despite having taken far longer to make it to the air (of course, we have no way of knowing how long it took Tribune to see the opportunity and how long it took them to act on it but either way Weigel looks good by comparison).

To put this in context, at least half of this story takes place after the collapse of '08, a downturn that hit advertiser-based businesses particularly hard. Furthermore, the story occurs in an industry that a large number of lobbyists and at least a few pundits were literally trying to kill. There had even been a New York Times op-ed calling for the government to eliminate over the air television and sell off the spectrum.

One of the great memes of the Great Recession has been that uncertainty paralyzes businesses. Even the possibility of a tax increase or some additional regulation -- both extremely mild by historical standards -- are enough to bring the economy to a standstill, but here's a market filled with unknowns under a credible threat of annihilation and we can still find a company like Weigel moving aggressively to establish dominance of it.

That's the other side of uncertainty. It allows companies to substitute boldness and decisiveness for money and market position and take advantage of opportunities that would otherwise be out of their reach.

[also posted at MippyvilleTV]

Thursday, October 6, 2011

CERN disproves global warming!

I exaggerate but not by much. Jon Chait has a wonderful time disembowelling this piece of bottom-scraping from the WSJ.

(The Churchill analogy alone is worth the price of admission)

Equal time for American Public Media

Having singled out NPR, I should mention that APM's Marketplace has been doing some extraordinary work lately. Today they ran the best piece I've come across yet on the business genius of Steve Jobs but the whole episode merits a link.

It feels strange talking about good journalism.

The unhappiest place on earth?

NPR has an excellent report out on J visas:

“You have 300,000 workers come to the US and they’re just lacking in protections,” Costa said. “Workplace protections, wage protections and we’ve see complaints where employers have been threatening people who complain about their work conditions with deportation.”

In fact, it’s even worse than that. In recent years, J1 workers have reported that their stays in the US were characterized by menial jobs, low wages, filthy living conditions, and a lot of economic exploitation.

J1 workers apply for the program and then pay between $3,000 and $6,000 to a sponsoring organization, accredited by the State Department. The sponsor organization places them with American companies. One of the biggest J1 employers is Disneyland.

“They work on rides, quick service food and beverage, housekeeping, parking attendants, merchandising, lifeguards, dispatch, and most importantly showkeepers- those are the janitors,” said Kit Jonson, a law professor at the University of North Dakota.

Jonson’s been researching the J1 labor force and says it’s become a very clever business strategy for American companies. They save on wages, state and federal taxes, healthcare, housing and pension plans.”

“For Disney those figures end up being really stunning,” Jonson said. “Disney’s saving in wages alone upwards of $18.2 million a year in hiring international workers. So international students are simply a lot cheaper than American labor.”

Both Kit Johnson and Daniel Costa say that especially now, when unemployment is so high, these jobs should be filled by local workers. But J1 workers are more attractive because on top of the cost savings, they’re less likely to put up a fuss. If they do, they’re easy to get rid of. Like a group of J1’s from Russia who came to work as lifeguards in Texas, and ended up begging in the streets when they weren’t paid.

Response to Comments

Comments from Trevor:

I agree that the logic of this argument is very strong, and it seems to me that the same argument applies to the morality of punishment.


and the link posted by Stuart Buck seemed to share a common theme. In the link, the authors argue that:

According to data provided by the California Department of Corrections and Rehabilitation, in 1977, parolees who were returned to prison or convicted of new crimes accounted for just 10% of California’s prison population. The percentage topped 20 only once prior to 1980. In 2009, however, the number was an alarming 77%, having held firm between the high 60s and low 80s since 1986.


I think that there is a real point here: vindictiveness is expensive. The previous focus in California prisons on rehabilitation and returning prisoners to society was based around minimizing losses of social capital. This is not to excuse the crimes (as most criminal activity is either selfish, mean or petty), but to point out that a focus on punishment is expensive and not especially good for the prisoners themselves. Clearly something was going right if only 10% of prisoners were coming back into prison. A return rate of 77% might actually indicate that we are simultaneously less safe and making life worse for the imprisoned.

Now, we can always find single examples of people who should never be released. Serial killers with psychiatric issues come immediately to mind. But we should not let the extreme example determine the policy for the median prisoner. Similarly, we have a 100% chance that at least one person on parole will re-offend. That would be true even if the crime rate in the parolees was less than that in the general population.

This is a hard issue. I have strong and irrational feelings of fear about many convicts. But the focus on punishment over rehabilitation may well have been a mistake.

Wednesday, October 5, 2011

"100,000 Tablets for School Children in New Delhi"

To be perfectly honest, the original post I had planned for this NPR story was rather critical of Apple, but before I could get around to posting it word came of Steve Jobs passing away and it seemed disrespectful. I'm sure we'll return to this subject later. In the mean time, check out this interesting interview from NPR's the World.

