Working Paper: NBER ID: w19604
Authors: Daniel K. Fetter
Abstract: The US home ownership rate rose by 10 percentage points between 1940 and 1945, about half the size of the net change over the 20th century, despite severe restrictions on construction during World War II. I present evidence that wartime rent control played an important role in this shift. The empirical test exploits features of the central authority's method of imposing rent control, which generated variation in the frozen level of rents for cities that had seen similar increases in rents prior to control. Greater rent reductions at the onset of rent control were associated with greater increases in home ownership over the first half of the 1940s. This relationship is not driven by differential trends in housing demand or other unobserved factors potentially correlated with variation in rent reductions. The results imply that if maximum rents had been set at their peak pre-control level rather than at a lower level, the increase in home ownership would have been 10 percent smaller. Combined with new data showing rapid house price appreciation during the war, this result suggests that 10 percent is a lower bound on the share of the "wartime" increase in home ownership that rent control can explain.
Keywords: rent control; home ownership; World War II; housing markets
JEL Codes: N00; N42; N92; R00; R28; R31; R38
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Maximum rents set at peak pre-control levels (E64) | 10 percent smaller increase in home ownership (R21) |
| Wartime rent control (E64) | Increased home ownership rates (R21) |
| 25 percentage point greater reduction in rents (R21) | Increased home ownership (R21) |
| Rent control (R21) | House price appreciation (R31) |