The CSAE Blog

Hi all,

It’s an early morning links round-up today, as I’m showing my visiting mother around all day (which mainly seems to consist of a tour of all the bookstores in the UK – if you ever wonder where I get my reading habits, it runs in the family). I’m going to let you all in on a secret: I am not a morning person, so bear that in mind when you read the inchoate and incoherent jumble of words below. If at work you ever want a thoughtful opinion from me, wait till after 10am. That’s no guarantee you’ll get it, but at least you’ll maximise your chances…

1. As I’ve mentioned many times before, I don’t tweet, so I missed this until a friend sent it to me: Chris Blattman asked people for ‘unpopular’ research findings in development. Unsurprisingly, Lant Pritchett had a few. What makes Lant a great communicator is that he finds the controversy and argument in what he wants to get across – he either thrills or annoys you into listening. This list is as good as you’d expect from him: he points out that democracies have a much tighter variance in growth rates than non-democracies, so most major growth and poverty-eradication experiences come from non-democracies (with an obvious flip side); that in absolute terms, it’s still better to be poor in a rich place than rich in a poor one, whatever Angus thinks; and that any within-country gaps in outcomes still pale compared to the across-country gaps.
2. Speaking of good communicators, I was pretty devastated when Hans Rosling died last year – he had a rare gift of being both an absolutely riveting speaker and being absolutely honest with data and evidence in building his discourse. Many great speakers rely on distorting analogies or reductio ad absurdium to get their points across. Hans used our own biases to surprise us with simple fact. Tim Harford writes about the book he was writing with his children, now published: unsurprisingly, the review is positive. I will be buying it, and encourage others to as well.
3. I had a secondary school teacher who loved quoting Keynes (more generally, he just loved Keynes, as the macroeconomist in me still does). Every few lessons, someone would point out that ‘in the long run, this system should reach an equilibrium’, to which he would inevitably shoot back “in the long run, we’re all dead”. But how long is the long run? David Evans investigates, and finds that macroeconomists consider everything from 9 to 10,000 years to be ‘the long run’. (I’m not typically a fan of path-dependency papers that say X happening 300 years ago led to Y today without any real analysis of why it didn’t change or what mechanisms translate the effects to the present day, but that’s a discussion for another day.)
4. Occupational segregation is increasingly part of the popular discourse on gender inequality in work, which is fantastic, as it’s slowly crowding out those naïve comparisons of average hourly compensation that have dominated the discussion to date (in developing contexts, Markus and his brilliant team in the World Bank do great work on this). This fantastic article by Maggie Koerth-Baker explains why this shift matters so much: it focuses our attention on a different set of root problems and solutions, and also allows us to see how uneven the progress to date has been. While things like education or outright discrimination have declined as explanatory factors in the gender pay gap in America, for example, they’ve declined more for white people than for others – so race is now a more important explanatory variable than it was 40 years ago.
5. The Economist on the development of an academic science community of Africans, in Africa.
6. Good research into the penalty you accrue in the labour market from being a refugee compared to an immigrant; Lant points out that development gains from movement far outweigh what you get from staying put, but not all forms of movement are equal. (And speaking of immigrants, the making of Little Punjab in Southhall).
7. What do you think the best thing about The Rock is?… It doesn’t matter what you think the best thing about the Rock is! The Ringer lays out all you need to make an awesome movie starring The Rock. Mainly, you need the Rock. Then you need an enormous enemy for him to lay the smack down on. And if you manage to pair him up with an albino gorilla taking steroids (not you, Brock), well.

Have a great weekend, everyone!

R

Hi all,

Sorry I forgot to warn you all that I wouldn’t be sending any links out before Easter: partly forgetfulness, partly the fact that my stomach had entirely overtaken my brain in anticipation of our Easter trip to Italy. The preparatory re-routing of blood to my belly proved a good move as much more of me has returned than left (a growth rate that would make China blush with shame). I’ve just spent hours catching up with my RSS feed. This one might be a bit epic, but on the plus side they start with a clip from Ferris Bueller and end with Justin Bieber’s moobs.

