The measles vaccination success

Sometimes, skepticism about philanthropy in developing countries is in order. Here's an interesting comment (via Chris Blattman) from Friends of Ethiopia, where Todd Johnson asked his friend Sammy (an Ethiopian entrepreneur) about the effect of NGOs:

"Africans don't see a reward system in place for being entrepreneurial. In fact, they view it as a matter of survival, not an opportunity to lift themselves out of poverty. Rather, what they learn at a very early age, is that in order to make good money, they should learn to speak English incredibly well and then maybe, just maybe, they can get a job driving for an NGO. In a few years, if they play their cards right, they might be able to land an NGO job as a project manager and even advance further."

Sammy's point was simply this. As a struggling businessman creating new start-ups, he could not compete with what NGO's were paying for some of the best and brightest. And even worse, he said, "by the time the NGO's are done with them, there isn't an ounce of entrepreneur left."

This makes it all the more important that we understand which forms of philanthropy are most efficient, with the greatest impact on welfare for the least expense and interference with the local economy. And I don't think that there is any better candidate than basic public health campaigns, particularly measles vaccination.

Relative to its impact, mass vaccination is perhaps the least economically invasive intervention imaginable. It costs very, very little for each vaccinated child (generally less than $1), and it's spread evenly throughout the population.

Meanwhile, the successes have been overwhelming. From 2000 to 2008, the Measles Initiative reduced mortality by 78 percent worldwide and 92 percent in Africa, with an estimated total impact of 4.3 million lives saved. Given that the initiative has invested $670 million in vaccinations to date, the cost per life saved has been roughly $150. (How's that for cost-effectiveness?) Even this, however, is an overestimate of the lifetime cost of the program, since the costs of vaccination are incurred upfront while the benefits dribble out over time.

You can see the decline in measles deaths in this chart from the World Health Organization's annual report on vaccine-preventable diseases:

Although measles vaccination has been a great success to date, more action is necessary. Measles eradication is possible, but it isn't in our near-term future, and in the meantime vaccination will need to continue for the declines in mortality to remain. This campaign—along with insecticide-treated bed nets, anti-tuberculosis drugs, and all the other low-hanging fruit in public health—is where our philanthropy should be overwhelmingly concentrated. I don't understand why it isn't.

Convergence in Eastern Europe

Curious about growth success stories in the 2000s (pre-Great Recession), I looked at which countries experienced annualized growth of over 5% in per capita GDP—adjusted by purchasing power parity—in the period from 2000 to 2008.

In my data, hoisted from the World Bank's invaluable World Development Indicators database, there were 38 such countries. 10 of them—Equatorial Guinea, Azerbaijan, Turkmenistan, Angola, Kazakhstan, Tajikistan, Chad, Mongolia, Uzbekistan, and Sudan—were driven in large part by natural resources, and don't reflect any general success in economic growth. Meanwhile, of the remaining 28, fully 14 are in Eastern Europe. Most of the population in this sample is in the remaining 14, especially China, India, Vietnam, and Ethiopia, but the astonishing uniformity of the growth boom in Eastern Europe is still very interesting to consider.

This map shows countries in Europe broken down by their level of per capita GDP growth from 2000 to 2008. There is remarkable geographic concentration: none of the already rich nations of Western and Northern Europe grew at a fast pace, while the Easternmost nations of the old Soviet bloc were the strongest performers of all. Poland and the Czech Republic, meanwhile, didn't grow as quickly as their counterparts to the East, but they started from a relatively high level.

In this region, growth is far from an elusive mystery; decent institutions, stability, and proximity to rich nations seem to have been enough to trigger massive advancement in living standards over the past decade. (And the apparent success of Belarus, the last full autocracy in Europe, even adds a wrinkle to our understanding of "decent institutions".) Investors can now be confident that their assets won't be destroyed through war or political turmoil, and low barriers to trade and movement allow knowledge to spread from advanced nations to developing ones.

Granted, the notion that countries will "converge" in overall productivity clearly can't explain every aspect of growth across the world. Many poor countries remain mired in seemingly intractable poverty, with mutually reinforcing economic and political dysfunction. Yet it seems to have been relatively accurate in Eastern Europe over the last decade.

