Ryan C Briggs

Gender balance at the Olympics

Fri, 12 Aug 2016 10:00:26 -0400

A few days ago I tweeted that I thought that “the ratio of male to female Olympians would be a decent cross-national indicator of women’s rights.” I was encouraged to actually check, and this post briefly summarizes the results.

First, I scraped data on participation at all Olympic games from 1896 to 2012 from Sports Reference.com. I’m only going to comment on the summer games here. Before we start, an important note: This is a silly throw away project and I in no way verified the quality of the Sports Reference data. Caveat Emptor.

To start, let’s look at how the gender balance of each team’s Olympic athletes changes over time. Each grey line is one of 225 countries that participated in at least one Olympic games. The black line is the cross-team average for each games. Note that the black line doesn’t tell you the overall fraction of women competing in each games because it counts every country equally rather than every athlete. Female participation is trending up and it begins a real climb in the mid 1980s.

In a very weak sense, this seems consistent with my hunch above: we know that from 1896 to the present women’s rights have improved and this seems to be tracked by the fraction of women in the Olympics. However, this is really weak evidence. Let’s compare across countries to get a better sense of how female Olympics participation and women’s political participation correlate.

This next graph only uses data from the year 2012. The y-axis is the fraction of all athletes are that are female in each of 205 countries in the 2012 Olympics. The x-axis is the fraction of seats held by women in each country’s national parliament. The blue line is an OLS fit line and the grey area is a 95% confidence interval. There is absolutely no relationship between these variables.

Hmm, so that is dispiriting. I really expected a relationship, but there is just nothing. In another surprise, I compared the same Olympic data against GDP per capita. Everything covaries with GDP per capita, right? Wrong. Again, there is just nothing here. Countries that are richer, and countries with more female parliamentarians, are no more likely to have more women competing in the Olympics than poorer countries or countries with more male-dominated politics.

My analysis finished here, but I had posted the web scraping code on twitter and asked other people to analyze the data. Happily, Adam Chilton did just that. I’ve posted his graphs below.

He is looking at a gender equality index and he gets basically the same finding as me for 2000. More gender equal places do not have more women on their Olympic teams. However, he finds that there was a relationship between gender equality and female Olympic participation in 1976. I have no clue what to make of this changing pattern. My main take away from all of this was that my original hunch was wrong. It appears that the ratio of male to female Olympians is not a decent cross-national indicator of women’s rights. The null hypothesis wins.

If anyone wants to replicate my work above, you can download the code to scrape the data, clean the data, and generate the graphs here.

Isaac is a lucky boy

Thu, 01 Oct 2015 15:20:14 -0400

About two weeks ago, my first child, Isaac was born. We spent a standard two days in the hospital and then went home. After about 5 hours at home, in his third day of life, Isaac developed a fever of 103 and we went straight to the ER. After a few days of tests, we learned that Isaac’s blood and kidneys were infected with enterococcus and that he probably had bacterial meningitis. This kind of thing is incredibly uncommon and no one really knows how he got infected. He was just incredibly unlucky…

…except that he really wasn’t.

For four years now, I’ve been teaching international development. One of my first questions to my students is: “If you knew nothing about yourself (e.g. gender, race) or where you would be born, when would you like to be born?” (I think I got this question originally from Angus Deaton?) The students can choose any time from the dawn of humanity to the present day. After a bit of thought, students quickly converge on “now” as the right answer. I like to stress to them that “now” is the right answer not only because the world is wealthier and more people have iPhones. Women are treated far better today than at any time in the past, more people have political freedoms today, and far fewer children die.

That final point felt important but abstract until a few weeks ago. Child mortality feels incredibly real now. Thankfully, in being born now and in America, Isaac is damn lucky. By just about any comparison, child mortality is the US is very low. It’s about ten times lower than much poorer countries like Malawi, and it’s also about ten times lower in America today than in the 1940s. Presently, about 10% of children in America with bacterial meningitis die. In the 1970s, the rate was about 50%.

Isaac is doing well. He is 13 days into a 21 day course of antibiotics and in his 16th day of life. He is eating well, he is above his birth weight, he has had no seizures, he has no signs of cranial inflammation, and he is doing regular baby stuff. He is expected to make a complete recovery. Untreated, he would have almost certainly died. He is damn lucky to have been born now in America.

I’ve known these statistics (well, aside from the ones on meningitis) for some time, but they carry a lot more emotional weight now. Most babies in Isaac’s situation still die. Development is the process of making that not so.

