Wednesday, December 15, 2010

U.S. census results

The NYT takes advantage of GIS to generate a map based on data from the Census Bureau for "Every City, Every Block", showing colour-coded patterns in race & ethnicity, income, education, and housing and families. The data is apparently from 2005-2009, so this is not the latest official census (at least, I don't think so). But it is extremely interesting — even apart from specific colour codes, it gives quite an illuminating view of population concentration and distribution across the country.

The link is to the default map, which is a view of greater New York City broken down by racial and ethnic groups, but you can zoom out and scroll over to any other part of the country, and click on the "View More Maps" button to see the other mapping options.

Tuesday, December 14, 2010

Conflict minerals - naming and shaming

The Enough project today released a report, "Getting to Conflict-Free", about corporate action on conflict minerals. The report is described in a blog post by Aaron Hall at the Enough project's website.

The most famous campaign to reduce the profitability of the mineral extraction industry in conflict zones is of course the one on conflict diamonds, which over time became visible enough to be referenced in several hit movies (including Blood Diamond, with Leonardo DiCaprio, and the James Bond installment Die Another Day) — a sign of high public awareness, if not always knowledge :-).

Unfortunately, the minerals that help fund the conflict in Eastern Congo are less photogenic as well as harder to track than diamonds. Gold, tin, tantalum, tungsten, coltan — all can be melted or ground down, so that physical tracking becomes impossible. The only way to track them, then, is to have a completely transparent supply chain which accounts for every gram of gold, tin, etc. that corporations use.

The Enough project's report is a first attempt to investigate how well corporations are doing in moving towards such transparency and accountability. The report ranks major technology companies on "percentage of progress toward responsible sourcing on conflict minerals". The top performer is HP, at just 32%, followed by Intel, at 24%. Worst on their list are SanDisk and Toshiba.

Although the blog post gives the key take-away findings, the report is worth reading for its discussion of the many challenges associated with achieving "progress toward responsible sourcing". (The report is not very clear, unfortunately, on the ranking methodology, but more detailed rankings can be downloaded from the Enough website here.)

Variation in educational achievement in the U.S.

Very interesting article in the Atlantic monthly by Amanda Ripley: "Your Child Left Behind". The article references a study, just published in EducationNext, by Eric Hanushek, Paul Peterson, and Ludger Woessmann which compares national educational performance (in mathematics) in several countries around the world to state-level educational performance in the United States (a more extensive version of the study is here).

The motivation for the research was that it is well-known that the United States as a whole performs poorly on such international comparisons (for the latest example, see here), but the United States is so large that it might well be the case that individual states within the U.S. outperform most countries. The researchers' finding, unfortunately for the U.S., is that the best performing state, Massachusetts, comes in 17th, followed by Minnesota in 20th place, with no other state in the top 25.

The study is interesting (and disturbing) in its own right, but it is also a nice example of bringing good social science methods to bear on questions that many people have impressionistic beliefs about but which are often difficult to test systematically. [As Hanushek notes in Ripley's article, one explanation for poor U.S. performance he often hears in his hometown of Palo Alto is "We're a very heterogeneous society — all these immigrants are dragging us down. But our kids are doing fine.")]

Numerology for idiots, EU version

While doing some research on the composition of the European Parliament I ran across a conspiracy theory so silly it is surprising anyone believes it. It is an outdated theory, but it continues to survive on the internet (nothing ever really dies on the internet, after all).

Here's the claim: seat 666 at the European Parliament is empty, waiting to be filled by the Antichrist (666 being the number of the beast). One source of this notion is Ian Paisley, the radical anti-Catholic Ulster politician, in an article dating to 1999. (Interestingly, 11 years earlier Paisley had denounced Pope John Paul II as the Antichrist when the latter addressed the European Parliament in its previous building). Other sources are here and here.

The allegedly incriminating list of seat allocations (with an empty spot at 666) dates to 1999, when seat 666 may indeed have been unallocated. So, however, were quite a number of other seats: the European Parliament in 1999 had 626 members, but occupied a debating chamber with seats for 750 (since expanded to 785) legislators (its previous home had seats for 600 delegates, so no number 666 there). This means that 750-626 = 124 seats were unoccupied, about 1 in 6.

The claim that the number of one particular unallocated seat is anything other than coincidental is of course idiotic. But it continues to survive, even though the actual seat has been occupied for years now (first, apparently, by Sajjad Karim, a British Liberal, who when asked about the significance of his seat number in 2006 was completely flummoxed :-). Current seating plans for the Strasbourg and Brussels hemicycles can be found here.

Still, a search on the terms "european parliament", "seat", and "666" yields about 108,000 results according to Google, and plenty of websites confidently assert that the seat has never been filled. (On the other hand, the European Parliament is just one part of the EU — a search on "European Union" and "antichrist" gives about 841,000 results!)

Thursday, December 9, 2010

More crowdsourcing, London

In the UK, Parliament is considering increasing university tuition fees. Students are protesting, as one might expect. Some from University College London (UCL) are maintaining a blog about it here. They have also jury-rigged their own crowd-sourcing setup with Google Maps: see their input form & the map itself.