In the past two years, of course, many of Eastern Europe's formerly high-flying economies have suffered from a severe slump; Latvia and its 20%+ unemployment and collapse in GDP come to mind. This seems, however, to be cyclical rather than structural—countries on the European periphery had prices that needed to come down relative to Western Europe, which was either impossible (for countries using the euro), not attempted (for countries maintaining pegs to the euro or dollar), or the precursor to a crisis of fiscal confidence (for countries like Ukraine that devalued and needed IMF intervention).

It remains unclear how quickly countries will emerge from the wreckage—most are returning to positive growth, albeit not full capacity. Regardless, there's no sign that the decline itself reflected something unsustainable about Eastern Europe's growth, rather than the unfortunate double blow of a monetary adjustment and worldwide recession. Convergence in Europe remains very real.

The logic of charter cities

Via Mark Thoma, I see that Aditya Chakrabortty is savaging Paul Romer's charter cities proposal in the Guardian:

Trouble is, the idea stinks. With little track record in dealing with poor countries, Romer has come up a grand scheme for lifting Africa and Asia out of poverty. What they need to do, he argues, is give up a big chunk of their land to a rich country. Policy experts from Washington can take over a patch of Rwanda, and invite along GM and Microsoft and Gap to come and set up factories. Poor countries give up their sovereignty in return for the promise of greater prosperity...

You think this is colonialism? For Romer, that "kind of emotion . . . can get in the way" (see what he did there? You have emotions; the elite economist has evidence). Sure, the poor people living and working in these new charter cities wouldn't necessarily have any democratic privileges such as the right to vote, but they could vote with their feet. And in the meantime, the Africans or the Asians would get the undoubted benefit of all this huge western expertise.

I agree that charter cities are a political longshot, but this isn't fair at all. Poor countries absolutely should not give up a "big chunk" of land to Western powers, and Romer isn't proposing that they should. Instead, the idea is to cede a small slice of currently undeveloped land to foreign administration. In this scenario, no one is forced to give up their rights—you're only under foreign administration if you voluntarily choose to move to the affected territory.

Just as importantly, this minimizes interference with the existing power structure. No regional administrator or bureaucrat is removed from office in favor of Westerners; foreign influence is confined to a previously empty patch of land. The idea is to avoid the opposition from entrenched interests that so often cripples attempts at reform. If this logic sounds unduly speculative, keep in mind that a similar idea has been a critical part of economic reforms around the world, including the spectacularly successful Special Economic Zones of mainland China.

In the end I still think charter cities are unlikely, if only because no Western nation has much incentive to be the benevolent administrator. But before we start crying about resurgent imperialism and ceding a "big chunk" of land to the West, we ought to understand what Romer is really proposing.

Statistical discrimination

In response to posts on credit scores in hiring by Yglesias and myself, Matt Zeitlin points out that discrimination can be perfectly rational (but nevertheless deserving of censure):

As we all know, labor market discrimination by race is both immoral and illegal, but it happens, and it doesn’t just happen because people are evil, but because there is an empirical element. Tim Harford had a great piece from a while back in Forbes where he talks about taste-based and statistical discrimination. (note: the piece is here.) The former is when employers just don’t like ethnic minorities and are straight-forward bigots and the latter is discrimination that happens because certain racial and ethnic identities are used as markers for other traits that matter to employers. Both, of course, are morally wrong and illegal, but the latter is hard to root out because, absent regulatory or legal pressure, employers will benefit from statistical discrimination, or at least they won’t suffer from it. From the perspective of the qualified black guy who can’t a job, there’s no difference between the two types of discrimination: he’s still being judged based on his group identity as opposed to his individual qualities.

For reasons stated in follow-up posts, I'm skeptical about the particular case of credit scores, but as a general matter this is spot on. I'm particularly glad that Matt brings up statistical discrimination, because it's a subtle but deeply important feature of labor markets. Not all racial discrimination is the result of uninformed prejudice; some of it, sadly, is rational for a profit-maximizing business, even though it remains destructive and morally distasteful.

The effects of statistical discrimination go beyond the obvious unfairness of judging an individual by his group identity. As economists have realized, there's the possibility that an entire racial group will be stuck in a bad equilibrium. If an employer believes that very few people of a particular race have invested in their job skills, he's likely to discount a good interview or test result from a member of that race as a fluke. Then, realizing that they're unlikely to get a good job anyway, people in that race will not invest in skills, confirming the employer's belief. It's a simple but lethal self-fulfilling prophecy.