1. How often do you get to preach free trade with the help of Ferris Bueller’s Day Off? But if you remember the greatest teaching scene in the history of movies (thankfully, no idiot climbs up on his desk to shout ‘oh Captain, my Captain’) you’ll know that the tariffs made famous by Ferris Bueller’s absence, those of the Smoot-Hawley Act made the Great Depression – anyone, anyone? – worse, not better. Planet Money uses this scene as a stepping off point to talk some of the dumber tariffs of all time here (transcript); they also investigate the effects of China’s retaliatory tariffs here (transcript, note that the US and China have now escalated the proposed schedules and China will be imposing tariffs on Boeing and soybeans), pointing out they’re hitting agriculture hard to hit Trump’s base. 538 used the impending trade war to teach us a little game theory, but found that their readers tended to get more vindictive as games repeated, rather than less, as the theory would predict. We might all be headed to an – anyone? Anyone? – recession, but at least we got some good reading.
2. Let’s cheer up by reminding ourselves how dumb our more popular political instincts are. Jonathan Portes summarises around 20 years of reading on the effects of immigration on the UK’s economy and points out that not only does it have no negative effect on employment in the medium/long run, it appears it doesn’t have any in the short run, either (Adam is saying ‘magic economics’ somewhere); it increases productivity; and increases wages for most of the distribution. Also, immigration increases in the US haven’t increased crime: if anything they’re associated with lower crime. And if the US does restrict its H1-B visas, not only will it disproportionately hit the most highly-skilled groups, it seems local people will not be employed in their stead – companies would just shrink.
3. Lant mentions immigration in this barnburner of a post, but really it’s about focusing on the big questions of economics. Elaborating on a point he made in our last economics conference, he points out that successful growth accelerations (or decelerations) have a much, much larger impact on welfare than even the best things donors can implement directly, like cash transfers or livelihoods work. I agree with Lant that the big questions are the most important ones, but the implied either/or isn’t there for me. Not everyone should be doing the same kind of work. We need more Lants, but not fewer Esthers.
4. Speaking of more Esthers, a series of podcasts with women in economics – the most recent one is Claudia Sahm. And researchers have found that in the movies, men are much more likely to be smart and make cognitively sophisticated points. Someone should show them Iron Fist, which should really be renamed ‘Leaden Brain’. Lastly, Justin Sandefur argues that girls education is not the answer to the gender pay gap; the problems lie elsewhere (he is not, obviously, saying we shouldn’t fund it – it is the answer to a lot of other things).
5. I’ve thought about this a lot, because I expect some of the things I want to research to throw up null results: how do you emphasise how interesting the lack of a statistically significant effect is? David Evans has answers. Also on Development Impact, Berk goes into great detail about why the new Give Directly results imply no long-term effect from cash transfers in the setting investigated.
6. Dietz is back – and he’s dissecting a paper from the fantastic Jane Humphries, who taught me years ago, when I was an 18 year-old.
7. “OK, so basic income is all about the freedom to say no. That’s a privilege for the rich right now. With a basic income, you can say no to a job you don’t want to do. You can say no to a city in which you no longer want to live. You can say no to an employer who harasses you at work . . . that’s what real freedom looks like.” There’s some I disagree with here (especially the implication that in developing countries structural problems can be ignored and cash will solve all), but a very good piece on UBIs.
8. Lastly, as a reward for wading through all those links, a definitive ranking of all the ridiculous tattoos on Justin Bieber’s ridiculous torso. Did you know he has what appears to be a lion wearing a tiny crown tattooed on one of his moobs? There’s a bear on the other one. He looks like the drawing wall at a kindergarten.

Have a great weekend, everyone!

R

Hi all,

I’m going to be honest: my brain is fried. The CSAE conference started on Sunday morning, and I was either in a session, picking brains for opinions about my own research design or pickling brains with Matt in The Bear pretty much constantly until Wednesday morning; then there was the one-day GrOW conference, about research on women and economic growth; and after that a half day in DFID where I spent some time lobbing grenades of ill-informed speculation at some of the smartest researchers out there on gender and jobs, growth and the economy, trying to work out what the frontiers for new research are. So, if this week’s links are even more incoherent than usual, that’s the reason; that and the fact that I’ve spent the day trying to get my own ideas into some kind of coherent shape for transfer of status. They are putting up an enormous fight against my attempts to impose order, like a group of highly persistent anarchists in my brain.

