[Originally a Twitter thread.]

I was inspired by Greg Allen’s analysis of US export restrictions to China to look up British attempts to control high-tech exports during the industrial revolution. I found David Jeremy’s article on export controls from 1780-1843. Mini-thread with random observations:

What struck me most was the profound difference in state capacity between the UK then and the US now, and how it affects the state’s ability to make export controls “a new strategic asset,” as Jake Sullivan put it. The UK’s customs and security agencies were so inadequate to the task of deterring and catching smugglers that most restrictions were eventually abandoned as unenforceable.

One vivid illustration: today even many foreign firms bend over backwards to comply with US restrictions for fear of penalties. Then, heavy machinery was illicitly loaded at major ports and insurance policies were available to cover the risk of discovery!

Britain’s diplomatic and intelligence services were incapable of consistently identifying law-breakers abroad, with the exception of an energetic consul in Philadelphia who once personally conveyed his reports to a British transport to prevent American merchants from intercepting them. The relative simplicity of the technology also made controls hard. Semiconductor-related equipment is made of thousands of different parts, many of which are themselves exquisite tech produced in third countries. Not so a spinning jenny.

I was most interested in the importance - and difficulty - of gauging the implications of technological change. Jeremy tells of a shift in the locus of technology from individual artisans, to small teams of specialized workers, and then to machine tools. But British policy did not keep pace with the growing importance of blueprints and hardware.The people became less important.

I’m a big believer in the importance of tacit knowledge but I’m open to the idea that technological change may make it less important in some fields. Similarly, we may learn that various strands of AI or quantum technology are more or less important than anticipated.

I’m no expert on any of this, but it gave me a new appreciation for the last point in Allen’s analysis: much relies on the capabilities of the bureaucracies that will enforce the new restrictions and the adaptability of the regulatory regime.