Meningitis is pretty rare, so if the thought of babies dying moves you then probably the best thing you can do at the moment is to donate to the Against Malaria Foundation. Giving out insecticide-treated bed nets is one of the most reliable ways of reducing child mortality.

PPS (April 3, 2016)

At about 6 ½ months old, Isaac is doing very well. He is at or above average on pretty much everything that we can measure. His first month of life was terrifying for us, but he really was incredibly lucky. Thanks to everyone for the well wishes.

Keep it simple

Fri, 26 Jun 2015 18:22:51 -0400

Chris Blattman recently argued that a lot of aid spent on skills training did little to help poor people. He also asked for “comments and criticisms,” and so Tom Pepinsky followed by calling for more ethnographic research aimed at identify specific constraints on households and more political economy research on institutions. The contention is that we need to focus more on the smallest stuff (households) and the biggest stuff (national institutions). Ken Opalo agreed and emphasized the importance of working with local elites in ways that make development goals compatible with their incentives. (My paraphrasing might be misrepresenting the authors, so do read all three posts)

Now, ethnographic analyses to actually understand what is going on and careful analyses of institutions and elite interests are all Good Things, but I’m skeptical of their ability to make aid work better and I’m very skeptical of any outsider’s ability to manipulate them.

I’m thinking out loud right now, but my reading of the literature on aid effectiveness suggests to me that aid works best when donors either do simple things over and over (polio vaccine) or produce global public goods (research to boost crop production). I think this emphasis on simplicity is more contentious than it sounds. To be clear: I think that institutions and local political battles are the most important things keeping poor countries poor. I also think that these things will not be changed by aid programs, and I think trying to influence them is probably a waste of money. In that sense, I think that aid cannot solve or probably even improve on the main factors that influence long-run development. I’m pretty sure that that should not be the job of aid. Instead, I’m for targeting aid to specific things that help poor people and where impact can either be measured or where we have very good evidence that the “treatment” works. This includes a range of health measures (e.g. vaccines, bed nets) or cash transfers. There probably is room to tweak these simple interventions based on a better understanding local context, but even then I worry about the ability of actual people to carry out these tweaks well. My hunch is that “tweaks for local context” without rigorous and iterative analysis and some kind of incentive-changing feedback loop probably won’t produce the learning that would make aid more effective.

The Drunk World Bank Twitterbot

Fri, 24 Apr 2015 12:36:53 -0400

I made a twitterbot that posts pseudo-random tweets formed from samples of text from the five most recent World Development Reports. My reasons for writing the bot, in order of importance, were: I wanted to learn something new in R, I thought that the results would make me laugh, I had a little bit of time to kill at ORD after the MPSA conference, and finally, I thought that the resulting development babble might make us think a little about how development experts communicate. Here is one example tweet:

It recognized that legal equality was not revealed as shown by the expansion of the variation in crime rates in agrarian economies.
— Drunk World Bank (@DrunkWB)
April 24, 2015

The nerdy details on what I did are below.

The Problem

My goal in writing the bot was to produce text that was recognizably from a development agency like the Bank but was also random enough to be funny. It had to have enough syntactical structure to sort of make sense, but it had to reach this point without any actual knowledge of grammar. This is a perfect application of Markov chains, and this entire project was inspired by episode 20 of Partially Derivative when Chris and Jonathan were talking about drunk beer reviews. As I write this, I realize that I must have lifted the idea of calling the bot “drunk” from Greg Reda, so hat tip to Greg.

I read Greg’s excellent post on Markov chains and the Markov chain wikipedia page and then googled around to find someone who had made a similar bot in R (most are in Python). I couldn’t find a Markov chain twitterbot written in R, but I did find a blog post that gave me useful sample code for posting text from R to twitter. Aside from these sources, it seemed that I had to solve the problem myself.

My Approach

I had no real experience with Markov chains before this project, and at first they sounded complicated. However, like a lot of mathematical ideas, they are remarkably simple once you get them. The fundamental idea is that we can make a chain of predictions where each prediction for the next turn (t+1) is informed only by information present in the current state (t). In my case, I knew that I wanted to chain together words that had local structure so that any set of three consecutive words would make sense. I also wanted to do this based on the language of the Bank, so I started by copying and pasting all of the text (excluding bibliographies and boilerplate text) from the five most recent World Development Reports into a plain text file. I then cut this text file into all consecutive three word segments (word triplets) and counted the frequency with which they occurred in the text. Most occurred once, but some were more common.

