Results for “model this” 3165 found
Politically incorrect paper of the month
African-Americans make up a larger proportion of students than teachers. Many educators say that as a result African-Americans students suffer because they lack role models and white students suffer because they lack diversity. In a newly published paper (working paper version), Thomas Dee (Swarthmore College) supports some but not all of this story. Using data from Tennessee’s Project Star, a very important experiment in which K-3 students were randomly assigned to small and regular sized classes, Dee finds that black students improve when they have black teachers. So far so good. Dee also finds, however, that white students improve when they have white teachers. Uh, oh. There goes the diversity is good for everyone story.
Dee is quick to point out that we don’t understand why students perform better with a teacher of their own race. If it is a role-model effect then why would white students perform poorly with black teachers – surely there are enough white role models to choose from that one more or less isn’t going to have an effect on the self-esteem of white students. Another theory, with some support from other studies, is that teachers spend more time helping students of their own race. Note that if it is the latter then better teacher training, to overcome natural biases, could improve the effectiveness of both white and black teachers.
The cite for the paper is Dee, Thomas S. 2004. Teachers, Race, and Student Achievement in a Randomized Experiment. The Review of Economics and Statistics 86(1): 195-210.
Does file-sharing hurt CD sales?
A new study by two researchers at Harvard Business School and the University of North Carolina, Chapel Hill, finds that sharing digital music files has no effect on CD sales. This is the first study that directly compares actual downloads of music files and store sales of CDs.
The authors, Associate Professor Felix Oberholzer-Gee of Harvard Business School in Boston and Professor Koleman Strumpf of the University of North Carolina, Chapel Hill, conclude that “File sharing had no effect on the sale of popular CDs in the second half of 2002. While downloads occurred on a vast scale during this period – 3 million simultaneous users shared 500 million files on the popular network FastTrack/KaZaA alone – most people who shared files appear to be individuals who would not have bought the albums that they downloaded,” say the authors…
Even in the professors’ most pessimistic statistical model, it takes 5,000 downloads to reduce the sales of an album by a single copy. If this worst-case scenario were true, file sharing would have reduced CD sales by 2 million copies in 2002. To provide a point of reference, CD sales actually declined by 139 million copies from 2000 to 2002.
Here is another interesting tidbit:
31 percent of all individuals who download music live in the United States. Other important countries are Germany with a 13 percent share of worldwide users, Italy with 11 percent, Japan with 8 percent and France with 7 percent. File sharers in the United States are particularly active. While they represent 31 percent of worldwide users, they download 36 percent of all files.
U.S. file sharers download files from all over the world. Only 45 percent of the files downloaded in the United States come from computers in the U.S. 16 percent of music files are downloaded from computers in Germany, 7 percent from Canada, 6 percent from Italy, 4 percent from the U.K. A legal strategy that focuses mostly on the United States is unlikely to change the supply of music files.
In other words, going after domestic uploaders, as the RCAA is doing, won’t cut off supply.
Here is one summary. Here is the original research.
My take: Yes I believe the result. Most downloaders are young or just sampling songs for kicks. But I doubt if this, legal developments aside, would be true five years from now. Over time I expect more people to forgo buying the CD, unless of course the law intervenes.
Addendum: Newmark’s Door offers some additional links. Larry Lessig argues for complementarity. Here is an article that copyright is too strict more generally, and yes The Grey Album is wonderful.
Was Nietzsche right?
In recent weeks there’s been a furor in the Washington D.C. area over lead in the District’s water supply. Today, the Washington Post (registration required) looks at why lead is bad for you and covers some of the science and public policy. That lead is bad for you is open and shut. Too much lead kills you and for kids, too much is not that much. But I am skeptical of recent studies that find that the worst effects of lead happen at the lowest levels of exposure.
Here‘s a typical newspaper account of one of those studies and a quote from a leading researcher on the topic:
“There is no safe level of blood lead,” said Dr. Bruce Lanphear, lead author of the lead study presented Monday at the Pediatric Academic Societies annual meeting.