100,000 Tablets for School Children in New Delhi

Market Efficiency

For a long time now I have been hearing about how markets set salaries based on the productivity and value of an employee. The ideal was that increases in executive pay were a function of the greater influence a top executive could have in the information age. It was a nice story. Unfortunately, it also seems to be untrue:

Companies have long hid the way they set executive pay, but in late 2006, the Securities and Exchange Commission began compelling companies to disclose the specifics of how they use peer groups to determine executive pay.

Since then, researchers have found that about 90 percent of major U.S. companies expressly set their executive pay targets at or above the median of their peer group. This creates just the kinds of circumstances that drive pay upward.

Moreover, the jump in pay because of peer benchmarking is significant. A chief executive’s pay is more influenced by what his or her “peers” earn than by the company’s recent performance for shareholders, according to two independent research efforts based on the new disclosures. One was by Michael Faulkender at the University of Maryland and Jun Yang of Indiana University, and another was led by John Bizjak at Texas Christian University.


As Kevin Drum observes:

Adjusted for inflation, cash compensation for line workers has actually decreased over the past few decades, and even when you include healthcare compensation it's grown only about 30% or so. In contrast, executive compensation over the same period has more than quadrupled.


This is clearly a result of market failure. It also suggests that the reasons for increases in executive compensation are entirely due to poorly designed compensation system and not because of market forces. Examples like this one are worth keeping in mind when considering whether a market-based outcome is really utility maximizing or not.

Tuesday, October 4, 2011

Effect Sizes

I think that this is very insightful:

They mistake small truths for large ones, and use the small truth to obfuscate the big one. So, the truth - that a few of the unemployed don’t want to work - is exaggerated and used to hide the bigger truth, that the vast majority of unemployment has other causes.


Mark and I have often discussed how effect size can be easily overlooked in modern debates. In epidemiology, for example, it can be the case that a drug has a serious side effect that is so rare that it basically cannot change the risk-benefit calculus. So, for example, statins can cause rhabdomyolysis (as an adverse drug side effect) despite have massive benefits on all-cause mortality (in secondary prevention of cardiovascular disease). But the rare side effect is often newsworthy and may discourage patients from seeking a beneficial therapy. Fortunately, we have randomized trials to sort out what the net impact of the benefits and risks of the drugs is like across a whole population.

I think lacking these experiments makes it easy to get focused on the details in macroeconomics. Policies that may increase utility across the whole population (e.g. immigration) may have costs to individual workers. Failing to properly specify the relative effect size of different interventions may lead to a focus on "second or third order effects". Or, even worse, to misjudging the net impact of a policy.

I think that might well be correct in the example above, as well. It is certain that there are people who would hire more if the minimum wage was to drop. But it is unclear that adjusting the minimum wage would have a major impact on the >9% unemployment rate we have in the United States. We may have to look elsewhere for solutions.

Now, implementing this advice is rough. Which is why I am pleased we have experiments over here in epidemiology.

Monday, October 3, 2011

The danger of deserve

Karl Smith:

On Facebook I think Robin framed the question as “how weak do temptations have to be before they make people less deserving of charity”

My clear answer would be that there is no level so low. Human suffering is bad. Reductions in human suffering are good.

Why humans are suffering is of concern to us in knowing when our interventions might be productive but it doesn’t affect whether they are warranted.


This is, in my view, precisely correct. The decision to impose a moral worthiness component to helping others is the source of a great deal of misery. For example, the moral implications of being HIV+ (in the early years of the epidemic) clearly reduced overall public health (due to stigma preventing patients from seeking care).

That isn't to say that some approaches that feel good may be counterproductive. If we are going to be good utilitarians then we really need to consider the global consequences of an action. Making one person better off at the cost of making many others miserable is typically a bad trade-off.

But, insofar as we can make lives better, does it ever make sense to ask if people deserve to have better lives?

Friday, September 30, 2011

Why I keep going on about rabbit ears

I'm starting a string of posts on this so I thought I'd take a few moments and explain why I think over-the-air TV is worth discussing.

1. I genuinely like the product

I get my television through an antenna on the top of my set and I can't think of a purchase I've been happier with. The picture is DVD quality and I get over a hundred channels, some of them very good, all for free.

2. It provides an excellent service for people who could use something nice

It has always sucked to be poor but in recent years we seemed determined to make it suck worse. Over the air digital television bucks that trend. Anyone with an old TV, a second hand converter box and a set of dollar store rabbit ears can have a source of entertainment, news and education (I get ten PBS channels). That may not seem like much to you but to a lot of people in this country, it can be a major improvement in quality of life.