To start the chain I draw one word triplet randomly, but with the probability of being selected being proportional to the triplet’s use in the WDRs. That three word triplet is the start of our tweet. Perhaps it is “Tailor technology to”. Next, find the subset of all word triplets whose first two words are the same as the last two words of the current tweet (technology to). In the current example, that set is small and includes only “technology to democratize”, “technology to local”, and “technology to make”. I then draw another tweet at random (but using the probability weights) from this subset and add the last word onto the current tweet. Perhaps it now reads “Tailor technology to democratize”. This process is repeated until you have a string that is long enough to be a tweet. In this way, we can build up a tweet where each additional word is selected only using information on the two trailing words. The result is the local structure that makes most tweets at least partially intelligble.

I’m leaving out a lot of details like: cleaning the WDR text to avoid things like page numbers, inset boxes of text, or tables; little tweaks to get the tweet to be the right length or avoid word repetition; or how I scheduled the posting to twitter; but that is the general idea.

Improving my approach

The naive approach above worked, but the tweets felt too random. Reading them, I could see that part of the problem was with the start and end of tweets. Tweets ending in words like “and” seemed off. My first solution was to include a list of words that would be cut from tweets before posting, but this was a hack and the list of “bad ending words” kept growing.

I realized that a much better solution was to scan the WDR text and isolate the word triplets that start or end sentences. I could then use those triplets as bookends for the rest of the chain. Finding these starting or ending triplets is a fairly simple problem, as the start of English sentences have capital letters that follow a period and a space, and all sentences end in a period and a space. After some pattern matching, I produced three tables of word triplets. One had the word triplets that start sentences and their frequencies, one had the middle words and their frequencies, and one had sentence ending words. I then worked these new tables into the matching procedure above. I would start with a draw from the starting triplets file and then fill it out with the middle words before capping it with a draw from the ending words table. I’m rather proud of the results:

Formal arrangements to share information on disaster response can worsen it.
— Drunk World Bank (@DrunkWB)
April 23, 2015

On the role of ideas in economic policy making

Mon, 25 Aug 2014 14:00:16 -0400

I was not aware of the recent success in the Guatemalan sugar industry. Since the 1980s, sugar production in the country has increased at an annual rate of about 7%, far outstripping similar countries. Rising production has occurred in tandem with better working conditions.

A new paper (ungated here) by Alberto Fuentes stresses the role that a “small team of managers motivated by Elite Solidarism, an interpretation of the Vatican II Catholic Social Doctrine” played in bringing about the changes. I haven’t read it carefully yet, but it strikes me as a good example of how ideas can have independent causal power. A more general version of this argument is well expressed in Dani Rodrik’s recent paper “When Ideas Trump Interests.” The two together seem like a good match for teaching about constructivism and the role of ideas in economic policy making.

Political aid targeting

Thu, 14 Aug 2014 13:00:00 -0400

I was listening to EconTalk the other day and Chris Blattman was being interviewed. The whole interview is good and I suggest you listen to it. After agreeing with almost everything Chris said, I found something to quibble with right around the end. Here is the quote:

“What weak states in Africa have—and will learn—to do is learn to target that money better towards supporters. […] The danger point with aid is when states that are receiving it are strong enough to use it effectively for their own political ends. I think right now a lot of states in Africa don’t have that organizational capability in the same way that states in Latin America have developed it. I think that is when it will become trickier.”

The comment reminded me of a recently published paper in World Development, Why there Should be No Political Foreign Aid Curse. In the paper the authors argue that there shouldn’t be an aid curse (where aid hinders democracy) because aid is less fungible, more conditional, and less constant than oil revenues. Both the authors of that paper and Chris seem to believe that right now, at least in Africa, donors are doing a pretty good job controlling where aid goes and therefore limiting the influence of local politics on aid allocation. Two recent papers (one of them by me), as well as other anecdotal evidence, have made me think that this view is probably wrong because it underestimates the power of recipients to direct aid for political goals.

The first paper is How Aid Targets Votes: The Impact of Electoral Incentives on Foreign Aid Distribution by Ryan Jablonski. Looking at Kenyan data from 1992 to 2010, Ryan shows that aid from the African Development Bank and World Bank was disproportionately targeted to areas that supported the sitting President. He also shows that aid helps Kenyan presidents win elections.