Edward Calabrese would not agree. Calabrese is a toxicologist at UMass-Amherst and a leading scholar of hormesis, the phenomenon that most if not all toxins are actually good for you at sufficiently low doses. This does not imply that you should start adding mercury to your eggs or lead back into your pots. But the impact of toxins appears to be U-shaped–good for you at sufficiently low levels then bad as exposure increases. Whatever doesn’t kill you makes you stronger.
Hormesis also implies that linear models or threshold models of toxic impact are misspecified and understate the impact of toxins over some ranges of exposure.
Here’s a Scientific American article on Calabrese and hormesis.
Here’s my take on the economics and policy implications of hormesis.
Good economics makes for strange company
As you may recall, I showed that the CEA’s employment forecast, wrong as it has been, was simply a prediction of a return to trend and that Paul Krugman’s charges that the CEA has become corrupt are irresponsible and unjustified. (Sadly, Brad Delong has also signed on to these charges.)
I find myself in surprising company. In Salon, the left-leaning James K. Galbraith presents almost the same graph and the same conclusion.
Was Economic Team Bush getting its job targets, as many suspect, from Karl Rove? Did the professionals turn prostitute, as Krugman charges in no uncertain words?….Our new chart shows that the Bush forecast did not imply unusually rapid job growth for an economic expansion. …the failure of the jobs forecast did not occur because economic recovery forecasts were abnormal. They were not. So far as we can tell, it did not occur because someone cooked the books, under instruction or otherwise.
Shouldn’t DeLong and Krugman be worried when they fall to the left of Galbraith?
See also Econopundit who writes, “every economist who, like me, runs models with new quarterly data giving alternate (and possibly less-optimistic) employment forecasts is owed an apology. I have falsified nothing, but my results have been similar to those shown in the chart. I have reported the numbers that resulted from running a known model openly available to any and all (including Paul Krugman) who are competent to use it.”
Thanks to The Big Picture for the Galbraith link.
Punkers for capitalism
With his mohawk, ratty fatigues, assorted chains and his menagerie of tattoos – swallows on each shoulder, a nautical star on his back and the logo of the Bouncing Souls, a New York City punk band, on his right leg – 22-year-old Nick Rizzuto is the very picture of counterculture alienation. But it’s when he talks politics that Mr. Rizzuto sounds like a real radical, for a punk anyway. Mr. Rizzuto is adamantly in favor of lowering taxes and for school vouchers, and against campaign finance laws; his favorite Supreme Court justice is Clarence Thomas; he plans to vote for President Bush in November; and he’s hard-core into capitalism.
“Punks will tell me, `Punk and capitalism don’t go together,’ ” Mr. Rizzuto said. “I don’t understand where they’re coming from. The biggest punk scenes are in capitalist countries like the U.S., Canada and Japan. I haven’t heard of any new North Korean punk bands coming out. There’s no scene in Iran.”
Here is a New York Times article, don’t forget to check out the pictures (password required). Here is a website for GOP punkers, they seem to approve of Reagan’s famous threat to bomb the Soviet Union. Or perhaps it is just irony. They stress that they are not libertarians because America is “at war” with the left, and the libertarian philosophy is not well-suited to fighting a war. Here is their cited critique of the Canadian health care model. Good economics, but these punkers, oppositional by nature, feel a kneejerk need to defend every action of the Bush administration. Here is the ConservativePunk.com website, which offers an interview with right-wing punker Johnny Ramone. Here is yet another site, which cites right-thinking punk bloggers. And will National Review be pleased that MyEvilMinion.com links to them approvingly?
My take: Punk music needs an idea of evil and an oppositional stance. So punkers will adopt every position of defiance they can find, including in-your-face right-wing politics. But in the long run? Remember what The Clash sung: “You grow up, you calm down, working for the clampdown…”
When is it rational to honor sunk costs?
Alex offers up some biography and describes his encounter with sunk costs. He asks the classic question: why honor a cost once it is sunk? Why not just go ahead and do what is best?