Some observers, like Bryan Caplan, argue just the opposite—that statistical discrimination may be self-reversing because it raises the return to education. Caplan points out that the return to education is empirically higher for black students than white ones; he surmises that statistical discrimination makes "counter-stereotypical" behavior particularly valuable. To an extent, I think he's right. The "bad equilibrium" story doesn't seem to lower the returns to obtaining easily observable credentials like college degrees. Perhaps this is because college degrees are such good indicators of competence that they overwhelm negative racial perceptions, or because in a world that erects massive social and economic barriers to obtaining a good education, managing to get a degree as a black student is much more impressive than it is for most white students.

Yet I think that Caplan's analysis is fundamentally incomplete, because it's too quick to equate skills with formal educational credentials. Suppose you're a student for whom college is out of the question (too much of a financial burden) but completing high school is likely. Even conditional on obtaining a high school diploma, there is plenty of variation in the effort levels you can choose. To the extent that it depresses job prospects for your entire racial group, statistical discrimination may also decrease the returns to verbal and quantitative skills that aren't captured in a simple diploma, and hence your incentive to acquire them.

And, of course, "skills" aren't just taught in school. There's a very important set of vocational, interpersonal, and general "noncognitive" skills that are equally decisive in determining job performance. Since these skills are particularly hard to measure in an objective way, they're the most susceptible to racial stereotyping and statistical discrimination.

In fact, this is where I suspect most of the problem lies. Certain categories of people are thought to lack the intangible skills necessary for a work environment. Of course, there is no innate reason why this should be true, but the perception damages incentives (and changes culture) enough to make itself true. And because there is no diploma for "having the right stuff," no formal accreditation can unravel the bad equilibrium.

It's a tough problem, and it's worthy of policymakers' attention.

It's already possible to discriminate against the unemployed!

I wasn't really satisfied with my explanation of why most arguments for banning the use of credit scores in hiring don't hold water. To refresh, Kevin Drum's main complaint with using credit scores is that it creates a "Catch-22" for unemployed workers: they have low credit scores because they're unemployed, which makes it hard to get a job, which means that their credit scores will deteriorate even more, and so on.

Now, there are two possibilities here (not mutually exclusive):

1. Employers are interested in information contained in credit ratings other than employment status (e.g. financial responsibility).
2. Employers are interested in employment status.

In my post I addressed possibility #1, and I pointed out that employers presumably understand that a low credit rating from a currently unemployed person carries different implications than a low credit rating from someone with a steady income. If businesses are interested in the "character qualities" weakly proxied by a credit report, they'll adjust for employment history in interpreting the numbers.

Possibility #2, meanwhile, isn't really an issue. If employers want to discriminate against applicants who are currently unemployed, they are already able to do so, and I don't think anyone proposes a change in policy here. So as long as you grant that businesses aren't stupid, Drum's specific gripe doesn't make much sense.

This doesn't mean there is no problem at all with using credit ratings for employment. The most worrying possibility would be that credit scores are an unusually good proxy for some attribute we don't think should be involved in hiring, like medical history. If this turns out to be true, we should certainly consider placing restrictions on their use in evaluating job candidates. For now, however, it's not clear that looking at credit scores is any more unfair than looking at college degrees, personal connections, or any of the innumerable other features of the hiring process that are inequitable yet widely accepted.

The astonishing story of BU and Seragen

When reading a very interesting 2008 article on college endowments from the Journal of Economic Perspectives, I came across this nugget about Boston University's disastrous investment history:

The perils faced by schools unfamiliar with alternative investments can be illustrated by the experience of Boston University. The school’s administration established its own venture capital subsidiary; in contrast, most leading universities have done the bulk of their alternative investing through outside partnerships, rather than trying to undertake these activities directly. The fund invested in Seragen, a privately held biotechnology company founded in 1979 by scientists affiliated with Boston University.

The result of this investment can be described based on Seragen’s filings with the U.S. Securities and Exchange Commission. As part of the school’s initial investment in 1987, it bought out the stakes of a number of independent venture capital investors, who had apparently concluded after a number of financing rounds that the firm’s prospects were unattractive. Between 1987 and 1997, the school, investing alongside university officials and trustees, provided at least $107 million dollars to the private firm. (By way of comparison, the school’s entire endowment in the fiscal year in which it initiated this investment was $142 million.) While the company succeeded in completing an initial public offering, it encountered a series of disappointments with its products. By late 1997, the University’s equity stake was worth only about $4 million. (emphasis mine)

That's right: Boston University threw an enormous fraction of its endowment at a single biotech startup, and lost almost all of it. This didn't kill BU—still a prestigious university with a prime location—but it did set the university back enormously.