The second paper is Aiding and Abetting: Project Aid and Ethnic Politics in Kenya by me. While Jablonski looks at aid from two donors across all sectors, I examined project from all donors in only two sectors. I also picked a time (late 1980s and early 1990s) when Kenya and its donors were on very bad terms. This is the time when donors cut programme aid to Moi pending democratization. During this tense time, donor project aid still went to the parts of Kenya that were part of Moi’s ethnic base. It seems pretty clear that, at least in Kenya, aid is targeted according to a local political logic.

The biggest issue here is generalizing from Kenya to other African countries. Kenya certainly has a higher organizational capability than many other African states, and this suggests that maybe Kenya is fairly unique in its ability to direct aid to politically important areas. However, Kenya in the 1990s was also unique in that it faced a small group of fairly homogenous donors (not much Chinese lending) who were not fans of the Kenyan regime. The fact that political aid targeting took place under these unlikely circumstances suggests to me that a lot more countries might be capable of this kind of targeting today.

Research shortcuts

Mon, 11 Aug 2014 11:00:29 -0400

Over the last few years I’ve developed a few tricks for getting articles and books quickly and I think that they might be of interest to other people. I’ll use examples from the two places where I’ve taught (AU and VT), but I imagine the basic ideas generalize to other institutions. I’ll also give examples of how to automate these tricks using TextExpander. The first trick modifies journal URLs to get fast access to articles and the second modifies library search URLs to get fast access to books. Both are only of use to people that have access to university libraries.

URL modification for journal articles

The first shortcut modifies journal URLs so that you access the article through your library. In my experience (n=2), this involves adding your library’s proxy URL after the end of the journal’s host name. In the case of AU, this requires adding “.proxyau.wrlc.org” and in the case of Tech, it involves adding “.ezproxy.lib.vt.edu:8080”. Both should ask for a login and then allow you to download the article.

For example, perhaps you might want to read Chris Blattman et al’s article “Generating Skilled Self-Employment in Developing Countries: Experimental Evidence from Uganda.” The URL for that article is “http://qje.oxfordjournals.org/content/129/2/697.” To access it through Virginia Tech, you would go to http://qje.oxfordjournals.org.ezproxy.lib.vt.edu:8080/content/129/2/697.

Adding text to URLs manually like this is tedious, so I altered a TextExpander script (from here) to automate the process. The script is below. My solution is clunky, but it has been consistently working for me. To change it for your institution, you would need to add your institution’s proxy URL in place of the Virginia Tech one (in bold below). You can find the URL for your library by visiting a journal webpage from one of your library’s databases. You can also just ask a librarian and they should be able to help.

#!/usr/bin/perl -w

#Initialise
use strict;
my($text);

#Get the text from the clipboard
$text =`pbpaste`;

#Insert the library string (by replacing .com with .com and the string
$text =~ s/.com/.com.ezproxy.lib.vt.edu\:8080/ig;

#Insert the library string (by replacing .org with .org and the string
$text =~ s/.org/.org.ezproxy.lib.vt.edu\:8080/ig;

#Output the text
print $text;

If you assign that code snippet to a shortcut in TextExpander then when you see an article that you like you just copy the URL, type the shortcut, hit enter, and download the pdf.

Fast and precise book search

Book searching should be done by ISBN numbers, but in my experience libraries make this time consuming by requiring you to click “advanced settings” every time you want to search. Rather than doing that, you can instead just paste in the library’s search URL and then add the ISBN. For Tech, the URL is “http://vt.summon.serialssolutions.com.ezproxy.lib.vt.edu:8080/search?t.isbn=” and for American the URL is “http://american.summon.serialssolutions.com/search?t.isbn=”. Your library hopefully has a similar URL. To find it, go to the advanced search settings for the library and then search by ISBN. Then look at the URL and grab all of it until the part that specifies an ISBN. Again, doing this by hand gets old and we can automate it with TextExpander. For Tech, you just add the following string to TextExpander as a shortcut: “http://vt.summon.serialssolutions.com.ezproxy.lib.vt.edu:8080/search?t.isbn=%clipboard%key:enter%”. When executed that string will take the ISBN number from the clipboard and search the Tech library for any book with that number. In my work, I usually find the book on amazon, then copy the ISBN from the amazon page of the book, then run the TextExpander shortcut.

I use these tricks daily and they have saved me a lot of time. Hopefully they’re of use to others too. If you improve on these ideas (the journal hack needs work), please send me a message @ryanbriggs.