1. The CSAE conference is amazing. The sheer range of research being presented by researchers from the US, Europe and Africa was mind-boggling, and logistically what must have been a nightmare to synchronise went off without a hitch. I wrote something about each of the three days: on the first, talking about Pamela Jakiela’s paper with Owen Ozier, considering the links between ‘gendered language’ and female outcomes; on the second I focused on Rachel Glennerster’s fantastic keynote, while noting that I think not all researchers are good at communication, nor should focus too much on becoming so; and on the final day, talking about optimisation, and how the limits to how people make decisions can help frame our understanding of the world. If that’s not enough, check out Markus and co’s set of one-line summaries of most of the research presented; and the #OxCSAE2018 tag on twitter. My only gripe: when the chair asks for a short question, please stop with the speeches disguised as questions.
2. Raj Chetty has just released the latest paper from his gargantuan equality of opportunity project, and it’s a barn-burner. He shows that while being born to wealthy parents is typically a great predictor of one’s own wealth, this is not true for black males, who have large gaps in their employment and wage results compared to similar white males. Paper here, NYT coverage here.
3. While we’re being pessimistic about the world, Seema Jayachandran blogs on the persistent and substantial gender gap in the kind of things men and women study in academic economics. Also depressing: a court case is raging in California, about who should pay for climate change. 538 have a fantastic summary of how it’s being argued, and it boils down to whether you blame fossil fuel companies for digging the stuff up, or society for wanting it.
4. I really liked this research from Japan, using an experimental design to elicit attitudes about migration from elderly Japanese without really making it clear that it was about migration at all. Two interesting things: first, it manages to partly separate out the economic and cultural objections to migration; and it manages to induce people to change their attitudes towards migration by providing information on the former.
5. In a kind of empirical economics version of Pacific Rim, Guido Imbens brings the ruckus to Angus Deaton’s critique of RCTs. I feel strongly that you should only read this while listening to the GZA verse of Bring da Ruckus.
6. Brian McCaig and Nina Pavcnik look at the effect of a trade agreement between Vietnam and the US on labour demand. We were literally discussing the need for more research on the demand effects on employment yesterday. This fits the bill.
7. Lastly, with March Madness in full swing, a reminder that DeAndre Ayton may not be in the Elite Eight, but he could still fold you up and play you like an accordion.

Have a great weekend, everyone!

R

Hi all,

These links start with deaths a disturbing amount these days. I must be getting old. I had thought Simspons characters don’t die, but apparently it’s quite common. Still, Stephen Hawking probably had a bigger impact on the world than Bleeding Gums Murphy, befitting a man who apparently said ‘my goal is simple: a complete understanding of the universe, what it is and why it is at all’. Imagine being a supervisor reading that in a research proposal. I’m not going to pretend to have read A Brief History of Time, though like everyone else I’ve written the odd note ripping off its iconic title. You can get flavour of his gift for communicating complex ideas simply in this excerpt, though.