How did the university make such an ill-fated decision in the first place? As a 1998 New York Times article explains, president John Silber (a "brilliant" former philosophy professor) decided that the startup—based on research by BU scientists—had a wildly successful future ahead of it, and pressured the university into investing massive amounts of venture capital through its endowment.

You might ask why John Silber—as opposed to, say, actual venture capitalists with actual experience in the biotech industry—considered himself qualified to make this decision. This question apparently never occurred to him. Nor did the prudence of investing a large fraction of a prominent educational institution's money into a single company in perhaps the riskiest business category in existence—a biotech startup.

Eventually, of course, the economics caught up with him. Intelligently administered university endowments do invest in venture capital, but only as one part of a carefully diversified whole. These venture capital investments, in turn, are spread between multiple funds, which are themselves composed of investments in many different companies and sectors, and managed by professionals with experience in the relevant fields. There was no reason to expect that Silber's intuition would outperform established venture capitalists in calculating a risk-adjusted price for Seragen's future income stream. And Silber certainly had no reason to repeal the established laws of risk management, under which a large, undiversified investment in a single security is unquestionably foolish.

Apparently, even many "smart" people like former BU president John Silber lack the foggiest understanding of investment strategy. It's borderline criminal that he had this power in the first place.

The surprising usefulness of middlemen

One common reaction to my argument about the inefficiency of a-la-carte cable is to stress that cable bundling is really just a holdover from a technologically backward era. In the past, it wasn't possible to sell stations individually. Now it is. In fact, very soon there will be no need to rely on cable providers at all. From a technical standpoint, there's no reason that television should be carried via dedicated "cable" or "satellite" companies, rather than simply piped over whatever internet connection a home happens to use. To many observers, the case seems clear: now that we have the technology to sell stations individually, the middleman is nothing more than a lumbering, soon-to-be-extinct monopolist.

But my point isn't that there is some technical need for Comcast or Time Warner to bundle (or even sell) content. Instead, I'm arguing that this practice is economically efficient. Someone has to pay the fixed costs of creating content; once these costs are paid, there's zero marginal cost to offering channels as widely as possible. I might only want cable to watch sports on ESPN or TNT, and you might only want news channels and cartoons, but the costs of creating all this content wouldn't magically go down if we had the option to purchase stations individually. Instead, assuming that the overall amount of content didn't decrease, on average we'd be paying the same fee for fewer options.

By bundling stations together and allowing us to access more content for the same price, the much-hated cable middleman has actually improved our choices. This arrangement is the accidental result of past technical limitations, but that doesn't mean it's not efficient.

No credential is perfect

Looking a little more closely at Yglesias's post on credit checks for job applicants, it's astonishing how many of the comments take the following form:

"But credit reports also reflect X, Y, or Z. It's not fair to judge applicants based on that!"

And so does practically everything else that businesses use to make hiring decisions! A college degree, for instance, is hardly a perfect signal of your work ability. Among other things, it can indicate:

1. Family wealth.
2. Whether you were lucky enough to live in a state with affordable, high-quality public universities.
3. Your sense of responsibility as a 20-year-old.

Yet I don't think I've ever heard anyone react that we should ban the use of college degrees in hiring. Sure, we'd like to devise alternative credentials that are useful to businesses and more accessible to the entire population. We'd also like to make college more affordable. But in the meantime, we quite reasonably allow businesses to hire based on educational credentials—even when those credentials carry grotesque inequities—because a college degree also provides useful information about an employee.

This is true even when a college degree isn't directly relevant. Most jobs that nominally require college education don't really demand anything beyond basic reading and analytical ability, and middle-school math (if that). We still accept that higher education will matter in hiring.

In a few particularly egregious circumstances, we acknowledge that some form of discrimination in hiring is so odious and societally destructive that it cannot be allowed—for instance, discrimination on the basis of race or gender. Forgive me if I don't think that credit score discrimination is clearly deserving of the same treatment. And if you think that it is, you have to explain why your argument doesn't extend to education, or work history, or the countless other imperfect and often unfair proxies that employers use to judge the suitability of job candidates.

