This dissertation focuses on the intersection of municipal finance and household finance. An average resident in the United States lives within the intersection of more than five sub- national government entities, each of which typically has autonomous taxing and budgetary authority. This work explores how fiscal choices made by these local governments affect the financial well-being of constituent households. The first half analyzes issues of equity in taxation and service provision, particularly with respect to local property taxes. The second half asks whether municipal capital structure choices, specifically relating to a particular form of public debt incurred by retirement systems, have a direct impact on household wealth.
The first chapter of my dissertation is coauthored with Carlos Avenancio-Leon. We document large, widespread racial and ethnic inequality in local property taxes. Using panel data covering 118 million homes in the United States, merged with geolocation detail for 75,000 taxing entities, we show a nationwide “assessment gap” which leads local governments to place a disproportionate fiscal burden on racial and ethnic minorities. We show that holding jurisdictions and property tax rates fixed, black and Hispanic residents nonetheless face a 10–13% higher tax burden for the same bundle of public services. For the median minority homeowner, this represents an additional cost of $300–$400 each year. At the 90th percentile of the national distribution, the excess tax burden is $800 per year. The assessment gap arises through two channels. First, property assessments are less sensitive to neighborhood attributes than market prices are. This generates racially correlated spatial variation in tax burden within jurisdiction. Second, appeals behavior and appeals outcomes differ by race. This results in higher assessment growth rates for minority residents. We propose an alternate approach for constructing assessments based on small-geography home price indexes, and show that this would reduce inequality by at least 55–70%. This project provides insight into how institutional discrimination can arise and persist. In this setting, at the intersection of housing markets and government policy, overt racial discrimination has been illegal since the Fair Housing Act of the 1960s. We show that outcomes which are demonstrably not race-neutral can still arise in a setting where policies are explicitly constrained to be race-blind.
The second chapter considers real economic effects of public pension underfunding. Most local governments in the United States sponsor a defined benefit pension plan for public em- ployees. Unfunded public pension liabilities represent a shortfall between contractually pro- tected commitments to future retirees and the assets held in trust to make these payments. As such, they are a particular form of public debt, ultimately backstopped by the taxpayer. State and local governments are subject to balanced-budget requirements, and therefore in- creases in unfunded pension liabilities imply, in expectation, the need to generate additional revenue or to reduce services at some future point. I test whether the increased expected costs represented by a shock to unfunded pension liabilities are capitalized into home prices. Using novel, hand-collected data on assets, liabilities, and fund flows for 200 of the largest county and municipal pension funds in the United States, I estimate whether increases in per-capita unfunded liabilities lower future house price growth. My measure of shifts in unfunded liabilities comes from large investment losses during the Great Recession. To address concerns that investment returns may be correlated with regional economic variables, I construct an instrument for fund investment returns from unexpected returns to broad asset classes. The identifying assumption is that the residual return for any broad asset class (extracted from a standard asset pricing model) is orthogonal to regional economic drivers. Using county-level measures of home price growth, I find little evidence of a strong link between pension obligations and home prices. I show that controlling for endogenous expenditure levels changes the sign of the estimated relationship, which is consistent with budgetary constraints implying a trade-off between spending on goods or services and pay- ments to retirement systems. Using microdata on individual home transactions, I find a clear negative link between pension liabilities and home price growth, along with strong evidence that these effects are larger for more valuable properties. This suggests wealth-heterogeneity in how local residents weigh the benefits of public expenditures against the costs of pub- lic debt, and also shows that municipal financial structure is directly relevant to household financial well-being.