1. Speaking of communicating complex ideas simply, there are some gems in this Tyler Cowen interview with Chris Blattman. I particularly liked this: “I usually think of the incentives as aligned for peace, so war is the puzzle to be explained”, but he also picks out the aspects of conflict that economists tend to understate. In particular, he mentions our failure to really get to grips with the psychic costs and impacts of war and how much about conflict comes from people making mistakes (more generally, despite the influence of behavioural economists, mistakes and their implications for organisational forms, regulations and individual economic interactions are still under-studied). The interview goes through lulls when Tyler shoehorns in all sorts of tangential detritus (designed to showcase how clever he is, rather than how clever his interviewee is?), but it’s worth reading. Related: Blattman savages community-driven development in this twitter thread.
2. What’s 22 trillion digits long and still getting longer? Fascinating piece about pi by Oliver Roeder, not least this: “A value of π to 40 digits would be more than enough to compute the circumference of the Milky Way galaxy to an error less than the size of a proton”. It also makes mention of Ramanujan, one of the most amazing people I have ever read about.
3. Two papers on migrants and education. The first, by Michela Carlana and co-authors, finds that simple aspiration-raising interventions can have a substantial effect on performance;  the second that large influxes of refugees does not compromise native children’s educational outcomes, rather improving them.
4. Planet Money make a point that was missing in most of the hand-wringing I read about Trump tariffs, specifically that the formal justification used for the tariff is ‘US national security’. This takes advantage of a WTO loophole for the establishment of tariffs, and may result in very specific form of tit-for-tat that could fundamentally undermine the WTO’s institutional effectiveness (transcript).
5. In news that will not exactly stun people who have transacted official business in much of South Asia, it turns out who you are matters for your outcomes. But interestingly, the mechanism may not be discrimination so much as an enhanced ability to pick ‘better’ economic partners when they’re from your own group – which might be an information or an enforcement effect, I would imagine.
6. This map of Africa made the social media rounds this week, giving the etymology of every country’s name. But before we exoticise the ‘Land of the Milking Nomads’ too much, let’s not forget that England means ‘Land of the Angles’ and Italy probably means ‘Land of the Young Cattle’.

Have a great weekend, everyone!

R

Hi all,

This week’s post is pretty much getting the last bit of me before I sleep for 48 hours to recover from the end of term. I’m just studying and I find it exhausting: I have no idea how the various lecturers manage to teach multiple classes, help their various charges negotiate incipient academic life and actually do their own research. But I can only sleep for a while: next Sunday is the start of the CSAE Conference. It’s going to be awesome, not only because of the Aid Thoughts reunion happening in the margins (and associated hangovers), but also for the incredible line-up. So much is happening in parallel, after the failure of my attempts to clone myself, I’m going to be glued to the internet to wait for David Evans’ inevitable (and inevitably brilliant) summary of all I missed. A side note: I’ll be blogging during the conference, hopefully both for CSAE and Aid Thoughts, so please do feel free to send me comments, rants or insights that you don’t have time to write up yourself: I promise to pass only the very best off as my own.

1. It was International Women’s Day yesterday so a few gender links to get us started, some incredibly depressing and some … well only slightly less so. Let’s start with ‘incredibly depressing’: Seema Jayachandran on India’s 21 million unwanted girls. It’s superb, looking at the sex ratio of last-born children in India and noting the behavioural implication – that parents keep having children until they have a boy. The trouble with this is that it means that even if there isn’t any discrimination in investment in children by sex, girls are born into larger families and on average receive fewer resources per head than boys. Also troubling: CGD ran pieces looking at the gender gap in US non-profits, and how it can be resolved. And more hopefully, Markus Goldstein on a very cool intervention in Norway that seems to improve gender norms among men, a key constraint to women’s economic empowerment. Lastly, and most trivially – the gender gap in IMDB movie ratings.
2. I’m quite pessimistic about our ability to resolve the man-made crises of the environment through behaviour change: bad behaviours are too entrenched, there are too many free-rider problems and too many of the really big culprits are organisations who face very strong incentives not to behave better in a world where aggressive regulation is pretty frowned-upon. A technological solution that is incentive-compatible, though? There must be huge optimism bias here, but MIT reckon they can get nuclear fusion to work within 15 years.
3. This link both reinforces and subverts some of what I said above: Bloomberg on how US Steel was the biggest culprit in its own descent into irrelevance. Apparently, powerful individuals in the industry resisted change, and the organisational incentives to get things right just weren’t strong enough to overcome that, leading us to where we are today: the industry begging to be protected so it can continue comfortably without moving with the times.
4. A really fascinating piece of work by Paola Giuliano and Nathan Nunn using climatic variation to test the proposition than societies in places with greater environmental instability are more likely to adopt new cultural practices – getting empirical estimates for an effect that evolutionary anthropologists have argued for a while.
5. All of Tim Harford’s five guidelines for thinking about statistics are unsurprisingly brilliant, but the first one is really important. He advises that you really think about your gut response to seeing a number  – really owning the way your priors and associated beliefs about the world are pushing you is necessary before you can be dispassionate about evidence.
6. Want to improve productivity? Make it less miserable to work. Sounds obvious, but apparently not everyone has clocked it. Also, ignore your phone. Which is easier said than done. My solution used to be to go out and celebrate a birthday. It would be days before I had a working phone again. Apparently, it works less well as the number of birthdays celebrated increases.
7. As someone who occasionally struggles with anxiety, I absolutely loved this piece by Kevin Love (yes, Kevin Love, the basketball player with the three stroke and outlet passes). He talks openly about his recent panic attack during a game and the emotional turmoil that led to it, and more generally about attitudes to mental health in competitive settings (academics, take note). Amazing.

