Saturday, October 31, 2009

Back in Uganda


This is the view of downtown Kampala from my room, which is on the 5th floor of a shopping complex. Wonderful.

Thursday, October 15, 2009

Economic development, Sherlock Holmes and medical science

These are the three things I think a lot about these days, and they're more related than you might think.

In the latest NYRB, Jerome Groopman (article not yet available online) reviews two new books on the intersection of medical science and sleuthing.

This statement often comes as a shock to most people when I say it: most medical doctors are not scientists. They are detectives and artists. Not only do most not participate in medical trials, but if you ask your medical doctor about the trials and studies the drugs they are prescribing are based on, few can tell you any details and will find the questions silly. Most see their profession as determining the problem, and after that, they simply follow the rule: when faced with x problem, prescribe y, then observe more and, if faced with z, prescribe some w.

Not that there's a problem with that. Division of labor works well here, and few people could keep all of this information straight in their head. Groopman has done a fine job in the past exploring the detective nature of medical doctors, and he deserves to be read for that reason. He also has written about the problems modern medicine faces for doctors to spend the necessary time on each patient, and for this argument he falls short.

In the NYRB article, Groopman laments the dwindling time doctors have to spend on patients, in large part becuase of the way their time has, in his mind, become commoditified by the system. Each minute a doctor spends with a patient is billed, and so most are forced to spend just enough time to figure out whats wrong, or close enough to it, then move along quickly.

Groopman is correct that doctors time has been commodified, but its not the system he should be blaming, its doctors. America pays its medical doctors far more than anywhere else in the world, mostly because this is what doctors have demanded in recent years. The commodification of doctors time is because doctors are very expensive, and thats mostly the fault of doctors.

But what does this have to do with economic development? The tools that medical science uses are very close to the tools I use in my work: the scientific method, close scrutiny of any claim, a bit of artisty and a dash of detective work.

I follow medical studies and, 9 studies out of 10 that I see quoted in the newspaper, I look up and observe the methodology closely. Many studies make more errors than you might think, and a close scepticism of any reported result is important. Even more importantly though, the tools of medical science have a lot to teach economists still.

Sunday, October 4, 2009

E coli outbreak

The NYT has a great video about the consequences and causes of an E Coli outbreak recently that left one woman paralyzed. I've blogged previously about the lack of proper government oversight of meat producers, and this is more evidence that we need more regulation.

Monday, September 28, 2009

Roman history

I recently finished reading Julius Ceasar's commentary on The Gallic War which was so interesting I quickly read through Goldsworthy's excellent biography Caesar: Life of a Colossus. Its one of the most interesting and balanced portrayals of the man ever written.

For anyone interested in Roman economic history, I also suggest this collection of research: The Ancient Economy, or a new post over at Brad DeLong on the end of Roman agriculture slavery.

Monday, September 21, 2009

More odd links

The posting will improve shortly, I promise. In the mean time, enjoy these odd links:

Sunday, September 13, 2009

Strange links

  • National Geographic has a wonderful interactive map of Manhattan where you can make direct comparisons with before and after the city was developed.
  • Its not the appendix's fault for being useless, its our modern world.

Saturday, September 12, 2009

Is the OLPC dead?

The OLPC has been getting a lot of heat lately in the blogsphere.

First, Timothy Ogden (HT PSD Blog) argued that the effect of computers on education is abysmal (in developed countries at least), and that proven education boosters, like deworming, are much more cost effective ($4/year versus $200). I will argue below that neither of these means much.

Then UN Distpatch argued the dream is over, with too many cost overruns and poor delivery. The founder makes a reply here, arguing there are some positive stories coming out. He also notes that the $400 netbook did not exist before the OLPC.

This has a personal note for me. For my neices first birthday, I bought her an OLPC through Amazon. I paid $400 in total, one for her, one to be donated.

Yes, I know 1 is too young to have a laptop. It has saved my sister plenty of headache though as my neice no longer tries to sit on her lap while my sister's on the computer. Instead my neice just pretends to be her mother with a mostly indestructible chew and drool proof computer. It also means that my neice will know how to type as soon as she knows how to spell, and computer programs, including the Linux opperating system, will not seem mysterious to her.

In the developed world, this matters a lot. Teaching at two colleges in the US has left me baffeled at the lack of computer literacy amoung those that finished highschool. I wouldn't be surprised if 90% of people from 18 to 25 in this country did not have a working knowledge of Excel, and only basic word skills.

But does it matter in the developing world? Maybe. Computers are not a necessity, but they are very much sought after. I get orders to bring 2-3+ laptops every time I go to Uganda. People want to be connected to the world, and they are desperatly seeking to learn how to do just that. Internet cafes and training centers are all over northern Uganda now.

I don't know about OLPC specifically, but I think it's too early to call the dream of cheap computers to the developing world dead. The impact of computers on education is more than the literature cited by Ogden. The effect of computer training in a developing country means a lot more than it does in a developed country because there are no other options in poor places. The general equilibrium effects could be massive, and this could take a long time to see.

And just because a computer is more expensive than another option is a meaningless comparison if we don't know the individual effects. In any case, why think of it as an either/or problem? Given the lost cost of deworming, I would suggest bundling them together in places that need the medicine.