Saturday, February 27, 2010

Poll: What is the appropriate role for online learning?

If you have been following my blog, you have noticed that I have been posting mini-lectures on microeconomics to YouTube (in an effort to freely provide a close substitute for my book and solutions manual). My YouTube channel has been surprisingly popular: So far, it has more than 230 subscribers, and my videos have been viewed over 35,000 times.

Aside from the public comments on my videos, I have received e-mails and other solicitations regarding my YouTube project. Some are from students who want me to help with their homework. Others are from online learning sites that want to feature my videos.

Without giving too much away about why I am posting these videos (you can see my introductory video here), I'm going to lead into the poll question for the week.

What role should online learning play?

(a) Online learning should take over. Classrooms are for dinosaurs who use actual pencils.
(b) In-class instruction is irreplaceable. Online learning is too passive to allow for real learning.
(c) Some hybrid is optimal. Information transmission is most efficient through the Internet, but questions are better to answer in person.

As with all polls on this site, the poll is open for a week. Please vote early and often (on the sidebar). Tell your friends, your followers, your subscribers and your students to vote as well. I'm interested in hearing what you have to say.

Wednesday, February 24, 2010

An application of marginal benefit equals marginal cost

Last week, I read an article about how University of Chicago's drinking policy remains unchanged in spite of a recent uptick in number of emergency room cases. Here's an excerpt.

There are currently no plans to change University alcohol citation policy despite a spike in alcohol-related ER visits last quarter, administrators said this week. The policy, relatively lenient compared to that of peer institutions, has resulted in a handful of disciplinary cases, but no arrests.

“The incidence of over drinking on our campus is still quite small compared to other schools,” Dean of Students Susan Art said. “It’s not a prevalent part of our culture in the College.”

According to crime statistics provided by the University of Chicago Web site, there were 14 disciplinary referrals for campus violations of liquor law during the years 2006–2008, and no arrests. According to similar information published online by the University of Pennsylvania, the UPenn campus had 73 disciplinary referrals and 54 arrests for on-campus violations of liquor law in the same time period.

Despite last quarter’s spike in emergency room visits due to excessive alcohol consumption, Art said that fewer than 20 students were sent to the ER as a result of alcohol intoxication last fall quarter.

“Certainly there is protocol for students committing violations. But we just don’t have that many incidents,” said Bob Mason, a UCPD spokesman. “Maybe that has to do with the size of the student population or maybe that has to do with the kind of people here at the University of Chicago.”

Mason added that the UCPD arrests underage drinkers. “Students are treated like any other member of the community,” Mason said.

First, I'd like to offer a clarification. The article is concerned with whether University of Chicago ought to involve the campus police when non-police authorities on campus discover underage drinking. From the sounds of it, Northwestern and UPenn like to involve the authorities in underage drinking incidents. UChicago doesn't. But, regardless of where you're caught drinking, if the police catch you, you're subject to a standard process (citations and arrests).

Clarifications aside, I'm worried about the economics of the article. We should evaluate a drinking policy based on what it accomplishes, rather than empty comparisons of total amount of drinking. What is the marginal benefit of having one policy rather than another? The article doesn't address this. Instead, we have two opposing red herrings.

UChicago students don't drink that much (Dean of Students Susan Art's position).
The recent increase in excessive drinking is a reason to change the policy (The article's position)

Both parties use the same logic.

More drinking ==> Stricter policy is optimal

Both parties make the same mistake. They don't think on the margin. The article tries to deal with the "total" problem, rather than address what the policy change would accomplish.

The relevant questions are not "how much do UChicago undergrads drink?" and "How much did UChicago underage drinking increase last quarter under the old regime?" The relevant questions are "How much would UChicago underage drinking change if the administration switched to a policy of citing and arresting?" and "How would the change in drinking last quarter be different under a cite-and-arrest policy?"

Thinking on the margin about UChicago underage drinking

Deterrence is the primary goal of a strict underage drinking policy. After all, the article is complaining that the kids who drink don't get arrested or cited for the alcohol violations. Deep down, I suspect the author wants to deter undergraduates from excessive underage drinking. She just didn't come out and say it.