Waka Waka

Mon, 19 May 2014 01:26:00 -0400

I don’t follow football or FIFA news, but I recently found out that the the new World Cup song isn’t so popular (see here or here). The response on twitter was an uptick in #WakaWaka, a callout to Shakira’s anthem from the 2010 World Cup.

Not only is Shakira’s song (with South African group Freshlyground) better, but tonight I found out that it also has a really interesting history. The chorus samples the song Zamina mina (Zangaléwa) from the Cameroonian band Golden Sounds. The band was formed in the 1980s by members of Cameroon’s National Guard.

It is hard for me to track down the actual history of the song, but my best reading of this is that the original song came about after WWII as an homage to African infantry. Scattered sources that I couldn’t verify also claimed that the song had anti-colonial themes. In the 1980s, Golden Sounds picked up the song and made the awesome version above. Shakira may have heard the song in Colombia, where it was apparently quite popular.

A sample of the song also appears at the end of the Vampire Weekend cover of the Springsteen song I’m going down.

Talk at the Foreign Service Institute

Wed, 05 Mar 2014 14:15:20 -0500

Earlier today I had the pleasure of visiting the Africa Area Studies program at the Foreign Service Institute where I gave a talk on the basics of foreign aid to Africa. I and a great time and got to share overly honest slides like the one below.

The students are in an intense, short program but seemed up to speed on major issues and asked great questions. I especially enjoyed one linking possible issues of aid dependence (and related governance concerns) to oil dependence.

The slides for the talk are available to download here.

Africa Rising and Aid Dependence

Fri, 28 Feb 2014 12:43:09 -0500

I’ve recently had a few people ask me about the idea that “Africa” is growing quickly and that aid is increasingly becoming irrelevant to development in Africa. While the “Africa Rising” narrative is a pleasant antidote to the strong pessimism of the 1990s, it can lead to wishful thinking about the decline in the importance of aid to Africa.

The simple fact its that relative to local resources, Africa still receives a lot of aid. The figure below shows net ODA to countries in Africa over five year intervals. The amount of aid is expressed as a fraction of local government spending.

First, note that we don’t have data on central government spending in many African countries. Second, note that aid plays a very large role in many countries. Of the 21 countries reporting data in 2010, donors spent about as much in the recipient country as did the local government. Keep in mind too that the local government spends money on both investments and consumption (wages of civil servants) while donors focus on investments only. In many African countries, donors still vastly outspend recipients when it comes just to investments.

Here is a similar figure that examines Kenya and Ethiopia over time. These are countries that we think of now as relatively well functioning with comparatively good governments, and again aid plays a pretty large role in both countries.

Donors spend about 1/5 as much as the Kenyan government and donors frequently outspend the Ethiopian government. Note also that even in these countries we do not have a compete time-series.

Africa’s recent economic growth is excellent, but it will take more than a few years (or decades) of economic growth to change the fundamental pictures above.

Jeff Sachs on aid and evaluation

Mon, 27 Jan 2014 00:37:00 -0500

Cashgate

Wed, 18 Dec 2013 14:41:00 -0500

In their reporting on cashgate, the CS Monitor bizarrely calls Malawi a model state. If by “model” they mean a state that should be imitated, then I don’t know anyone who thinks this.

When thinking about the “cashgate” scandal in Malawi, it is important to remember that there is very little that is new here. Aid has been diverted in Malawi for a long time. What was new was the idea that Joyce Banda was serious about cutting down on graft and reducing the privileges of the President (e.g. selling the jet). That might still be true. It is probable that the activities falling under cashgate started before her time and it is possible that she was unaware of them.

I’ll give one quick example of aid diversion in Malawi. This one comes from my dissertation and happened under Muluzi in the mid-to-late 1990s. When Muluzi won the election in 1994, education enrolment in Malawi was terrible (around 67% of primary age school kids were in school). Muluzi responded to this by removing schools fees, and thereafter an additional one million children entered the school system. The system did not have the money to respond to this influx, and after a short time lag, donors responded with money to build new schools and supply materials.

The contracts for school construction went through the Ministry of Education, and were largely awarded to close contacts of UDF politicians. These contracts were awarded without bidding and were often paid upfront. The contractors took the money and didn’t build the schools (although they sometimes built the foundation). The World Bank gave 11.8 million USD for school construction during this time and aimed to build and furnish 1,600 new classrooms and 75 new schools. In the end, they actually built 858 classrooms (including 340 which were “unfinished”) and 59 schools (just under half were furnished). They spent all of their money.