Brian at CrookedTimber writes the following, citing a paper by philosopher Thomas Kelly:
The main idea (roughly stated) is that since the value of an action is partially determined by what happens in the future…our current actions can be sometimes justified by the redemptive value they confer on past actions.
What does this mean concretely? Here is one example from Kelly:
One might prefer that, if others have made significant sacrifices in attempting to realize some valuable state of affairs S, then their sacrifices not be in vain. That is, one might prefer that these sacrifices causally contribute to the realization of some valuable state of affairs…Interestingly, one sometimes is in a position to determine, by one’s own actions, whether the past efforts of others will have been in vain. This is true, for example, when it is within one’s power to finish some valuable project in whose service others have labored, but which they are now not in a position to complete. Let us say that when one acts so as to prevent the past efforts of others from having been in vain one redeems those efforts.
In other words, you don’t want to admit that you shouldn’t have started your blog. And how about this?
Gilbert Harman…observes that, so strong is our desire to see our own past efforts play a role in bringing about valuable ends, we will often adopt new ends, carefully tailored, so that our past efforts can be seen as instrumentally valuable means to the achievement of these ends.
Let’s not forget the game-theoretic rationale for honoring sunk costs: You might honor sunk costs so that others do not perceive you as wasteful, or so that others perceive you as constant and reliable. Robert Nozick argued that we follow through on sunk costs as a kind of self-discipline, to prevent ourselves from initiating too many stupid undertakings in the future. If you self-signal that you will follow through on your commitments, you will be more careful in accepting commitments in the first place.
By the way, I have found that women honor sunk costs to a greater degree than do men. Furthermore women often do not like it when men announce that something is “only” a sunk cost.
The bottom line: Once your model of choice is at all complex, no one knows what a sunk cost means any more. So a theoretical scolding of those who honor “sunk costs” is not completely well-defined. That being said, there is still the empirical question of whether most people attach too much weight to previous plans and have a status quo bias. The experimental evidence suggests that we are more rigid than we need to be. The propensity to honor previous commitments may have efficiency properties, but we cannot discard this proclivity when we ought to.
CDs as loss leaders
An increasing percentage of compact discs are sold in mega-chains, such as Best Buy or Wal-Mart, as loss leaders. Offer the CD at a very cheap price, and hope that the buyers also take home a television set. This practice is the central reason why Tower Records recently went bankrupt.
Loss leader CDs push music in a more mainstream direction. The impulse buy is for the TV, the musical purchase is planned, which favors established stars with new releases. Sudden impulse buys of unknown musical products, by definition, do not bring people into the store. In essence consumers have decided they would rather bundle hit musical releases with TV sets and computers (the Best Buy model), than with more obscure musical releases (the Tower model).
Consumers with mainstream musical tastes are better off, but how about consumers who prefer the niche products? On the downside, hit CDs are cross-subsidizing obscure CDs to a lesser degree than before.
Nonetheless not all hope is lost for buyers with indie tastes. Amazon.com and other Internet services offer a wide variety of releases and lessen the need for such a cross-subsidy. And keep in mind that the cross-subsidy went two ways. The customers who prefer music from Madagascar no longer have to cross-subsidize the Eminem displays in Tower. The CDs are held in Amazon-linked warehouses, which is cheaper, even once you take shipping into account. Furthermore the desire to build up the Amazon brand name cross-subsidizes obscure products of all kinds, many of which Amazon makes little or no money from.
The other key musical trend of our time is illegal downloading, which hurts the top artists most of all. Indie releases use the Internet for publicity, and world music artists learned to live without copyright protection a long time ago. Legal downloading, through iPod, subsidizes music of all kinds. None of the iTunes songs make money for Apple, rather music of your choice (if they can get the rights) is a loss leader for hardware. Most people buy iTunes, not for the latest hits, but to hold a diverse mix of their past and yet-to-be-known future favorites. And most satellite radio channels do not play hits but rather serve niche tastes. XM offers a wide variety of channels, in part to make its brand name well-known and focal.
The bottom line: Don’t worry about music as a loss leader. Cross-subsidies all over the place, and point in many differing directions. But at the end of the day, both the demand and supply for musical diversity are alive and well.