In my view, citing and arresting kids would probably have a bigger effect on underage drinking at UChicago than "dealing with it as a health issue." People respond to incentives, and calling underage drinking a "health issue" is not exactly going to deter an overworked undergraduate from partying hard after midterms. There's probably a deterrence effect of strict punishment, and therefore, there might be a benefit to the cite-and-arrest policy.

On the other hand, there are clear costs of having a cite-and-arrest policy.

On one front, UChicago police have their hands full in protecting students from other members of the community. The south side of Chicago is not especially safe at night, but it is better when the UCPD are on their patrols. Dumping a bunch of petty underage drinking cases on the UCPD is going to make them less effective in this task (because they'll have to divert some resources).

On another front, it's possible that issuing a cite-and-arrest policy could increase excessive drinking. It's true that we would expect the total amount of drinking to decline if the University issued a cite-and-arrest drinking policy, but most of this will happen on the extensive margin. People who don't value partying that much won't have their one or two beers. They'll have none at all.

But, it isn't clear what this will do to excessive drinking (which takes place on the intensive margin). If your punishment doesn't depend on how much you imbibe, you might as well get hammered if you're going to drink at all. Someone who would have 8 drinks might switch to 10 drinks because they want to take advantage of the fact that they're already breaking the law.**

Without some concrete data on the effects of other universities switching from lenient policies to more stringent ones regarding drinking, I don't know where to stand on the issue. That said, I think the above discussion identifies the relevant margins.

A note on economics

Economics is often criticized for "restating common sense." This complaint is at the top of the list of complaints that people offer while taking an Econ101 course. The students will say, "It's just common sense. Economics isn't hard at all." After making this complaint, many students will proceed to fail the exam. How can this be? First, economics isn't so easy. Most people find mastering economic logic to be a difficult task. Second, common sense isn't so common. Many people lack it -- even if they recognize that economics is purely and "merely" logic.

** Note: I realize this "imbibe more because I'm already in trouble if I get caught" motive is not thinking on the margin. Under most assumptions, a rational utility maximizing drinker would stop at 8 in either case. But, I have two points to counter this. First, I think this notion of how to behave is commonplace. Second, in the real world, we might not be dealing with a rational utility maximizing drinker (though I'll admit that if a "rational utility maximizing drinker" exists, he's probably at UChicago).

Saturday, February 20, 2010

Poll: Which two countries would you take in the winter Olympics?

A couple of professors I know made a bet over which two countries would garner the most medals (not just gold medals) in this year's Winter Olympics.

Professor 1 bet on The United States and Canada
Professor 2 bet on Germany and Austria

At current tally, Prof 1 leads Prof 2 by a margin of 28 to 18. Professor 1 told me that he thought he had a chance to win with the U.S. and Canada this year because the new Olympic sports (i.e., the Halfpipe and kindred events) tend to favor the United States. That said, the race isn't over yet. Germany and Austria are traditional powerhouses in the Winter Olympics, and there is still over a week left in the Olympics.

So, I'm going to make it my poll this week (with a couple of additional pairings thrown in), but I'm going to eliminate the starting advantage that the U.S. and Canada currently have. Instead of making the poll about the entire Olympic Games, I want to know how you think these pairings will do for the remainder of the Olympics.

Which pair of countries do you think will win the most medals FROM FEB 20 to THE END OF THE GAMES?

The United States and Canada (current tally 28)
Germany and Austria (current tally 18)
Russia and the United States (current tally 25)
Norway and France (current tally 17)

The poll is open for about a week (on the sidebar --->), and you are welcome to post any write in candidates in the comments. Vote early and often. I'm interested in hearing what you have to say.