Hail Matthew Yglesias

His reaction to calls for banning employers from checking credit scores:

"Drum and the Slacktivist both call for making this illegal, and at first blush I’m inclined to agree. But at the same time I try to adhere to the principle I outlined here and resist the urge to call for regulating the business practices of private firms when the issue isn’t pollution or some other case where the externalities are clear. After all, it seems like either this credit check business is a sound business practice (in which case allowing it is making the economy more efficient and ultimately building a more prosperous tomorrow) or else it’s an unsound business practice (in which case competition should drive it out). But this really does seem fishy and I’m curious to read more on the subject."

It's hard to deny that there is some legitimate business value to checking credit scores. On average a lower credit score does indicate a weaker sense of financial responsibility, although there is certainly plenty of unrelated individual variation as well. (Isn't this true of any metric for evaluating employees?) Still, Drum and the Slacktivist think that use of credit scores in employment is creating a "Catch-22" whereby unemployed workers have bad credit—because they're unemployed—and thus can't get a job, which only worsens their credit and job prospects even further.

Of course, you'd imagine businesses would realize that weak credit scores from an unemployed applicant don't carry the same meaning as weak credit scores from an applicant with a steady work history, and evaluate them accordingly. In this light, Drum's point is really just "I think that employers are so stupid that they won't realize credit partially reflects employment history." Maybe that's true, and if after a careful review of the evidence we find that employers are incapable of using credit scores in a remotely rational manner, then we can consider regulatory action.

In the meantime, however, it's deeply destructive to run around making wild assumptions about businesses' incompetence. As Yglesias says, that's the kind of liberal I'm not.

Better data: a modest proposal for education reform

One of the tragedies of education reform is that the evidence on so many important questions is so thin. Are private schools more effective on average than public schools? It's impossible to say: most commonly cited studies derive their results from ghastly regressions that attempt to isolate the effect of schooling by tossing in countless demographic "controls." The unobservable qualities of students—motivation, parental engagement, and so on—are ignored, despite their presumably pivotal role in driving students' school choices.

Randomization provides a far more reliable source of statistical evidence, but for obvious reasons it's not common in education (with a few notable exceptions). Yet there is one relatively frequent situation where randomization is the standard accepted practice: a lottery for students applying to an oversubscribed charter school. And in a recent working paper, Angrist et al. use the lottery for admission to a KIPP charter school in Lynn, Massachusetts to identify the school's impact on students.

The conclusion? While "KIPP Lynn applicants have (pre-lottery) math and reading scores that are 0.39 and 0.44 standard deviations below the state average and somewhat below the LPS average," the authors estimate that:

Our results show overall reading score gains of about 0.12 standard deviations for each year a student spends at KIPP, with significantly larger gains for special education and LEP students of about 0.3-0.4 standard deviations. Students attending KIPP gain an average of 0.35 standard deviations per year in math; these effects are slightly larger for LEP and special education students.

In other words, a single year of KIPP education can almost completely close the math achievement gap between Lynn students—primarily low-income minorities—and students in the highest-scoring state in America.

This isn't conclusive, of course: although randomization is the best way of identifying causal effects, there are still a few possible weaknesses in the data, and we only have results for one school. By any standard, however, this is tremendously encouraging news. If we could replicate this result—or even a result half as strong—in KIPP charter schools across the country, the impact on education policy would be profound.

This research should be easy. Anywhere in America where a lottery is held to determine entrance to a charter school, records should be kept and linked with students' performance. If education researchers across the country could access a dataset with large-scale randomized variation in school assignment, our understanding of what really works (and what doesn't) in charter schools would be vastly improved. I can't think of any reform that offers such overwhelming potential benefit at such low cost.

In defense of Fahrenheit

We have weird measurements in the United States. 1760 yards to a mile, 3 feet to a yard, 12 inches to a foot; 16 ounces to a pound; 4 quarts to a gallon, 2 pints to a quart, 16 fluid ounces to a pint. They're ludicrously hard to convert, and I know nothing about a "pint" except that it's a measurement of beer.

Amid all this mess, people often criticize Fahrenheit. Granted, Fahrenheit seems less "natural" to people accustomed to powers of ten; why set freezing at 32 and boiling at 212 when you can use 0 and 100 instead? But there's really no practical disadvantage, because we never need to do arithmetic with temperature. Miles, yards, feet, and inches are annoying because they're a pain to convert, and they don't cleanly match up with our volume units in the way that a cubic meter translates to 1000 liters. None of these concerns are relevant for temperatures—we don't convert them, and we don't add them together.