How to stay together
What best predicts whether a marriage will last?
The crucial predictors, say the researchers, are the presence of facial expressions that accompany emotions such as contempt. Gottman says that just watching a couple and looking for this expression, described as a sideways pull of a corner of the mouth accompanied by rolling eyes, is enough to make a good guess about a couple’s suitability. “This is our best predictor,” he says. “Contempt is the sulphuric acid of love.”
We are told that the entire model has a 94 percent success rate in predicting divorce.
Here is some positive advice for Valentine’s Day:
Gottman may also have stumbled across the secret of a lasting relationship – simply ignore the nasty comments from your partner. He says that courting couples tend to ignore negative statements and pay more attention to positive remarks. Once married, this trend often reverses, although couples that remain together into their sixties retain this outlook.
How about gay couples?
…gay and lesbian couples, as well as heterosexual couples that do not marry, hold on to the positive value of courtship better than straight partners who get hitched, he says.
For more information, read The Mathematics of Marriage: Dynamic Nonlinear Models. Are they joking with that title?
How does the porn industry stop digital pirates?
Just as you can download songs, so can you download digital images. Many hackers, for instance, circulate Playboy’s photos around the web. So how do porn services make money?
Porn services will sue people who sell their images for money, but they don’t usually go after users who share files amongst themselves, read this recent story. The industry is already based on churning out large amounts of product very rapidly and very cheaply. By the time one image is being pirated, another set of images is being promoted in any case.
Furthermore many people have very particular fetishes. They don’t just want images, they want images of a very specific kind. (Use your imagination to fill in the blanks here.) Mood matters as much as the specific practice. Often there is no simple way to describe your fetish and get the proper images to download. So you go to a paid site that specializes in your “thing” and you pay them to select and present it. Free pornographic images are common, but selection and context remain as valuable, albeit cheap, services.
Might this work as a business model for the music industry? Songs without titles? How about groups without names, for that matter? Hard to imagine. Groups need monikers to spread their fame by word of mouth. Peers wish to listen to and talk about the same music, whereby most porn viewers are content to have a more private experience. Furthermore the number of popular songs is limited and fans could assign their own names to the material, thus enabling downloading. Producing a good song is harder and more costly than producing a good porn shot, which again brings us back to a reliance on names and reputation. That being said, when it comes to untried bands, which exist in great profusion, the porn model may be precisely the future direction of popular music.
The Economics of Orgasm
I’ve been an economist for so long that I don’t flinch when the paper abstract starts as follows:
“This paper models love-making as a signaling game. In the act of love-making, man and woman send each other possibly deceptive signals about their true state of ecstasy. Each has a prior belief about the other’s state of ecstasy. These prior beliefs are associated with the other’s sexual response capacity…”
Or if that is not enough for you: “In this paper, love is formally defined as a mixture of altruism and possessiveness. Love is shown to alter the man and the woman’s payoff functions in a way that increases the equilibrium probability of faking, but more so for the woman than for the man.”
Here is the full paper. I could go on with quotations, but why don’t we look at the empirical results, drawn from an extensive data set and questionnaire:
1. 72 percent of women admit to having faked it in their current or most recent relationship, for men the number is 26 percent.
2. You are more likely to fake an orgasm if you are in love. “It was the men I deceived the most that I loved the most,” said Marguerite Duras.
3. Being in love and faking are less positively correlated for men than for women. Perhaps men want to look like studs, regardless of the seriousness of the relationship.
4. Women mind less if their partners fake orgasm. (Might some be positively relieved to have it over?)
5. Faking is correlated with age, but in complicated ways. It depends on whether you love your partner, whether you are a man or woman, and how old you are.
6. The more education you have, the more likely you are to fake orgasm. I found this to be the most interesting result.
The author, Hugo Mialon, is on the job market right now and he has a forthcoming co-authored AER piece, plus a revise-and-resubmit at the Rand Journal. His dissertation is “Five Essays on the Economics of Law and Language.”
