We provide novel estimates of the location, timing, magnitude, and determinants of the start of the previous U.S. housing boom. The housing cycle cannot be interpreted as a single, national event, as different markets began to boom across a decade-long period, some of them multiple times. A fundamental factor, income of prospective buyers, can account for half of the initial jump in price growth, while expansion of purchases by underrepresented minorities cannot. The start of the boom also was financed conventionally, not by subprime mortgages. The latter's share did rise sharply over time, but only after a multiyear lag.

The income distribution in many developed countries widened dramatically from 1970 to 2000. Some scholars argue that income inequality contributes to a host of social ills by undermining voters' willingness to support public expenditures. In contrast, we find that growing income inequality is associated with an expansion in government revenues and expenditures on a wide range of services in U.S. municipalities and school districts. Results are robust to a number of model specifications, including instrumental variables that address the endogeneity of the local income distribution. Our results are inconsistent with models predicting that heterogeneous societies provide lower levels of public goods.