THREAD (~50 tweets). My potted history of institutions in macro-development thinking, at the time of the AJR intervention. I had most of these thoughts earlier, but this week they have crystallised. I don't discuss AJR specifically, but I do have brief thoughts on their impact.

I don't think anyone has ever quite expressed it as such, but Richard White's 2011 history of the transcontinentals is tantamount to a scathing indictment (from a left-populist perspective) of a vastly socially costly industrial policy failure & misallocation of resources.

Just read a (non-economics) working paper which identifies a big puzzle but then brags about solving it at long last. Yet there's nothing of the kind. The proposed 'solution' is so trivial as to merely restate the original puzzle and push the mystery farther back into the past.

When cost (including opportunity costs) did not matter, the USA massively overtook the rest of the world in shipbuilding market share.

but the PRC deliberately ate into Japanese and Korean market shares for strategic reasons! The welfare effects are irrelevant!. The USA too can increase global market share if welfare just didn't matter!

From the Barwick et al. paper that's being widely cited, one can make the case that (1) Chinese subsidies stole market share away from the world's most efficient ship producers (Japan & Korea) so (2) global welfare was reduced even if Chinese welfare may or may not have increased.

The current industrial policy discourse is so strange. Most people who argue for it today, do it on the grounds of strategic & geopolitical necessity, so welfare effects are irrelevant. But other people who support it keep debating the welfare effects!.

Another interesting finding from the same paper is the comparison of racial income gaps in the USA v France. Racial income gaps are smaller in France primarily because the overall income distribution is more compressed. But positional inequality is similar between the 2 countries

Frenchmen of Indochinese origin have the smallest income gap with the white 🇫🇷 pop. In fact 🇫🇷-born Asians have higher incomes than whites, except when controlling for education & location (i.e. 'should' have even higher incomes since more educated & cluster in richer regions)

Congratulations to @dkedrosky & @nunopgpalma on their "Brazilian gold influx impoverished 18th c Portugal through Dutch Disease & delayed industrialisation via a political-institutional resource curse" paper. But double congratulations to Davis for his first academic publication!

Taiwan's machine tools industry. Had it very early on, whilst it was still an agricultural exporter.

Development is indexed by GDP per capita. Japan ca 1930 had about the same income as, say, El Salvador in the 1970s. Such a poor country capable of building a war machine so "early in its development" did not climb the ladder so much as leapfrog it!.

i.e., there is NO "ladder of development". You have countries which originate frontier technologies, plus follower countries which rapidly catch up to the frontier & start contributing to it, plus countries who receive the tech & almost never expand the frontier.

People forget that Japan -- to fight a war it was destined to lose -- built its own massive arsenal of tanks, ships, aircraft, bombs, etc. By the 1930s, Japan was no longer importing tools & equipment to build such things. They already had a big capital goods sector.

And it's another reason the United States was never a 'developing country' in technological capabilities! It has always had a capital goods sector, early in the 19th century. The USA was capable of supplying its own machinery from almost the very beginning of industrialisation.

H/T for the paper here

The capital goods sector is a very leading indicator of later development. If you're producing industrial equipment "early in your economic development", that's a sign of what's to come later. But most countries don't do this, in fact, even middle-income countries.

( Can't find this now but I once had a short thread on Taiwan's machinery industry which started exporting machines across Asia in the early 1970s ).

All countries which manufacture specialised capital goods, already had an equipment-manufacturing industry quite "early in their development". Japan started producing automatic looms in the 1930s. China was already exporting machines in the 1990s (to other poor countries).

Not exactly shocking. Even after substantial deindustrialisation, richest countries dominate the manufacturing of specialised equipment. But there is a belief that countries "move up the ladder of development" into making capital goods. But nothing is farther from the truth!.

Disaggregate deindustrialisation by sector!. Over the course of economic development, the industrial share of employment & value added has an inverted U shape in the aggregate, & this shape is repeated in each subsector, EXCEPT one (capital goods)