OK, the orgasm stuff is not his most marketable side, but Hugo seems to be a guy with both ideas and good technical skills. Hire this man. If we had a slot at GMU I would be pushing for him, even though he probably fakes his orgasms.
Thanks to Newmark’s Door for the initial pointer to the paper.
Frequent flyers and the new world of music
Recently I suggested that the on-line music world had yet to settle on a workable business model. Now Gary Leff reports the following new arrangement, whereby you use frequent flyer miles to buy music:
Sony has partnered with United Airlines to introduce another business model
for downloadable music — paying for songs with alternative currency
(frequent flyer miles) rather than money.Details are still being worked on, I understand, and the website is still a
few months from launch, but it looks like Sony and United will offer
consumers both
* the ability to buy songs while earning United Mileage Plus miles, making
United’s loyalty currency a reason for consumers to pick the Sony site over
rivals (and tapping a 37 million member marketing database on top of Sony’s
own lists).
* the ability to pay in miles rather than money, which will be perfect for
infrequent flyers with a small unused stash of miles (a free ticket starts
at 25,000 miles – a song may cost 100 or 250 or 500 miles, price has yet to
be set).
Keep in mind, the recorded music market is a mere $12 billion or so a year. For purposes of comparison, Kraft sold for $13 billion. Southwest has had a capitalization comparable as well. Tobacco advertising for one year is about $11.5 billion. Now I don’t expect the whole music market to be driven by frequent flyer miles. But neither is it obvious that the best way to proceed is to first sue people and then get them to fork over $12 billion into your coffers. The music industry is small relative to the economy as a whole, and relative to advertising as a whole. Here’s hoping for some new and creative solutions to the property rights problem.
Why 99 cents per song?
Apple’s iTunes charges 99 cents for every song downloaded. Why? Is Outkast’s “Hey Ya” really worth no more than a creaky Pat Boone ballad?
Some artists object to this “one price fits all” model. A star may feel it cheapens the value of her wares, or that she simply deserves a higher return.
An alternative business model asks users to donate to the artist, depending how much they like the song. For one service, you can pay as little as $5 but it is suggested that you pay more. The average payment is running at $8.93, though this is a small and self-selected group using the service. In any case not all songs go for the same final price. The service is called Magnatune: We Are Not Evil, check out their web site.
Yet another idea would use an auction system. Listeners could bid for song downloads, with the price determined periodically by supply and demand. We would then expect the songs in highest demand to bring the highest price. Note also that when bands sell their concert recordings on-line, they don’t generally all charge the same prices.
Alternatively, songs may be like books. You charge a low price at first, to stimulate a snowball of fan demand. Bestsellers sell for less, per page, than academic books. (Imagine a professor boasting “Stephen King’s books sell for a mere $6.99; my books sell for a royal $75 a piece.”) In this case the supplier would flood the downloads market with copies, so that the price of the more popular song would be less, not more, despite higher demand.
Different movies sell for the same prices. Either Return of the King or the latest bomb both go for $8.50 at the same theater. This practice has long puzzled me. Perhaps the low price satisfies a fairness constraint, and also helps generate a snowball of fan demand, as with books. It might make more sense to expand the number of screens for the movie rather than raise the price. And hit movies pull people into movie theaters more generally, which spills over into demand for other movies.
The big change may come when downloads are used as advertising. Pepsi is expected to give away up to one million downloaded songs, through iTunes, in connection with the Super Bowl. Coca-Cola may be entering the market as well. Keep in mind that the recorded music industry is small in size relative to corporate advertising budgets. Perhaps corporations will become patrons of music, giving away songs wrapped in an advertisement.
The bottom line: iTunes is just one business model, and it has yet to prove itself. Apple is making money off the hardware, not the songs. Returns will plummet once the hardware business becomes more competitive. It remains to be seen how the downloads market will evolve, but do not expect a mere extrapolation of current trends.
Why can’t you find your favorite song?
… fans who venture onto any of the pay music sites will not find the most popular band ever, the Beatles. They will not find other top-selling acts, such as the Dave Mathews Band, Garth Brooks, the Grateful Dead, AC/DC and the Cars.
