This dissertation introduces methods to measure the drivers of housing supply and investment across heterogeneous U.S. housing markets, and then offer empirical estimates of their causes and consequences. In the first chapter, I estimate which demographic changes caused postwar suburbs to adopt a widespread land use control --- the minimum lot size. I develop an algorithm detecting bunching on lot sizes, where minimum lot sizes bind developers from building denser housing. Applying the algorithm to national assessor records, I estimate for 7,000 local governments which lot size controls were first adopted there and how they limited residential density over the last 80 years. I find a sizable response in lot size adoption and restrictiveness with just Black population growth from 1940--1970, when four million Black Americans left the South for economic opportunity. The rise in central city Black composition in non-Southern central cities accelerated minimum lot size adoption while explaining binding density controls applied to at least 830,000 housing units. Additional evidence rationalize the effects of Black migration as suburban governments zoning to maximize their neighbourhoods' market values when there was high demand for racially segregated public goods. In the second chapter, I documented segmentation in a major component of residential investment: remodeling projects. I show that households making major remodels live in higher-quality homes within a housing market, as defined by their self-reported home values. Remodeling demand has a price elasticity exceeding 2, suggesting real constraints on investment are responsible for segmentation within markets. I then provide evidence that the real constraints are based on credit risk. A rise in house price indices by one percent raises remodeling propensities as much as a percentage point rise in leverage decreases them.