Thread by @Noahpinion

1/In today's @bopinion post I yell at Lant Pritchett for yelling at Banerjee, Duflo and Kremer:

bloomberg.com/opinion/articl…

2/Economist Lant Pritchett unleashed a ferocious smackdown on the three Nobel winners in a recent Facebook post.

This is a level of shade I haven't seen top economists throw since the Macro Wars of 7 or 8 years ago.

facebook.com/lant.pritchett…

3/This smackdown garnered a surprising amount of approval from #EconTwitter:

https://twitter.com/JWMason1/status/1191933156137619459

4/In a nutshell, Pritchett's argument is that the only thing that really reduces global poverty is economic growth, which RCTs can't test, so by using RCTs, Banerjee Duflo and Kremer hurt the field of development economics.

5/So, on one hand, Pritchett has a point, which is that when poor countries become richer, poverty falls.

If you want your people to be less poor, make your country less poor.

(Note: Everyone who pooh-poohs GDP as a measure of human welfare needs to reckon with this fact.)

6/So yeah, if we knew how to turn poor countries into rich ones, we'd have a magic wand. It would be much much much more powerful than any of the little targeted health and education stuff that people like Duflo and Banerjee are doing.

Yay for magic wands!!

7/But we don't have a magic wand, do we?

Yeah, economists make big grandiose growth models. But most of these - like the Solow Model - are more about telling us what we CAN'T do than what we CAN do. (Solow model tells us we can't get rich just by saving more.)

8/Growth models typically end up just applying some label to the "magic stuff" that makes poor countries into rich ones.

"Technology"!

"Human capital"!

They don't tell us how to actually get more of these things.

econ2.jhu.edu/people/ccarrol…

9/(One exception is Paul Romer's endogenous growth model, which tells us to spend more money on research. But that's a prescription for rich countries, not poor ones! Research spending is not what's going to make Tanzania get rich.)

10/How do we get more "human capital"? Universal public education, obviously. Every single poor-country government knows this. Every single development agency knows this. They're all trying to do it already.

Also we know that that, by itself, isn't enough to make a country rich.

11/What about foreign aid? Is that a magic wand that makes poor countries rich?

Jeff Sachs bet his legacy that it was. But it wasn't.

psmag.com/social-justice…

12/How about market liberalization? Is that a magic wand that makes poor countries get rich?

The legacy of the "Washington Consensus" suggests that no, it is not.

web.stanford.edu/group/scspi/me…

13/How about industrial policy? Is that a magic country-enrichening wand?

Possibly. Right now it looks like the best candidate (not that many economists are studying it or recommending it!).

But we're still a long way off from a general model.

imf.org/en/Publication…

14/So Lant Pritchett would have development economists chuck their rink-dink little education and health policies and spend their careers searching for the Magic Wand, the Ultimate Secret of how countries get rich.

Given the track record, is that a valuable use of time??

15/Yeah, we shouldn't completely give up the search for the Ultimate Secret. But having every single development economist chase after that dream is just going to lead to a lot of economists bellowing at each other in seminar rooms in support of their own pet Big Ideas.

16/The anti-RCT people always accuse the RCT people of being like a drunk looking for his keys under a lamppost because that's where the light is.

But what Pritchett wants economists to do seems more like the drunk forgetting about his keys and looking around for a free car!!

17/It's like telling engineers that by studying how to improve the efficiency of solar cells by 0.5%, they've given up on the quest for fusion power.

Should every engineer be working on the holy grail of fusion? Is that efficient? No.

18/Pritchett's proposed program feels like hubris. The work of Duflo, Banerjee, and the other "randomistas" is humble and incremental, like the work of a plumber.

There's absolutely nothing wrong with economists being like plumbers.

19/It makes perfect sense for lots of development economists to focus on the small stuff. To develop programs that help real people in tangible, verifiable ways.

Sneering at anyone who doesn't quest after the Ultimate Secret is not a good look.

(end)

bloomberg.com/opinion/articl…