The Bank’s Implementation Completion Report makes for a fun read (really). For example, it goes on about how “the Government failed to provide the necessary oversight”, “accounts were not well maintained”, and “records were not properly kept.” At one point the Bank notes that under the local implementing agency “there was virtually no supervision of works on site or management of the on-going contracts certificates and performance” and that they were “not following the contractual procedure and their valuations resulted in over compensation to the contractors.”

In all, the Bank spent all of its resources and managed to produce about 1/3 of its expected outputs. The Bank wasn’t unique here. Other donors were taken in too but they aren’t as transparent. I won’t get into it, but we know that the cash went to UDF politicians (because the ACB eventually busted the scam. See here, gated). The point is that this was in 1999. Malawi was (and is) known for this kind of graft and the only thing that was new under Banda was the idea that some of these trends might have changed. That is looking less likely now.

2013–14 Plans

Mon, 12 Aug 2013 14:23:56 -0400

I’m happy to announce that I’ll be spending the 2013–14 academic year at American University’s School of International Service. It will be nice to be able to stay in DC and keep working with such great colleagues and I’m really excited to be teaching courses on African political economy, African politics, and international development. I’ve posed the syllabi for the fall courses on my teaching page.

This semester I’ll be presenting a poster on foreign aid politics in Kenya at APSA on Friday, August 30th at 2pm. I’ll also be presenting work on the impact of colonial institutions on education in Ghana at ASA on November 23rd. If anyone wants to meet with me at either conference, be sure to send me an email or a message on twitter.

Mapping aid to Malawi

Mon, 17 Dec 2012 13:09:00 -0500

Below is a simple map of district-level aid to Malawi since 2000. The information on aid came from AidData and the map file came from the UNSALB. The code was written in R. There is nothing overly complicated about the map, but it was one of the first times that I sat down with R and figured out how to make a map.

I learned a few lessons along the way:

While the geocoded data from AidData are amazing, I really wish that they broke down disbursements by year. Unless I’m missing something, then all that you get in the Malawi dataset are date of commitment, planned date of completion, and amount disbursed to date. I don’t want to complain, but data by year would be really useful to researchers.
It is hard to find trustworthy shapefiles of maps with subnational boundaries. I tried using a few dodgy ones (look at the odd boundaries in the south) before stumbling across the UN Second Level Administrative Boundaries data set project. I will start there next time that I need maps.
If you are mapping in R, maptools is your friend. If you are merging datasets, then I found it far easier to use join (part of the plyr package) than merge. I still am very much learning R, so interpret my recommendations accordingly.

Worldstat

Tue, 30 Oct 2012 16:10:00 -0400

On marginal revolution I came across a new Stata package by Damian Clarke. It makes it very simple to show both maps and time series graphs of World Bank data. Even better, because it uses wbopendata it can work with any of the World Bank’s open databases. This gives you access to thousands of variables.

After installing (type “ssc install worldstat”), I graphed net ODA as a percent of GNI to Africa (type “worldstat Africa, stat(DT.ODA.ODAT.GN.ZS)”). The cleaned up version of the graph is presented here. Damian deserves a lot of praise. He’s made it very easy to make beautiful, useful representations of data. I can see myself using this frequently for slides.

Graphing the Volatility of Aid

Fri, 20 Jul 2012 19:15:00 -0400

Foreign aid is very unpredictable. This unpredictability makes recipient government planning more challenging than it ought to be and lessens the value of aid. One of my research interests is the effect of aid volatility or large swings in aid on different political outcomes in recipient countries. One challenge of this work is conveying the magnitude of aid volatility.

The figure above is my current best attempt at showing just how volatile aid can be. Each year shows two bars, each representing an aid change in a country in sub-Saharan Africa in that year. The blue bars show the largest aid increases and the red bars show the most extreme aid decreases. I’m not particularly good at presenting data, so if you have criticisms please pass them along: @ryanbriggs.

The data used are ODA disbursements and I subtracted out technical assistance and debt relief. The sample of recipients contains all countries in sub-Saharan Africa between 1961 and 2008. I measured aid volatility as percentage changes from the aid level in the previous year. This way of measuring aid changes will produce large values if the previous year had an unusually small amount of aid. For example, Zimbabwe went from 0.46 million in aid in 1979 to 227.42 million aid in 1980, for an increase of roughly 49,000 percent. To avoid these situations, I dropped any observations where a country received less than 50 million dollars of aid. This means that all of the percentage changes in the figure above are happening on a base of at least 50 million (2008) USD of aid.