In the end, judging whether Fahrenheit or Celsius is superior is really just a matter of convention and taste. While Celsius is slightly easier to remember, I don't think that recalling two short numbers poses too much of a challenge. Indeed, Fahrenheit has the advantage of matching up better with the temperatures we encounter outside: the range from 0 to 100 degrees Fahrenheit roughly corresponds to the most common temperatures on Earth, while temperatures above 50 degrees Celsius almost never happen.

So... absolutely, abolish the mile and retire the gallon. I welcome our metric overlords. But don't touch Fahrenheit—it's just as good as any other scale.

China has (almost) the highest fertility rate in East Asia!

Take a look at the 2009 CIA World Factbook figures on total fertility rate (summarized in the second table here) and you'll see a startling pattern at the bottom of the list:

223. Japan: 1.21
224. South Korea: 1.21
225. Taiwan: 1.14
226. Singapore: 1.09
227. Hong Kong: 1.02
228. Macau: 0.91

Out of the six lowest-fertility countries in the world, five are located in East Asia and the other is majority Chinese. In fact, despite its one-child policy, China has the highest fertility rate, 1.79 children per woman, of all countries in East Asia except North Korea (which probably shouldn't count).

It isn't too surprising that China is higher: declines in fertility are heavily correlated with economic development, and despite its overwhelming success over the past three decades China remains far poorer than the six countries listed above. But the shockingly low fertility rates in the East Asian nations most culturally comparable to China offer a strong hint of China's demographic future. Shanghai, China's richest city—and our best example of what a richer China 20 years hence might look like—has a fertility rate of only 0.88!

As I've pointed out before and believe even more strongly now, population projections like the UN's World Population Prospects—which assumes that all fertility rates will eventually converge to 1.85—are likely to be very, very wrong about China's demographic outlook...

Zero marginal cost and Microsoft Office

In my last post, I made the case that a la carte cable wouldn't necessarily save consumers' money: instead, the price of each individual channel would skyrocket, the market would unravel, and consumers would find themselves paying a similar amount of money for a far less impressive set of choices.

This results from a feature common to most kinds of electronic content: zero marginal cost. When you're buying a bundle of cable channels, you're not paying for the cost of distributing each channel to you—as long as cable is already installed, that's virtually zero. Instead, you're supporting the fixed costs of creating the content in the first place. If a la carte cable leads each consumer to subscribe to only 10% of channels, each channel will be forced to raise its price to support the cost of programming (and quite possibly go out of business in the process). In the end, the funding for the content has to come from somewhere, and a thinner base of subscribers means either less content or higher fees for the same content.

To better understand the economics of zero marginal cost, it's useful to look at another example: the software industry. In the age of the Internet, software is virtually free to distribute—it's a classic case of zero marginal cost. Yet commercial software certainly isn't free. Software makers need to recoup their costs somehow, and the obvious way is to charge consumers for using the product. What happens, then, when a company offers multiple related products, each with a large share of the market? It could bill consumers a la carte, pricing each piece of software separately. Or it could bundle the software, charging a single price for a large package.

Microsoft Office springs to mind. As it turns out, Microsoft offers the different components of its Office suite both individually and as a bundle. The prices, however, are quite revealing. Buying Word 2010 standalone costs $150, which is the same as the price for the entire Office Home and Student 2010 suite, including Word, Excel, PowerPoint, and OneNote.

Why doesn't Microsoft offer a cheaper version of Word for home use? Because it has nothing to gain. Many home consumers, presumably, are only really interested in Word. If Microsoft charged less for Word alone, these consumers would happily snap up the option. But it doesn't cost anything for Microsoft to throw in Excel, PowerPoint and OneNote as well, and even the consumers who mainly want Word will be happier with access to every program.

In fact, the only reason Microsoft sells different bundles at all—rather than offering every conceivable feature in one version of Microsoft Office—is that it wants to price discriminate, especially between home users, small businesses, and high-powered business clients. Sophisticated businesses are willing to pay much more for Microsoft Office than home users. Hence Microsoft sets aside one cheap version restricted to home users by the terms of its license, and also differentiates between "Microsoft Office Home and Business" and "Microsoft Office Professional" to extract revenue from the businesses who are most willing to pay. But while it offers these different tiers of software, it never provides each individual program as a cheap a-la-carte option. The individual prices always add up to far more than the price of the bundle.

