Economics & Marginalia: December 18, 2023

Hi all,

These links are winging their way to you late, and you can blame CGD Europe’s Christmas festivities; it’s not exactly that it was a bacchanalia of epic proportions (at least it wasn’t by the time I left), but it tore a chunk out of my working week, especially as I brought my son to the first portion, his first trip to the office. He insisted on wearing a ‘work outfit’, consisting of a shirt, cardigan and corduroy trousers, which suggests he is going straight into the Season 1 Lester Freamon stage of his career. I’m afraid it was not a good preparation for real working life, involving much cake, Secret Santa and a mid-afternoon nap. The second portion was even less representative, involving karaoke (in which my dulcet tones were not put to work) and jenga.

1. Still there are some advantages to being late: not least that this Planet Money podcast on the background to Florian Ederer, Kyle Jensen and Paul Goldsmith-Pinkham’s EJMR paper dropped (transcript). It’s probably less interesting about EJMR itself (of which much has already been said) than about the circumstances under which research ideas come about: random conversations and chance discoveries play an important role. Though this paper doesn’t say anything causal about the world (for example, it can’t show whether going on EJMR makes people more toxic), I feel like this ‘chance discovery’ mode of research origination is more common in world where plausible exogeneity plays an ever greater role in publication prospects. Previously, one would often talk about a big question that warrants further investigation, and then embark on the best way of getting at an answer. Nowadays ‘identification’ (of a causal effect) is as often the first stage. This could be a plausibly exogeneous variation: what can we study with it. The best papers, of course, do not trade off between these methods: researchers spend a great deal of time thinking about important things (as I happen to believe the ubiquity or otherwise of EJMR in economics is) and then one day have the exogeneous variation eureka moment and start a paper.

2. The DI job market papers series is winding up, but two very good ones came out this week. First, I was a big fan of Sergio Puerto’s paper on how seed developers do not get the most out of the seeds they develop by failing to account for and cater to the heterogeneity of farmers: he finds a system in which farmers choose which seeds to use from multiple new varieties shows higher take-up of new seeds, and higher farmer productivity. It demonstrates two things I’ve banged on about in the past: first, using the market a little bit more in designing incentives for innovation can be very powerful; and as central planners, we typically have less or less valuable information than we think we have (though that is a disadvantage of central action, not a reason not to do it at all). I was also a big fan of Geetika Nagpal’s work on zoning changes and affordable housing in India.

3. I’m going to use this example in teaching about public procurement next year: a city contracted the design of its bus system out to an algorithm and caused chaos. There are lots of reasons why it didn’t work out, but this is the central one: it matters more that you choose the right thing to optimise than the most efficient way of optimising it. They contracted for the wrong thing (fewer bus routes, rather than a better bus system, subject to cost); from there, the only way is downhill. You only have some control over how fast you go.

4. Christmas is coming, so it would be remiss of me not to include at least one link about books (the best possible Christmas gift; if you know me well enough to get a Christmas present from me, there will be at least one book involved). This isn’t economics, but LitHub on giving books as gifts is wonderful, though it doesn’t go fully into strategy. My approach is that you have two options: if you know someone’s taste well enough, you can get them something they wouldn’t buy for themselves. Book gifts can push you to something you will plausibly enjoy, but is outside your wheelhouse (I once bought an anthropologist friend of mine who is allergic to numbers a book about statistics, which I know she read because she’s quoted it back to me). If you don’t have that level of knowledge, the safe thing is to buy a book that brings you joy, and then try and share a bit of that with receiver (the most recent book gift I gave was Rumpole of the Bailey, a ray of curmudgeonly sunshine if ever one existed).

5. I have largely gone off Tyler Cowen’s writings on MR (an opinion factory on hyperdrive), but it is always worth listening to Fuschia Dunlop, and Cowen asks some excellent questions in this Conversation with Tyler.  

6. This is wonderful: Tim Harford riffs on the distinction Oliver Burkeman has made between ‘telic’ and ‘atelic’ activities: the telic having defined end goals and the latter having none, and can be enjoyed without being clouded by an end. It made me think of why I like birdwatching. Even if I ever get close to seeing every species of bird in the world (I will not, nor will I ever seriously make the attempt), it wouldn’t be the end. Once you’ve really begun to love birdwatching the enjoyment of watching a flock of sparrows darting up and down from a building ledge is just as meaningful as an 80% obscured glimpse of Ross’s Turaco in a forest in Malawi. I think my son has a streak of this in him: we spent an couple of hours building magna-tile houses yesterday, and every time we completed one, he simply started again: the building was the fun, not the house he built.

7. It’s not uncommon to feel that the times we grew up in were a kind of wonder-period, much better than modern times where everyone does things just a little bit wrong (and in a very old-man-shakes-fist-at-sky conversation at the Christmas party I discussed this with respect to teenage rebellion with a couple of colleagues), but sometimes it’s objectively just correct. And Planet Money discuss how advertising really was catchier and better in the 1990s than it ever has been since (transcript). It wasn’t that the ‘90s was full of geniuses with an inexhaustible supply of memeable catch phrases (whassup?) and jingles, it was that with everyone watching live TV, it was worth the effort to invest in them. Listening to this prompted me to google one of the adverts that has always stuck with me, something I remember vividly to this day: I’d often thought it must have come out when I was 10 years old or so, but no: apparently, Perrier managed to inveigle their slogan (“Once you’ve discovered Perrier… nothing else will do.”) into the brain of a three-year-old and it’s still there nearly four decades later.

The links will return in January. Have a great weekend, and a great new year, everyone!

R