P-Insight · Fab Housing Thesis

When a chip fab is announced, does local housing actually move?

A common pitch: a multi-billion-dollar semiconductor fab lands in a ZIP code and rents and home values take off. We test that claim directly. For every major U.S. fab with a documented public announce date, we line up the local Zillow home-value (ZHVI) and rent (ZORI) series against the fab's own metro and the U.S. national series — same calendar window — and measure the abnormal growth that is left after the nationwide 2020–2024 boom is differenced out.

Honest verdict
Loading…
The abnormal-growth picture local minus control

Each statistic below is the median, across the studied fabs, of local cumulative growth minus control cumulative growth over roughly three years after the announcement. A positive number means the fab's ZIP outran its control; zero means it tracked it. The metro comparison is the cleanest test — it nets out regional housing conditions.

Read this honestly: a median home-value tilt of a couple of percentage points over three years is real but tiny. It is the same order of magnitude as the normal spread between any two ZIPs inside a metro. Rent shows no consistent fab premium at all. We can see a faint, mostly-positive lean — we cannot call it a clean, reliable fab effect.

Per-fab event studies
Fab ZIP
Metro
U.S. national
Announce month (index = 100)
Why this is hard — and where we stay honest caveats
Method event study