Advocates for a-la-carte cable almost always use the same basic argument: why should I have to pay for channels I don't watch? You could just as easily adapt this rhetoric to apply to Microsoft Office: why should I have to pay for programs I don't use? Why doesn't Microsoft offer a home version of Word for less than the price of Office? But if you understand the economics of zero marginal cost, you'll realize the fallacy. Offering an additional channel to a cable subscriber doesn't cost a network anything. Including more programs doesn't cost Microsoft anything. By bundling, they're able to deliver more content for the same price—and that's great for consumers.

More choice is less choice: The strange economics of a-la-carte cable

I've rarely seen an abuse of economic logic as dramatic as the push for a-la-carte cable. The economics of zero-marginal-cost services like additional cable channels are extraordinarily unintuitive, and otherwise sensible prescriptions like "more choice" do not necessarily improve consumers' actual options. In fact, debundling of cable is quite likely to lead to worse choices for a similar amount of money.

To see why, let's forget about cable for a moment and imagine a city with 20 parks, strewn throughout various neighborhoods. Suppose congestion isn't an issue; each park is large enough that an additional visitor doesn't materially diminish the experience of the others. The city park system is required to recoup its costs by selling yearly passes.

At first, the park system sells yearly passes to all 20 parks in a single package, for $100. All 100,000 residents purchase the passes, allowing the parks to cover $10 million in annual expenses. Life is good.

But residents start to grumble—after all, they don't visit all the parks. Why should they have to pay for the maintenance of parks they don't visit? In response to this eminently reasonable complaint, the park board offers a new pricing scheme, where passes are sold for each park individually. At first it sets a low price for each pass: $20 a year. Sadly, after a year—and an unpleasant report from the county treasurer—the price for each park is increased to $100. Residents pay the same amount of money for access to only a single park.

Horrors! What happened? As it happens, each family was mainly interested in visiting the park nearest to its neighborhood; although occasionally it might want to visit parks elsewhere in the city, few other parks were worth the $20 annual pass. On average, families purchased only two passes: one to the park closest by, and one to another popular park across the city. Needless to say, this wasn't enough to cover the annual expenses of the park board: its revenues declined by 60 percent!

To compensate, the board initially voted to raise the price for a pass to $50 a year. Its consultants, however, ran several surveys and discovered that at a price of $50, the average resident would only buy passes to 1.5 parks. This, of course, forced a further price increase to increase revenue, leading to another drop in estimated demand. After several excruciating meetings with the consultants, the board realized that the minimal price sufficient to balance its budget was $100. At this point, the average resident would purchase only one pass: the sole worthwhile investment was a pass to the nearest neighborhood park.

This outcome was unambiguously worse for city residents. After all, they derived some benefit from being able to visit other parks—just not enough to justify an extra $100 for any one pass. They now paid the same amount of money but received far fewer choices in return. A policy that purported to promote choice actually destroyed it.

Substitute "niche programming" for "neighborhood parks" and you have the essence of cable economics. I hardly care about MTV; take away Comedy Central, however, and I'll be steamed. Many viewers feel the opposite. In an a-la-carte world, we'd each choose to buy only our favorite channels. Their prices would skyrocket on a thinner subscriber base, to the point we'd be paying nearly as much as we did before. I'd be just as poor financially, and I wouldn't even be able to indulge myself with Jersey Shore—surely a bad deal all around.

Is this analysis of cable economics a little one-sided? Of course. There are legitimate arguments for why a-la-carte cable might be beneficial; maybe it would make content providers more responsive to viewer interests, or shake up some cozy monopolies. But the typical argument for a-la-carte cable is still dreadfully wrong. In a world where additional content has zero marginal cost, the freedom to choose the content you buy has a good chance of leaving you with less choice in the end.

The environmental vapidity of Mitch Daniels

Last month's glowing Weekly Standard profile of Mitch Daniels has one sadly revealing excerpt:

Beyond the debt and the deficit, in Daniels’s telling, all other issues fade to comparative insignificance. He’s an agnostic on the science of global warming but says his views don’t matter. “I don’t know if the CO2 zealots are right,” he said. “But I don’t care, because we can’t afford to do what they want to do. Unless you want to go broke, in which case the world isn’t going to be any greener. Poor nations are never green.”