Wednesday, February 17, 2010

Some thoughts on health care

This is a quote from a NBER working paper (link is gated) by Fogel and Lee:
Saving lives, as important as it is, and as effective as modern medicine has become, is not the main activity of physicians and other health professionals. As I have already indicated, it is likely that past public reforms, improvements in nutrition and other living standards, and the democratization of education have done much more to increase longevity than has clinical medicine. The main thing physicians do is make life more bearable: to relieve pain, to reduce the severity of chronic conditions, to postpone disabilities or even overcome some of them, to mend broken limbs, to prescribe drugs, to reduce anxiety, overcome depression, and instruct individuals on how to take care of themselves.
Fogel and Lee go on to document that the United States -- a much richer nation than Britain -- has much higher demand for "access to specialists," "individually tailored health services," "inpatient care," and freedom of choice among alternatives. Each of these types of services is expensive, but Americans -- even those who economize on health care -- demand such services, whereas European nationals do not (at least not to the same extent). For example, Europeans are more willing to put up with long waiting lists for hernia operations (sometimes exceeding two years) when the typical wait in the United States is a few weeks.

Based on this discussion, Fogel and Lee conclude:
And so, what is viewed as "essential" health care in the United States includes items that in other cultures would be regarded as wasteful luxuries. This misunderstanding of the American system is relevant to the proposition that 15 percent of Americans are "uncovered" by health insurance. "Uncovered" does not mean untreated. The uninsured see doctors almost as frequently as the insured. Nor is it clear that the effectiveness of their care is always less than those who have insurance. The uninsured are treated in public clinics and in emergency rooms, which (although they lack the conveniences of insured care and may have long queues) provide competent services, both standard and high tech.
This paper sounds like it was written in response to the current health care debate, but it wasn't; it was written in 2003. Despite being written 7 years ago, this paper is now at the top of my list of must-read health care articles. I also recommend Friedman's How to Cure Health Care (2001).

Saturday, February 13, 2010

Valentine's Day

Here is an excerpt from an article on the History of Valentine's Day.
While some believe that Valentine's Day is celebrated in the middle of February to commemorate the anniversary of Valentine's death or burial — which probably occurred around 270 A.D — others claim that the Christian church may have decided to celebrate Valentine's feast day in the middle of February in an effort to 'christianize' celebrations of the pagan Lupercalia festival. In ancient Rome, February was the official beginning of spring and was considered a time for purification. Houses were ritually cleansed by sweeping them out and then sprinkling salt and a type of wheat called spelt throughout their interiors. Lupercalia, which began at the ides of February, February 15, was a fertility festival dedicated to Faunus, the Roman god of agriculture, as well as to the Roman founders Romulus and Remus.
The article goes further to describe Lupercalia:

To begin the festival, members of the Luperci, an order of Roman priests, would gather at the sacred cave where the infants Romulus and Remus, the founders of Rome, were believed to have been cared for by a she-wolf or lupa. The priests would then sacrifice a goat, for fertility, and a dog, for purification.

The boys then sliced the goat's hide into strips, dipped them in the sacrificial blood and took to the streets, gently slapping both women and fields of crops with the goathide strips. Far from being fearful, Roman women welcomed being touched with the hides because it was believed the strips would make them more fertile in the coming year. Later in the day, according to legend, all the young women in the city would place their names in a big urn. The city's bachelors would then each choose a name out of the urn and become paired for the year with his chosen woman. These matches often ended in marriage. Pope Gelasius declared February 14 St. Valentine's Day around 498 A.D. The Roman 'lottery' system for romantic pairing was deemed un-Christian and outlawed.
It would be interesting to have data on these pairings to determine how often they actually ended in marriage. I doubt the success rate was higher than speed dating or being high school sweethearts, but the lottery system might have some promise (just so long as you're not stuck with your unlucky draws for life).

The blind-date lotteries look like a centralized system of blind dating, which might not be so bad if the goal is help people find their best match. Lottery dating might help people meet potential mates who they would have never thought to approach because they were "out of their league." Aside from broadening horizons, these lotteries may have increased the number of connections a person made with members of the opposite sex. This further increases the chances of finding a marriageable mate.

Most people who showed up to the blind-date lotteries were probably just looking for a good time. Yet, it's possible that meeting a surprisingly fun person at a lottery could lead to a happy match -- a match for life. Nine years ago, I wasn't looking for marriage when I approached a friend's friend at her locker, but I am sure glad I asked her on that date.

