IPA’s weekly links

Guest post by Jeff Mosenkis of Innovations for Poverty Action.

A big thanks to all the folks who’ve donated to IPA’s anti-poverty work before the year end (you can also donate through Dean Karlan’s Facebook fundraiser through tomorrow, credit to his brave daughter on that one.)

  • Thirteen prominent economists offer their favorite econ papers of the year, but the paper making a splash this week is from Melissa Dell and Pablo Querubin, comparing two approaches to combatting insurgency during the Vietnam War. They exploit the fact that the U.S. Army and Marines had authority over strategy in different parts of the country and took different approaches. While the Army used an overwhelming bombing campaign, the Marines took a “hearts and minds” approach, embedding troops in villages for security and conducting development projects. They conclude the bombing increased the insurgency and worsened attitudes towards the U.S., compared to the hearts and minds approach. Or as someone commented, it turns out you can’t bomb somebody into liking you.
    • Their method for finding similar villages (that were either targeted for bombing or not) to compare got a standing ovation at an NBER talk:

[Newly] Declassified Air Force histories document that one of the factors used in allocating weekly preplanned bombing missions was hamlet security (Project CHECO 1969). A Bayesian algorithm combined data from 169 questions on security, political, and economic characteristics into a single hamlet security rating. The output ranged continuously from 1 to 5 but was rounded to the nearest whole number before being printed from the mainframe computer.

The study estimates the causal impacts of overwhelming firepower by comparing places just below and above the rounding thresholds, with being below the threshold used as an instrument for bombing.

  • While many people have advocated for framing predictions about the future in terms of probabilities, David Leonhardt uses a Kahneman story to explain that most people have difficulty thinking probabilistically, and usually just round up or down to a yes or no. Instead he suggests that stories explaining why things might plausibly turn out one way or the other could work better. Similarly, earlier this year Robert Shiller wrote that economists, lost in numbers, have failed to appreciate the human instinct for storytelling. He went through newspapers before and after the great depression to gauge sentiment at the time, and concludes that economists rarely account for the power of people’s mental narratives to drive economic cycles. But he suggests this is an area where economists might learn something from the humanities.
  • I mentioned a few weeks ago that the U.N. was sending a special rapporteur on extreme poverty and human rights on a tour examining extreme poverty in the U.S. His observations on terrible conditions for the poor in the U.S. were pretty stunning. But he also concludes that one reason for America’s lack of action in the face of such suffering, or even for implementing policies that make poverty worse, is the stories other Americans tell themselves about the poor:

He found that stereotypes serve to undermine the poor — and are used to justify not coming to their aid. “So the rich are industrious, entrepreneurial, patriotic and the drivers of economic success. The poor, on the other hand, are wasters, losers and scammers,” Alston told NPR. As a result, he says, many people believe that “money spent on welfare is money down the drain. Money devoted to the rich is a sound investment.”
He spoke to politicians and political appointees who were “completely sold on the narrative of such scammers sitting on comfortable sofas, watching color TVs, while surfing on their smartphones, all paid for by welfare.”
But Alston says he met people working full time at chain stores who needed food stamps because they couldn’t survive on their wages.

  • In his job market paper NYU’s Kevin Munger describes a randomized trial where he used twitter bots/fake accounts to remind uncivil people to be nicer in political discourse, and it worked. So maybe there’s hope for humanity?
  • Scientists, you’ve been caught. Adding new information to Wikipedia entries led to that new information being cited more in the scientific literature.
    • It also makes me look with more interest at the trial Chris Blattman did at the request of Wikipedia, using his course to have students write or improve Wikipedia articles.
  • One of the better cryptocurrency efforts I’ve seen: you can donate your extra CPU cycles to mine cryptocurrency for the Bronx Bail Fund. (Caveat: I haven’t tried or vetted it yet, and yes I know about the electricity), h/t Colin Rust.