Given the obvious promise and appeal of using Google Maps for such purposes, one wonders whether Google is working towards incorporating a similar set-up as one of Google's own products (of which they have more and more). It is probably not a coincidence that they're already a partner of Ushahidi, the non-profit interactive mapping initiative I mentioned last month.

The illusion of secrecy

Earlier this week, students at Columbia's International and Public Affairs school were advised not to discuss any WikiLeaks material online. The recommendation came from a Columbia alumnus who now works at the State Department, and who argued that ""Engaging in these activities would call into question your ability to deal with confidential information, which is part of most positions with the federal government."

As the linked article notes, Columbia's administration quickly reversed itself on this issue, fortunately. Nonetheless, the episode reminded me of Solzhenitsyn's First Circle, in which Soviet scientists must lock away their top-secret research documents in a safe every night, so the imperialist spies won't be able to steal them — except this top-secret research consists mainly of Western science journals, since Western science is far ahead of the Soviets. (It's been years since I read Solzhenitsyn, so apologies for a somewhat hazy description)

It is similarly absurd to pretend that the material leaked by Wikileaks is still in some way secret or confidential. Fortunately for Columbia students, their administration is a little less delusional than the Soviet regime was; still, their instinctual reaction is telling.

Seasonal migration as foreign aid

David McKenzie at the World Bank offers an intriguing suggestion about "The Most Effective Development Intervention We Have Evidence For": under certain conditions it appears that seasonal migration trumps any other intervention. New Zealand has a short term migration program which, apparently, comes "with minimal displacement of native workers, and overstay rates of less than 1%."

The blog posting links to the actual research paper, as well as some other related links. Comments on the posting note that it is likely to be much more difficult to find political support for such an intervention than it is for foreign aid. Moreover, keeping the overstay rate low requires active involvement and monitoring on the part of the home as well as the host country. The United States currently has a seasonal worker visa program (H2A), but it is not clear how much monitoring takes place — I was unable to even find estimates of overstay rates, for example (but plenty of law offices willing to advise those who have overstayed on what to do next :-).

Sunday, December 5, 2010

Aid and public opinion

The general political wisdom in Washington DC is that the public does not like foreign aid. However, it has long been known that the public does not actually know much about the U.S. aid program.

For example, surveys consistently find that while a majority of Americans would like to see the U.S. aid budget cut, a majority of Americans also wants the U.S. to give much more aid than it does. This apparent contradiction arises from the fact that most Americans believe the U.S. gives far more aid than it does.

The pattern has been familiar since 1995, when the Program on International Policy Attitudes found that the median estimate of the U.S. aid budget was 15% of the total federal budget, at a time when the actual outlay was a little under 1%. The median preferred share of the federal budget to be allocated to aid was 5% (i.e. on average 1/3 of what people thought was the current level but 5 times the actual current level).

A follow-up study in 2001 found "no decline in the public's extreme overestimation of the amount of the federal budget that goes to foreign aid." Indeed, in this second study the median estimate was that 20% of the federal budget went to aid, while the median desired allocation was 10% of the budget. Meanwhile, the actual aid budget had not changed much and continued to be just under 1%.

(The follow-up study also noted that in a different survey, 66% of respondents said that t00-high foreign spending was a "major reason the economy is not doing better than it is.")

Little has changed in the past decade, as poll results just published by World Public show. Indeed, the overestimation of U.S. aid continues to increase: now the median estimate of the share of the federal budget that goes to foreign aid is 27%, with the median desired allocation still at 10% (actual aid volume continues to be close to 1% of the federal budget).

The survey also found that the lower one's education attainment, the higher the median estimate of aid's share of the budget — "Among those with less than a high school education the median estimate was that foreign aid represented an extraordinary 45 percent." This may help account for the erroneous belief in some circles that it should be fairly easy to cut the federal budget in difficult economic times such as these. In practice, of course, even a complete elimination of the foreign aid program (something few people advocate) would have only a small impact on the deficit.

Saturday, December 4, 2010

Good links

Just cleaning out some bookmarks:

Very funny Mad Lib satirizing ICT4D arguments (see earlier post here).

Egypt has a second round of elections this weekend (the first round was last week). Good article by Gwynne Dyer on "Why elections are a charade in Egypt under Hosni Mubarak", and some background by Eric Trager on "Egypt's Machiavellian electoral gambit" (from just before the first round).

Critical article by Marlise Simons in the NYT a week or two back about problems the International Criminal Court has encountered in pursuing its first trial, that of Thomas Lubanga, who stands accused of committing various war crimes, including the conscription of child soldiers.

Funny article in the NYT by David Segal about how there is no such thing as bad publicity on the web for ecommerce businesses.

Preventing genocide

Interesting review by Richard Just at the New Republic of Geoffrey Robinson's new book “If You Leave Us Here, We Will Die”: How Genocide Was Stopped in East Timor. The book's title is an obvious allusion to Gourevitch's classic We Wish to Inform You That Tomorrow We Will be Killed With Our Families: Stories from Rwanda, only in East Timor's case genocide was (largely) avoided.

