Economics & Marginalia: May 17, 2024

Hi all,

It was a long, hard slog, and there have been a number of setbacks (not least days on end of merciless grey skies and rain), but London has finally settled into a real, actual summer. That means two things: first, everyone is in a ridiculously good mood. We get three weeks of sunshine a year, and it can cause such un-English behaviour as public smiling, conversations with strangers and random acts of goodwill. And secondly, it means that CGD will begin its ‘summer Fridays’ policy soon. Summer Fridays mean that if there’s no urgent business to attend to, we can take the day off and spend it with our kids, gardening or just sitting in the sunshine with a bottle of wine (all three together are not wholly advised). The links usually count as urgent business for me, but if there’s a good test match on, it does mean that your usual service may occasionally be interrupted over the next few months. It’s a glorious act of felicity that the English cricket summer reaches its crescendo just as the NBA playoffs finish (my bet, for what little it’s worth given my track record, is that this is the year the Celtics break through again, but I hope I’m wrong, for 77’s sake).  

1. He had me at ‘Balassa-Samuelson’. Oliver Kim has a super blog post, which I discovered via Pseudoerasmus, on African price levels. He argues that the puzzle, identified in a 2013 CGD paper by Alan Gelb, Vijaya Ramachandran and Christian Meyer that I have long loved, that African price levels are too high relative to their GDP per capita, which may be behind their relatively high wage rates, may not be a puzzle after all.  He suggests that the simple relationship proposed in the Balassa-Samuelson model (a relationship that, in a roundabout way, underpins my belief that exports are effectively the key to sustainable and inclusive economic development) is too simple, and the introduction of an agricultural sector can explain the results Alan, Vij and Christian discovered. But the story doesn’t end there: as Pseud points out in his twitter thread (read all the replies, this is excellent stuff), even correcting for the points Oliver makes, African prices may be higher than, say, Asian prices. A lot comes down to how agriculture develops, something I’m hoping to write something on for CGD soon.

2. And while we’re on the subject of African economic transformations, this piece by Maggie McMillan and Harun Onder is also very good. It dampens some of the enthusiasm I’ve shown for the idea of a tradable services-led transformation path (one I discussed in this paper with Charles Kenny and Brian Webster). Their argument is that there remain important drawbacks to a tradable service-led model: specifically that the export performance of services has been disappointing, they have attracted relatively little FDI and they generate less domestic tax revenue.

3. Man, I hate it when bloggers take the easy way out and choose to write about easy, simple issues, like how to restabilise the Sahel. Ken Opalo’s article on this, a problem that has stymied countless analysts and armchair diplomats (as well as real, actual diplomats), is typically excellent and well-written. This sentence in particular resonated: “[t]he biggest obstacle to clear thinking about stabilizing the Sahel at the moment is the fixation on electoralism as a magical solution.” I think that message has started getting through, at least to the people I speak to at the sharp end of policy, but maybe there’s some selection bias at work there. Ken identifies a number of popular bad ideas that should be avoided, and a few better ideas that might yield results. It’s less clear how to implement them, but if it was easy or obvious, it wouldn’t be necessary to write this piece.

4. Speaking of political violence and solutions, Thiemo Fetzer and Stephan Kyburz have a good VoxDev piece on the value of local democracy in forestalling violent conflict over resource rents. I must admit to some scepticism and desire to learn more about this example, but it’s well worth reading.

5. Two good ones from Development Impact: first, David McKenzie continues his annual rundown of the major economics and development journals and the experience of those submitting to them, a great public good. And he also interviews Jing Cai, a great researcher who we’ve hosted at CGD’s Future of Development series; it’s full of interesting reflections.

6. I regularly give a presentation about using evidence in policymaking: I’ve delivered it to policymakers, to undergraduates and to postgrads in public policy courses. One of the slides is just the words “there is never a bad time to ask about the data”. Two great examples of this. First, Saloni Dattani from Our World in Data has a really good deep dive into the data on maternal mortality in the US. It looks like it has been trending upwards recently, prompting a great deal of comment. But Saloni shows convincingly that this is just an artefact of a change in the way maternal mortality is counted, one that has been rolled out across states only slowly. Since the change increases the number of counted maternal deaths, the slow rollout means it looks like a gradual increase for the country as a whole. Always ask about the data. And secondly, Andrew Gelman reports on work that digs into the details of why Americans report such sharply diverging stories about their economic circumstances.

7. Lastly, I came across a fun paper (I can’t for the life of me remember how anymore), which uses Bollywood and Hollywood movies as a dataset from which to measure different social norms across cultures. I had many, many thoughts about this one, but my first was: I cannot think of a data source I would be less happy to define our cultural moment than these sorts of movies. Bollywood has many things to offer (a long history of absolutely banging tunes and some incredible dances, for example), but I’m not 100% certain how well it reflects the day-to-day life of the country. It made me think, though—if AI is going to be learning about us from our most ubiquitous pop culture, which we love for reasons other than realism, future historians producing their articles in Chat GTP-50381 are going to have an odd view of our lives.

Have a great weekend, everyone!

R