They will find that top-selling acts Madonna and Red Hot Chili Peppers sell their songs by the album, but not as singles.
They will find some musicians on one service, but not on others. They will find puzzling choices: Led Zeppelin fans can buy a 47-minute spoken-word biography of the band online, but no Zeppelin songs because the band has not licensed them for sale on the Internet.
Why are these potential gains from trade not being exploited?
1. Some artists are holding out for a higher price or better terms. This can mean either a better cut for the artist, or the artist does not like the “all songs for 99 cents” model of iTunes.
2. Many artists feel that selling songs on an individual basis takes them out of proper context or cannibalizes sales for the album.
3. Pre-1998 contracts do not specify Internet rights to the songs. Assignment of Internet rights can require the underlying contract to be renegotiated.
4. Renegotiations must involve both the performer and the songwriter.
5. Often the relevant parties cannot be found or are otherwise difficult to deal with. One executive said: “You can be sure the heirs are a son and daughter who aren’t talking to each other and one of those two is getting divorced.”
Here is the full account. You will find stories of high transaction costs, poorly defined property rights, and stubborn holdouts, all the classic predictions of institutional failure theories.
The bottom line: Selection, not just price, remains a big advantage for non-legal downloading. If iTunes and related services are to make it in the long run, they will need to offer near-universal choice of music.
Google gets better
Enter an airline flight number — for example, “united 80,” — and the popular search engine will provide links to reports on that flight’s status at Travelocity.com and Fboweb.com, including maps showing its progress.
Type an area code into the search box, and you’ll be pointed to a MapQuest.com map of the general region that area code covers. A U.S. Postal Service package tracking number yields a link to a delivery-status page at the Postal Service’s Web site. A vehicle identification number will call up a page describing the car’s year, make and model type.
Or you can type in a universal product code number — minus the dashes, but including any tiny numbers appearing to the far left or right under the bar code — and Google will look up the product’s full name, then generate a list of Web sites selling the item or providing other information about it.
Check out Google’s own explanation, or this article from The Washington Post. The real question is where search engines are headed, and whether Internet gatekeepers will get more or less centralized. I have already predicted that Google essentially has peaked, though I will confess I used Google to find that very link.
Marriage Mathematics and Political Change
John Gottman has spent decades studying how married couples interact. His most striking finding is the tendency of couples at risk of divorce to have markedly different interaction styles. His recent book, The Mathematics of Marriage, summarizes his observations of married couples and presents a parsimonious model of marriage (see here for Slate’s review). The highlight of the research is that couples where the dominant mode of interaction includes criticism, contempt, defensiveness and stonewalling are very, very likely to divorce. Successful marriages involve a great deal of mending and reworking of the relationship. The mathematics links some theories about emotions and interaction to this observed pattern.
What I find interesting is the implication for thinking about politics. Let’s assume that political order is a sort of “marriage” between state and citizen. At least from the perspective of the citizen, it’s a relationship that can be broken, if warranted. This is a premise of many normative theories of revolution – the citizens have a right to a new government if they feel the written and unwritten rules have been violated. Unfortunately, what we know about exactly how this happens – moving to abandon the social contract – is sketchy at best, although political scientists and sociologists have a hunch that it involves some combination of repression of the population and a de-legitimizing of the government, which itself might have multiple causes.
Gottman’s approach to studying relationships offers a useful way to think about these issues. Gottman’s point is that there may be varying sources of the emotions that destroy marriages, but the road to divorce usually starts in the same place – once spouses have learned certain interaction strategies, they create hard to change feedback loops. Similarly, governments and populations that learn certain strategies for interacting with each other probably set up hard to break cycles leading to long term stability or perpetual crisis. The nice thing about Gottman’s analysis of marriage is that the math predicts stability or decline, and not much in between – a non-trivial prediction. The same prediction for states is that states tend to be on a tough to change road to constant crisis (like in Africa and the Middle East) or stability (like in the US). Switches from one path to the other should be infrequent and difficult, which seems to describe the world pretty well.