This is a poll post, so I better get to the poll.

What do you think of Valentine's Day?

(a) Just a Hallmark holiday
(b) A wonderful celebration of love
(c) Merely singles out singles
(d) Too much pressure on previously-happy couples

The vote begins now, and the poll (on the sidebar -->) is open for a week. Vote early and often. Tell your sweetheart to vote, and remember that voting is anonymous. At any rate, I am looking forward to seeing what you have to say.

Wednesday, February 10, 2010

What is math?

Here is a quote from an article I read last week.
Most mathematicians at one time or another have probably found themselves in the position of trying to refute the notion that they are people with "a head for figures," or that they "know a lot of formulas." At such times it may be convenient to have an illustration at hand to show that mathematics need not be concerned with figures, either numerical or geometrical. For this purpose we recommend the statement and proof of our Theorem 1. The argument is carried out not in mathematical symbols but in ordinary English; there are no obscure or technical terms. Knowledge of calculus is not presupposed. In fact, one hardly needs to know how to count. Yet any mathematician will immediately recognize the argument as mathematical, while people without mathematical training will probably find difficulty in following the argument., though not because of unfamiliarity with the subject matter.

What, then, to raise the old question once more, is mathematics? The answer, it appears, is that any argument which is carried out with sufficient precision is mathematical, and the reason your friends and ours cannot understand mathematics is not because they have no head for figures, but because they are unable to achieve the degree of concentration required to follow a moderately involved sequences of inferences. This observation will hardly be news to those engaged in teaching mathematics, but it may not be so readily accepted by people outside of the profession. For them the foregoing may serve as a useful illustration.
The article is a classic in economics by Gale and Shapley. The title? College Admissions and the Stability of Marriage. If you "have a head for inferences," I suggest you read it [here]. It is an interesting and influential paper, and it is only 7 pages long.

Saturday, February 6, 2010

Poll: Do you value a product more if it is "American"?

There has been a recent hubbub in the blogs I read about recent "Buy American" provisions in an agreement between the United States and Canada. Here's the background:
The U.S. and Canada, its largest trading partner, reached a preliminary deal to settle what had become an acrimonious dispute over "Buy American" provisions in the U.S. stimulus package.

The deal, if approved, will give companies on both sides of the border access to government procurement contracts at the state and local levels. U.S. Trade Representative Ron Kirk said the increased access for U.S. firms in Canada would be worth billions of dollars in contracts.

Sound good? Not to Don Boudreaux. Here's some of the letter he wrote to the Wall Street Journal.

In other words, Washington agrees that it will spend Americans’ tax dollars wisely – that is, get the most value in return for each dollar spent – only if Ottawa does the same with Canadians’ tax dollars. If instead Ottawa had stubbornly insisted on wasting Canadian taxpayer dollars by refusing to buy lower-priced goods and services from Americans, Washington would have boasted of its commitment to continue wasting American taxpayer dollars by refusing to buy lower-priced goods and services from Canadians.

This sort of Dali-esque surrealism of government behavior is ignored by the public and the punditry only because it is all too common.

He's not the only one on my blogroll who has come out writing against these agreements. Here's what Jeffrey Miron thinks about the issue:

Even if this issue gets resolved sensibly, it should never have arisen in the first place. When the government builds infrastructure, it should do so at minimum cost (quality adjusted). The Buy America provisions interfere with that objective and risk killing jobs when our trading partners retaliate.

A side issue (the subject of today's poll) is whether people value American products more because they were made in America. That is, are you more likely to buy a product if you know it was made in the United States (or Montana)? That brings me to the poll question of the week.

How do you value a product if it is "American made"?

(a) I like it about the same
(b) I like it better
(c) I like it worse

The poll is open for a week. Vote early and often (on the sidebar -->). I'm interested in hearing what you have to say.

Wednesday, February 3, 2010

Institutions, weather, culture and growth

I have been doing some reading on economic growth for my economic history class. One of the readings is the discussion (book review and interview) surrounding the publication of the 1998 book, The Wealth and Poverty of Nations: Why Some Are So Rich and Some Are So Poor by David Landes.