Have a great weekend, everyone!

R

Hi all,

Greetings from Narnia! My perpetual pessimism about an endless winter has finally been borne out. I’m writing this e-mail on the bus between Oxford and London, looking out at a landscape almost entirely white, punctuated by the odd treetop and broken down car (please, take great care on these roads!). We just drove past a frozen lake. Seriously: this isn’t cool. Somebody, please kill that goddam groundhog.

1. About a year ago, I gave a presentation discussing the likely implications of the Trump regime on global trade, and said that there was much more noise than signal when trying to actually forecast his economic policy, concluding that a serious trade war wasn’t imminent. Well, it sure seems to be now: the US has imposed substantial tariffs on steel and aluminium, and the EU has already begun making retaliatory threats. This could be the start of something very bad, but it also could also be something of a prelude to not very much. The last time the US implemented similar tariffs, they were abandoned after nine months when it became clear that the cost of retaliation and the benefits gained were insubstantial.
2. I love food, a fact that can be easily observed by either becoming my friend or looking at a time-lapse of my waist over the last five years; crap research that scaremongers about eating or tries to turn the enjoyment of food into some sort of disorder really bothers me, especially when it turns out that it’s had the life p-hacked out of it. Stephanie Lee uncovers a litany of malpractice in the research of a Cornell academic who has done a lot of this kind of work. It prompted an uncharacteristically forgiving response from Andrew Gelman which is well worth reading. Side note: this piece of bacon-assassination is completely cack-handed, too: bacon is killing us, really? When, to increase my chances of getting bowel cancer from 5% to 6%, I’d need to eat two rashers a day for the rest of my life? That sound you just heard was me laughing into my Gino’s bacon sarnie.
3. Tim Harford jumps on the McKenzie, Bloom et al. bandwagon, reporting on the paper I linked to last week, and investigating why so many good practices firms use don’t seem to be picked up and emulated by other ones. I’ve been thinking about this too, and it strikes me that given how much noise there is in observing firm performance, it’s unlikely that even successful firms have a full understanding of why they’re successful, let alone their competitors. In such a context, any imperfection of competition allows for the possibility of good practices going unremarked and uncopied.
4. David Robalino, one of the World Bank’s jobs gurus, on the perils of measuring the ‘cost per job’, something DFID have had to grapple with extensively too. He uses general equilibrium modelling to come up with some very high estimates that nevertheless ring true. Well worth reading.
5. Get smarter in three minutes, Chris Woodruff edition: Chris distils a serious amount of accumulated knowledge into a pithy summary of what makes firms improve, coming to a conclusion that economists should have tattooed on their forearms for ease of reference: competition matters. In similar vein: everyone needs to be quiet for five minutes and read Francis Teal.
6. This week in “um… I’m going to need about three years to think about this… then maybe I can get back to you”: the conditions in which mice were conceived seem to affect their future characteristics, through an identifiable genetic mechanism. So many jokes to choose from (what the hell were Donald Trump’s parents thinking about? Were mine reading the FT?) but my thoughts are best summed up thus.
7. Lastly, I never thought I’d say this, but screw FiveThirtyEight readers. How dare they rank the greatest song in movie history a measly 71st out of 81? Clearly, they are much more a muppet than a man (or a very manly muppet). Also, apropos of absolutely nothing, LeBron James did this yesterday. I repeat: my thoughts.

Have a great weekend, everyone!

R