Hi all,

So, it’s been a fairly eventful week, so I’ve been more or less glued to the news. This has had the welcome effect of helping me notice all sorts of fantastic stories (like the woman who responded to a politician’s sexist joke by running against him – and winning, and the guy who forced a plane to make an emergency landing because of his infidelity), but at the cost of bringing so much to my attention that this week’s links could easily be thirty bullets long. I’ll try and spare you that particular form of torture, but I make no promises: of all the failings that I’ve been accused of in the past, brevity is notably absent.

1. This is basically my perfect paper: an RCT about using cricket to investigate the effects of collaboration and competition on discrimination. Matt Lowe randomly assigned Indian men into mixed-caste or same-caste cricket teams, and then randomly assigned opponents and finds that collaborative cross-caste contact increases cross-caste friendship and reduces caste-favouritism when allocating rewards. What’s more, this has effects on efficiency and leads to better teammate selection in future matches. I was having a conversation recently about how seemingly trivial topics can generate important findings – this could potentially be one (or I’m just massively over-weighting the importance of these results because cricket is obviously more important than life itself).
2. Another Matt, Collin, here considers some of the implications of machine learning for replication and research transparency.
3. This didn’t get much coverage that I saw, but is really worrying: Zitto Kabwe, a Tanzanian politician and former chair of the Public Accounts Committee was arrested for presenting economic analysis that questioned the Bank of Tanzania’s growth figures. I have no idea if Zitto’s analysis is right (I haven’t seen it), but it’s pretty terrifying if questioning official statistics can get you arrested – when I lived in Tz, it would have made me a criminal on a near daily basis. Justin Sandefur goes into the details of Zitto’s analysis, here. As an aside, Chris Adam points out the importance of keeping an eye on domestic credit to the private sector, one of my go-to indicators on the economic health of a country, in a quote Justin includes.
4. Gabriel Zucman makes the case for a world financial registry in the Grauniad. He’s one of the best researchers on inequality out there, and well worth reading.
5. I really liked this: Martin Williams (at the Blavatnik School of Government) summarises his research into why so many public sector projects in Ghana remain unfinished. He shows that corruption is probably not the main reason: firms tend to do more of the work than they’re paid for, not less. Rather, he points the blame at the changeable nature of political decisions – as the political alliances underlying a project unravel, the money allocated to it runs the risk of be siphoned away to start a project with better political backing, but which may itself be subject to the same effect in the future. The paper is underpinned by an amazing data set, too.
6. David Evans reads 147 development papers and summarises them in one sentence each. It seems like a magic trick, but he does this kind of thing regularly. It’s incredible – click, read, comment and thank him. It’s an extraordinary public good. Related: he examines the geographic coverage of these papers here.
7. Lastly, two food links: First an amazing Reuters investigation into North Korea’s food markets (some of the food designed out of necessity looks amazing!); and second, an economist teaches about international trade using goulash. And with that, I’m going to have lunch.

Have a great weekend, everyone!

R