From what I can glean from these auxiliary sources, the book seems worthwhile. In the book, Landes tells a detailed economic history of Europe, including how weather patterns affected the optimal form of production (slave or non-slave labor). These broad weather patterns influenced the institutional framework (i.e., the rules of the game) that these societies set up. As Barry Eichengreen says in his review:
Europe's Industrial Revolution was, at the deepest level, a product of the Gulf Stream. The continent's mild summers permit intense physical activity, unlike the tropics, whose heat and humidity force even the most energetic to seek shelter from the midday sun, and where the incentive to find others to do hard labor accounts for the concentration of wealth and ultimately explains the rise of slave society, a form of economic and social organization incompatible with capitalist growth. As the author puts it with characteristic bluntness, "Where society is divided between a privileged few landowners and a large mass of poor, dependent, perhaps unfree laborers-in effect, between a school for laziness (or self-indulgence) over against a slough of despond-what is the incentive to change and improve?" Europe's cold winters suppressed pathogens and pests and rendered parasitism the exception, increasing the capacity of its natives for work. The continent enjoyed just the right amount of rainfall, between the extremes of desert, where crops died of thirst and topsoil eroded away, and the torrents of the Tropics, where jungle and rain forest crowded out settled agriculture.
More than just setting different legal rules of the game, this characterization of how weather ultimately became important is really a statement that informal institutions (norms, ethics, cultural rules of society) rule. Landes singles out the Protestant work ethic that was exported from Northern Europe to America:
The Reformation was a fundamental threat to the established church, which regarded intellectual and political novelty as subversive and persecuted nonconformists, prohibiting study abroad and stifling the spread of scientific knowledge. Landes argues that the anti-Protestant backlash sealed the fate of southern Europe for the next three hundred years. Southern Europe, in turn, exported its shortcomings to South America. In North America, by contrast, geography and the culture of dissent carried the day. Abundant free land created a society of small farmers and well-paid workers of unparalleled individualism, self-reliance, and initiative. Ample natural resources and an extensive consumer market, itself the product of a relatively level income distribution, led to the American system of manufactures, a form of industrial organization in which raw materials were intensively used to churn out standardized products using what ultimately developed into mass-production techniques. Even the American South, where climate and geography had encouraged the use of slave labor, quickly shed the lingering effects of its anticapitalist system once air conditioning freed it of its climatic handicap.
As an economist, I appreciate the appeal to the rules of the game. Indeed, I believe that institutions rule in terms of explaining differences in economic growth across nations. I also believe that these rules of the game are difficult to quantify and test. It is one thing to tell a coherent story that matches the facts. It is a much bigger accomplishment to also document the causes of those facts with a systematic analysis of the historical record and data. I think that's why the 2001 paper by Daron Acemoglu, Simon Johnson, and James Robinson (The Colonial Origins of Comparative Development: An Empirical Investigation) is so important.

These authors found a systematic test of Landes proposition that institutions rule (but they don't single out the Protestant/anti-Protestant dimension). The 2001 paper (commonly referred to as AJR) uses differences the settler mortality rate during colonization to predict differences in the types of institutions that were set up. If a lot of settlers died upon arrival, the colonizing country would set up extractive institutions (like slave-plantation systems) to exploit the resources without endangering their countrymen to the harshness of the wild. If not many settlers died upon arrival, the colonizing country would adopt institutions that make for a pleasant society.

And, if there is one thing economists know about institutions, it's that they change slowly -- So slowly that the institutions set up in the 1600s left a permanent mark on economic development four centuries later. Indeed, even after many revolutions, the countries that were settled under extractive institutions still have extractive institutions that closely resemble those set up during colonial times. Countries -- like the United States -- that were settled by people who wanted to live here are much better off by comparison.

When I first read the review of Landes' book, I wasn't impressed because I had previously read AJR. Then, I looked at the publication date. Landes documented his hypothesis in 1998 before AJR developed the test in 2001. That gave me a deeper appreciation for what the 1998 book purportedly accomplished. Next up? Maybe I'll read the original book.