The figure reveals that each year some countries experience very large changes in aid. Social scientists often focus on averages—and average aid volatility is large enough that it is a problem—but the tails of this distribution are really scary. Each year of the figure shows two instances of real countries getting knocked around by aid changes, so I’d say that tails of the distribution matter. Some countries went from over 50 million in aid one year to less than 0 the next. Many other countries saw their aid double or triple in one year, and one saw a six-fold increase. How can a government plan future expenditures if a large fraction of their revenue can either quadruple or be cut in half. This is what is at stake in the calls for donor coordination to reduce aid volatility.

Update: For those who asked, I don’t have the data to measure aid changes as a percentage of government revenue or expenditure. To the best of my knowledge, these data do not exist across time for most of the countries in SSA. If I normalize aid changes as a percentage of GDP, I often see swings larger than +/- 5% of GDP.

A final note: it is hard to believe that these kinds of aid changes don’t affect recipient country politics, right?

RCTs and validity in other sciences

Tue, 29 May 2012 14:23:24 -0400

Randomized control trials (RCTs) are the gold standard for some kinds of development interventions, but they are naturally limited in how broadly they can generalize to new contexts. This point was recently raised in an excellent debate and discussion between Abhijit Banerjee and Angus Deaton.

I mentioned this before and drew a distinction between doing RCTs with people and doing RCTs with larger social units within countries. Countries are far more heterogeneous than people, and so it is easier to generalize from some sample of people to the rest of humanity than it is to generalize from some sample of villages to all villages.

The reason I’m bringing this up again is that I didn’t realize how far medical science has moved towards increasing internal validity at the cost of external. From the Aid on the Edge of Chaos blog:

Ironically, “the canonical tenets of ‘scientific excellence’” are threatening to undermine the whole enterprise. One rather shocking – for me, at least – example relate to the latest developments in research on mice, where a lot of resources and funding have been poured into the cloning of genetically identical animals in order to enable fully controlled, replicable experiments and rigorous hypothesis-testing. However, the findings of this research have turned out to be useless when applied in the real world of diversity and change.

I recommend reading the whole article. It is sadly comforting to know that every science has these problems.

The African music on my ipod

Thu, 05 Jan 2012 10:40:52 -0500

Roving bandit recently made a few music recommendations, and Michael Clemens asked for more. I’ve been traveling in Ghana, Kenya, and Malawi the better part of the last year and I’ve managed to collect a fair amount of African music from gospel to rap. Much of this will not be new to seasoned listeners, and I’m ignoring a lot of classics, but this is what I’ve been enjoying. I’ll list the artist, country, and a song that I really like:

Brenda Fassie (South Africa) - Vul'indlela
Magic System (Côte d'Ivoire) - Premier Gaou
Miriam Makeba (South Africa) - Pata Pata
Bad’s Diom (Mauritania) - Koune Ak Sa Vibes
Napo De Mi Amor et ses Black Devils (Benin or Togo) - Leki Santchi
Onipa Nua (Ghana) - I Feel Alright
Ebaahi Soundz (Ghana) - Oshitℇ
The Very Best (Malawi) - The Warm Heart of Africa
Letta Mbulu (South Africa) - Welele
Rose Mhando (Tanzania) - Yesu Nakupenda
Roger Damawuzan (Benin or Togo) - Wait for me
Sir Shina Peters & His International Stars (Nigeria) - Yabis

Finally, I heard sawa sawa le, by Flavour, every day in Nairobi and frequently in Ghana. It was easily the song that I heard most in my travels and I became curious about it. The song is actually a dance remix of Nwa Baby, and Nwa Baby is a cover of an earlier highlife song (called sawa sawa le) by Rex Lawson. You can find more info and listen to the earlier versions here.

If you want more music, the blog Awesome Tapes from Africa has a large, otherwise hard to find selection of older music for download. Unfortunately, I don’t know the legal situation surrounding the tapes and I am quite sure that the artists aren’t getting any money out of it. I would also recommend the labels Analog Africa, Soundway, and Syllart productions. Honest Jon’s also has a great selection of African music for sale.

New Design and Public Data

Tue, 03 Jan 2012 16:37:45 -0500

I’ve decided to usher in 2012 with a blog redesign. I won’t be regularly writing on the blog until I am back in Washington around September, but I’m still really happy with the new design. Aside from looking pretty, I now have the space to post some of the data that I’ve been collecting over the last nine months.