Just notes that "the ability to ignore human rights, to regard mass killing coldly and cynically, is not the province of any one political party or even any one ideology," a conclusion that echoes Samantha Power's argument in her devastating A Problem from Hell. He also points out that, in the end, preventing genocide was not particularly costly, either in economic terms or in political (good-)will. But it does require that the regime sponsoring a (potential) genocide be vulnerable, as Indonesia was in the late 1990s.

The obvious question is whether the Sudanese government is similarly vulnerable, should things get out of control in Southern Sudan after the scheduled referendum on independence a few months from now. Richard Just thinks not, and I am inclined to agree. But that does not mean genocide cannot be prevented; only that it requires a stronger commitment (and some advance preparation, for example in terms of lobbying at the UN) on the part of those who would do so. An opportunity, perhaps, for the Obama administration to improve its standing on human rights? Hope springs eternal!

"Psychological waterboarding"?

It turns out that post-9/11 prisoners at Guantanamo were routinely given a controversial malaria drug (mefloquine) known to carry a serious risk of side effects including paranoia and ideas of suicide. Seton Hall's law school has published a report discussing the case (press release here). Mark Denbeaux, director of the Seton Hall Law Center for Policy and Research, comments: "At best it represents monumental incompetence. At worst, it’s torture."

If the administration of the drug did not arise from incompetence, the report suggests that not only would this policy "likely satisfy the legal definition of torture" (p. 4), but also, if used as part of a program of enhanced interrogation, "would be the psychological equivalent of waterboarding." has a parallel article on the issue, which quotes Remington Nevin, a Major in the U.S. Army Medical Corps as calling the policy "pharmacological waterboarding." Nevin has been doing a lot of research on the adverse effects of mefloquine, so he knows of what he speaks.

I buy the argument that, if deliberately administered for non-medical reasons, the policy can be characterized as torture. But there are many types of torture, and I am not convinced that waterboarding is the physical "instrument" whose effects are most similar to this medical "instrument". One sort of suspects that both Nevin and Seton Hall latched onto the comparison because of the resonance of the term waterboarding in the public discourse.

Either way, this is rather disturbing news, and it is important to get some answers about the motivations for the administration of mefloquine. Of course, given the Obama administration's reluctance to pursue such answers in other torture-related areas, there is little reason for optimism.

P.S. Interesting discussion of the Army dispensing mefloquine to its own soldiers in the Army Times here. Given the available evidence at this point, I'd put the odds of incompetence vs. torture at about 50-50.

Thursday, December 2, 2010

(Information) technology and development

The Nov/Dec issue of the Boston Review has a great forum on "Can Technology End Poverty" about the promise and the limitations of the clunkily named ICT4D sector (information and communications technology for development). In the lead article, Kentaro Toyama argues that, contrary to the hype, ICT4D can only magnify the effect of motivated and capable human beings working on development — it cannot substitute for or replace them. (For a similar point about the promise and pitfalls of crowdsourcing, see earlier post here).

A number of prominent academics and participants in various ICT4D initiatives offer their responses. Archon Fung's argument about the possibility of exploiting the socio-economic biases of technology is particularly intriguing. The final contribution in the forum is Toyama's, who responds to the comments from the other contributors. Toyama is not optimistic about Fung's suggestion that technology be specifically designed to help improve the lives of the poor, but thinks he may be on to something in suggesting that technology be used to support public goods.

I think Toyama is right in pointing out that "inevitably, the rich, skilled, and socially connected will make better use of [technologies] than the rest." But this need not be as big a problem as he seems to believe it is. Fung's argument, I believe, is that one can develop specific technologies that the rich, skilled, and socially connected would not have a strong incentive to use, because the technologies supply something (possibly a public good, but not necessarily so) that they do not need (either because they already have it, or simply because it is not one of their needs).

The challenge, of course, is that this requires a lot more thought than simply throwing laptops or cellphones at a developing country problem. But promoting such an approach would be a good complement, I believe, to Toyama's central prescription: that, "when deciding how to allocate resources between technology and human capital, [we] invest first in the factor that is most lacking," which, in a developing country, will often be the latter.

Wednesday, December 1, 2010

Repaying sovereign debt

Ireland's EU/IMF bailout plan places the full burden of the crisis on the borrowers, rather than making the lenders pay something too (something that should have been judged both fair and prudent). For both economic and political reasons, this is going to make successful resolution of Ireland's crisis that much more difficult. Barry Eichengreen has a scathing piece in Germany's Handelsblatt newspaper on this today (translation here; the German title is far better than the English one, though).

Given the high likelihood that Ireland will need to return to the negotiating table for some new solution, concerns about the handling of potential similar problems in other eurozone states are mounting. Willem Buiter, Chief Economist at Citigroup, has a good economic overview of the problems facing European sovereign debtors in particular: Sovereign Debt Crisis Update. Figure 34 in this article is particularly interesting (or worrisome).

Key points from Buiter's conclusion: "There is no such thing as an absolutely safe sovereign," "the distinction between public and private balance sheets can become blurred in a crisis." Not grounds for great optimism, in other words.