Place-based industrial policy and semiconductor clusters
Industrial policy and place-based policy are both rising in the economics journals, and governments now subsidize chip plants in specific towns, from Arizona to Kumamoto. Yet only 4 of 127 recent papers on such policies mention semiconductors. Do these plants create lasting local clusters, or expensive islands?
What prompted this
In economics, "industrial policy" had 17 papers in July–September 2026 against 11 expected and "place-based policy" 10 against 4; across all journals "innovation policy" grew too (30 against 16); see the trends page. Since October 2025, 90 papers carry an industrial-policy keyword (59 in economics journals), and 13 of them carry a place-based, cluster or regional keyword. Of the 127 papers with either kind of keyword, only 4 carry a semiconductor or chip keyword.
Questions
- Around new subsidized fabs, do local suppliers, wages and new firms grow beyond the plant itself, and for how long?
- Do housing costs and labour shortages eat the gains, as in some earlier large-plant cases?
- Do the subsidies change where the next plants and suppliers locate?
Data that could help
Plant announcements and subsidies (CHIPS Act awards in the US, Japan's support for fabs in Kumamoto and Hokkaido), local employment and business registers, land prices and patents.
Next step
Compare the municipalities around the TSMC-led plant in Kumamoto with similar municipalities, before and after its 2021 announcement, on employment, new firms and land prices.