Maybe but the Japanese capture of the memory market was just as much about low costs of capital. Semiconductor manufacturing is about as capital intensive an industry as exists.

I'd note that the Koreans now own this market. Same strategy. Commoditized industry thus barriers to technological entry are not high - and low costs of capital.

I lived and worked in Japan for 6 years as a software engineer and left speaking and reading Japanese well. However I was in a joint venture quantitative investment mgmt firm and not manufacturing. Finance has its own particular views over, actually, all industries. You'll note that an important measure in the paper is stock prices.

I might add to the author's analysis: with all these countries (Japan, Taiwan, Korea and China) you have to watch the demographic curve and in particular see just how many new university graduates arrive in the economy each yr. That curve is now declining for all them - but did it start declining for Japan earlier than Taiwan and Korea? Or is the time frame involved (since the GFC) too short?

Traditional economics saw inputs as: land, labor and capital. Current understanding is more nuanced but that doesn't invalidate l, l and c. Less labor, less growth.

Yields are a closely guarded secret for semi companies, since that determines the lowest price a company can sell at and still make some money.

High interest during the 80s really hurt a lot of US companies, though. One example is Caterpillar vs. Komatsu for construction equipment. Cat had a price premium in early 80s, then interest rates were super high, the dollar popped, and that created a window for Komatsu to grab market share.