Housing's Two Americas, Mapped by Inventory
This chart plots each of the 100 largest U.S. metros by two numbers: current for-sale inventory as a share of that metro's 2018-2019 average, and year-over-year growth in the Zillow Home Value Index. Metros where inventory recovered past pre-pandemic levels are mostly seeing prices fall, and metros where inventory never came back are still seeing gains; the fitted line summarizes how much of the difference this one variable explains.
What does it show?
Where for-sale inventory is back above pre-pandemic levels, home prices are falling; where it never recovered, they are still rising. Inventory recovery explains half the difference in price growth across the 100 largest metros.
Methodology
For each of Zillow's 100 largest metros, for-sale inventory in 2026-07 is divided by that metro's average inventory over 2018-01..2019-12 — the full pre-pandemic span the series offers. That recovery ratio is regressed against year-over-year growth in the Zillow Home Value Index (all homes, smoothed, seasonally adjusted): slope -0.066 points of annual growth per point of recovery (standard error 0.0066), R² 0.51, p 1.6e-16, n = 97. Robustness, actually run: the same ratio measured 12 months earlier predicts the following year's growth with R² 0.62 (slope -0.067, n = 97) — stronger than the coincident fit, so inventory leads at the metro level. The regression is descriptive, not causal: inventory and prices respond jointly to demand. Inventory is smoothed but not seasonally adjusted; the two-year baseline damps but does not remove seasonality. The inventory series begins in 2018, so 'pre-pandemic' means its first two years. Extreme metros: Austin, TX holds 167% of pre-pandemic inventory with prices -5.2% on the year; Bridgeport, CT holds 38% with prices +5.7%. Zillow revises history when methodology changes, so a refresh can move every point. Data provided by Zillow Group; terms verified 2026-07-31.