The new data section of the webpage contains election information, government budget documents, and reports on government spending and various aid projects in Ghana, Kenya, and Malawi. I’ll be adding more documents from each country, as well as documents from Zambia, over the coming months. Despite being technically public and free to access in each country, I found many government documents quite difficult to access. While I’ve greatly enjoyed my field research, my hope is that the next researcher doing this kind of work won’t have to travel just to get some basic information on foreign aid and government spending.

Happy new year!

Power to the People: Village Electrification in Ghana

Fri, 17 Jun 2011 14:21:00 -0400

I’ve been in Ghana for about 5 weeks researching electrification projects, and I thought it was about time that I talk about it. In the late 1980s and 1990s, electrification projects became a priority for the Ghanaian government and a large amount of donor funding went towards village electrification. My starting point is a study and government wide plan in 1989. To simplify slightly, there were two main components in the government’s electrification strategy. The first were Self-Help Electrification Projects (SHEPs). SHEPs provided villages with electricity if they were already near high voltage lines and if they could locally provide some of the resources for electrification. Usually it worked out so that if the community could erect the wooden poles for power lines and wire a third of the houses, the government would do the rest. There were a number of successive SHEPs from the 1990s to present.

The second component was the National Electrification Project (NEP). This was an ambitious project that aimed to electrify over 400 villages in Ghana, including all district capitals between 1993 and 1998. It was expected to add over 100,000 new connections to the grid and it was building out the infrastructure that enabled many more SHEPs. When it ended in 2001 it had added 89,000 connections, which is pretty impressive. Conflict in the north was the main reason for missing the connection target and deadline.

I’m starting out with a basic question: Who got electricity?

There are two ways to answer this. The first it to examine where the government spent resources for electrification. The second is to look at surveys that track who actually had electricity at certain points in time. Below I have regional results for both approaches. To examine who actually got electricity, I looked at results from Ghana Living Standards Surveys in 1991/92 and 1998/99. Both surveys asked respondents to name the main source of light for their home. First, I have a map that shows the percentage of respondents in 1991 who said that they lit their homes with electricity from the national grid:

In 1991, only about a quarter of all Ghanaians had electricity and the spread of electricity was very unequal. The Ashanti and Greater Accra regions had decent access to electricity, but the east and north of the country lagged far behind. The NDC, who was in power from the start of this period until the 2000 election, was explicit about wanting to close this gap. Let’s see if they succeeded.

The map below shows the answers to the same question in 1998:

By the end of the decade, Volta and Western region—and to a lesser extent the surrounding southern regions—made large gains and closed much of the gap with Ashanti. Accra still has a huge lead in electrification with 82% of respondents using it to light their homes. At the time of the survey, much of the planned electrification in and above Northern region was delayed due to conflict. To envision the full impact of electrification from 1991-2000 you should mentally make the northern regions a little darker.

To make the changes more obvious, the map below shows only the percentage point increases in electrification. Here you can see Volta and especially Western region stand out. Western region went from 22% of respondents using electricity for lighting in 1991 to 49% in 1998. Volta went from 10% to 28%.

Finally, I have regional information on financial allocations under the NEP. I received the financing details from the World Bank. The Bank funded about 45% of the NEP, bilateral donors funded about 25%, and the government of Ghana provided funds for about 10% of the project. I don’t know who funded the remaining portion (it is marked as “other”), but it is likely other smaller bilateral donors. Below I have a graph that compares the allocation of money against the the actual percentage point increase in electricity usage.

The graph is slightly misleading because a dollar of funding for electrification might go further in regions with a smaller population or area. On the other hand, a lot of the resources here went to fixed costs such as clearing land and building power lines, so the population problem may not be so bad. There are a few things to note about the graph.

First, Greater Accra and Ashanti were allocated very little money and saw very small gains. They also, however, had high baseline rates of electrification. Second, Western and Volta saw the largest increases in electrification, but Volta didn’t receive that much funding through the NEP. This is an interesting result, as it shows that Volta must have either had a large number of SHEPs or must have received resources through other channels. Third, a lot of money was spent in the north of the country and, as I said earlier, the GLSS in 98 was completed too early to capture it. The correlation between NEP funding and actual electrification is 0.87 if I don’t include Northern, Upper West, and Upper East regions.

Well, that’s it. I’m not going to make bold (or any) conclusions until I have more district-level data and better controls. People familiar with Ghanaian politics, however, can probably spot some interesting patterns.