Sigh. As far as visible pollutants are concerned, Daniels is mostly right: rich nations have managed to control emissions that cause localized problems like smog and acid rain, while a miserably poor city like Dhaka suffers from stifling pollution. But with carbon dioxide, it's the exact opposite: superficially cleaner rich nations are the worst offenders, as demonstrated by this Wikipedia map of per capita emissions statistics from the International Energy Agency:


See the deep red country in the Western Hemisphere? That's us—the United States has higher emissions per capita than any other large nation. Poor countries, meanwhile, emit next to nothing: Bangladesh, for instance, has less than 2% of the United States's per capita emissions.

Daniels' ignorance on this question—not to mention his casual dismissal of "CO2 zealots"—reflects a sad failure to understand the difference between carbon dioxide and "pollution" in general. Daniels and his ilk see the crowded, smoggy capital of a developing country and assume that it's playing a large role in carbon emissions—after all, it looks polluted. In reality, countries like the United States, Canada, and Australia contribute to climate change in an overwhelmingly disproportionate way. Until they take leadership to resolve the global collective action problem, climate change is likely to continue unabated.

Could Joe Arpaio's dad be a legal immigrant today?

If you've ever had the misfortune of listening to Maricopa County sheriff Joe Arpaio—last caught pledging to do a "sweep" for undocumented immigrants the instant the new Arizona law goes into effect—you might be forgiven for assuming that he was anti-immigrant.

But don't worry—Joe Arpaio isn't really anti-immigrant. He's just anti-illegal immigrant, as he hastens to explain:

"My mother and father came from Italy legally and made a good life for themselves. They worked hard but they were here legally. I have no problems with people coming into the U.S. to become citizens, that's what made this country great, but you must come into the country legally -- not illegally."

Ah, the wonders of legal immigration.

It turns out that Arpaio's dad, a native of Lacedonia, Italy, came to the United States via Ellis Island in 1923. Fortunately for him, he came in 1923, because just a year later Congress passed the Immigration Act of 1924, one of the most appalling and xenophobic acts in its history. The Act of 1924 was the first permanent limitation on immigration to the United States, the direct ancestor of our current system—and, by design, it made legal immigration by most Italians virtually impossible.

Congress had already passed the "Emergency Quota Act" in 1921, which limited immigration from any country to 3% of its current nationals in the United States. The 1924 act was visibly harsher, pushing the ratio down to 2%. But the new law also made a subtler change: it redefined the relevant population by using the census results from 1890, not 1920. Why? Because 1890, you see, was before Southern and Eastern Europeans began coming to America in large numbers. Before there were so many Italians...

Nicola Arpaio was a very lucky man.

Now imagine that a young Mexican laborer wants to come to the United States, much as the elder Arpaio did. Like the 22-year old Arpaio, he doesn't have immediate family in the United States (the route for almost all legal immigrants from Mexico), or possess any special education or skills. How can he immigrate legally?

The diversity visa lottery doesn't apply; Mexico sends too many family-based immigrants to qualify. The H-2A and H-2B temporary worker visas allow laborers to come for a few years, but sends them back afterward, never offering any path to permanent residence. Meanwhile, lacking advanced education or training in a skilled occupation, our aspiring immigrant is ineligible for every type of employment-based green card but one: the "other workers" subcategory of the EB-3 visa.

The fact that the worldwide cap for these "other workers" is 5000 each year might seem discouraging enough. But there is another complication. Through mid-2001, undocumented immigrants in the United States were allowed to apply to adjust their status to either a family or employment-based green card. This was a good policy in principle, of course—undocumented workers should have the opportunity to right their status. Congress, however, didn't adjust the total number of green cards to meet this new source of demand. Predictably, the waiting list for other workers has been backlogged at 2001 ever since.

This also means that only a fraction of the green cards in this category are given to new, legal immigrants. Indeed, from the State Department Visa Office's report for fiscal year 2009, we can determine exactly how many "other workers" green cards were given to residents of Mexico, rather than immigrants already in the United States adjusting their status.

The answer? 23.

You have to hand it to Joe Arpaio. Yes, his father legally immigrated to the United States—just one year before the force of anti-Italian bigotry made it almost impossible for his countrymen to follow. And now, if a young Mexican resembling Nicola Arpaio wants to come to the United States, Joe's response is that he should stay in Mexico, get in line, and hope that he's one of the lucky 23 granted passage from a country of 110 million.

After all, he wouldn't want to be illegal.