Determinants of county-level rent levels: a Bayesian model averaging approach

Abstract

ABSTRACT This paper uses Bayesian Model Averaging (BMA) to account for model uncertainty and to identify the robust determinants of county-level rent levels for a panel of 111 counties and independent cities in Virginia from 2008 to 2017. Of the fifteen determinants tested, the results show that single-family home prices, per capita education spending, and educational attainment are robust and positively associated with rent levels, while unemployment rate, population density, labour force participation, and the percent of the population that is Latino are robust and negatively associated with rent levels.