3. There are other people trying to take that service away

Some would like to sell off the part of the spectrum used for television. Economist Richard Thaler even wrote a New York Times op-ed on the subject, explaining how the sale could solve virtually all of society's ill. Think I'm exaggerating?
Here's a list of national domestic priorities, in no particular order: Stimulate the economy, improve health care, offer fast Internet connections to all of our schools, foster development of advanced technology. Oh, and let’s not forget, we’d better do something about the budget deficit.

Now, suppose that there were a way to deal effectively with all of those things at once, without hurting anyone... I know that this sounds like the second coming of voodoo economics, but bear with me. This proposal involves no magical thinking, just good common sense: By simply reallocating the way we use the radio spectrum now devoted to over-the-air television broadcasting, we can create a bonanza for the government, stimulate the economy and advance all of the other goals listed above. Really.
(I'm tempted to go off on multiple tangents here about how these sales of public land of have a way of going badly for the government and the tax payer, or how the VHF part of the spectrum isn't actually that useful for mobile applications, or how oblivious men like Thaler are to what life is like in the bottom decile, but Rajiv Sethi has already written the definitive rebuttal so I'll just leave it with a link and a recommendation to follow it.)

At the risk of sounding paranoid, I very much doubt that this idea simply popped into Thaler's head. We live in an age subsidized discourse. Whenever you read a news story or opinion piece that seems to come from a lobbyist's desk, you can generally assume that it originally did. I'm not saying that Thaler was paid to hold these opinions -- I'm sure he wasn't -- but I'll bet good money that the experts he relied on were, either directly on indirectly.

There is a huge imbalance of money in this conflict. A number of big and deep-pocketed corporations are gunning for OTA broadcasters. Some would like to carve up the spectrum. Other would just like to get rid of the competition.

4. Competition is good

And over-the-air television plays a vital role in maintaining competition in the world of live TV. If access is limited to cable/phone lines and satellite, the business will always be dominated by a handful of very big and powerful companies and since these companies fall in the middle of the supply chain, we have to worry about both monopolistic and monopsonistic effects.

The industry currently runs on the up-sell model: get people in the door with a twenty or thirty dollar a month plan (sometimes helped along with some opaque pricing), then make the package crappy enough to get people to move up to a more expensive tier. OTA television presents a potentially deadly threat to that model because, in most markets, OTA is actually better than cable's basic package. It has more channels, better content and less compression (this may be less applicable to satellite). Cable has been able to ignore this threat up until now because most consumers are unaware of what they can get for free, but if word gets out cable companies will have to start giving customers considerably more value for their money.

5. There's a story here

And it has been woefully under-reported. New technology. New markets. Small but smart players like Weigel Broadcasting coming up with innovative business models and easily lapping major media companies. For me that's way more interesting than reading about Facebook throwing money at some business problem.

(also posted at MippyvilleTV)

You should read this

Science is about discarding models, not making them

Performance based teaching

Via Megan McArdle, comes this gem:

It would be naïve to assume that the persons subjected to variable pay-for-performance would accept the respective criteria in a passive way and fulfil their work accordingly. Rather, they spend much energy and time trying to manipulate these criteria in their favour. This is facilitated by the fact that employees often know the specific features of their work better than their superiors. The wage explosions observable in many sectors of the economy can at least partly be attributed to such manipulations, eg when managers are able to contract easily achievable performance goals.


Arnold Kling goes on to link this with education reform:


When a remote authority sets incentives, people respond by manipulating the system. This fact is poorly understood by education reformers who are fond of pay-for-performance and national standards, by health care reformers who are fond of paying for quality, and by financial regulators. In fact, the quoted paragraph provides an excellent description of the financial regulatory process under risk-based capital. The banks spent much energy and time trying to manipulate the risk-based capital regulations in their favor. They got what they wanted, in terms of risky portfolios backed by little capital.

The Hayekian story here is that effective compensation practices require local knowledge and tacit knowledge. In a large company, you give a middle manager a fair amount of discretion in compensating his or her staff. If instead you try to implement an automated bonus system, you will get gamed.


I think that this hits at the heart of the concerns Mark has been expressing for a while. Test-based systems that are implemented at a very high level encourage all sorts of behavior, and it is quite possible that manipulating the system will be easier than actually improving performance. Even worse, bad performers (gaming the system) have an advantage over good performers as they can get top scores for less total effort.

Bed performance drives out good performance and things get worse. I think Megan is very right to be skeptical about how easy it is to reform systems once they adopt this management style as the new backbone of the labor force are the people who thrive in gaming the system and they will resist change back to older approaches.

I wish we'd look more at the history of countries like Russia for how difficult it is to make top-down reform and economic control work at the national level.