What Makes a Landlord? Ownership of Real Estate U.S. Households
31 Pages Posted: 21 May 1997
Date Written: February 1997
This paper seems to be the first positive analysis of the holding of real estate assets by households. Most providers of rental housing in the United States are unincorporated private entities, either households or partnerships. Understanding these landlords is important for understanding the rental housing market.
This study examines the decision to own income property in a non-corporate form, and how much (if any) of one's wealth to invest in such real estate, defined as claims to land and/or improvements that are not used for the owner's own residence and are not held through shares of common stock. Certain characteristics of real estate as an asset should influence a household's decision about how much real estate to hold, and indeed whether to hold any. Disadvantages include illiquidity, lumpiness, high transactions costs, and idiosyncratic risk. Advantages include the opportunity to profit by personal oversight of the asset, low-cost local information, inflation hedging, and leverage. Starting in part from the literature on ownership of stocks and bonds, I offer some preliminary hypotheses about the effects of wealth, inflation fears, earnings, schooling, region, tax rates, health, race and nationality, marital status, and mobility on real estate ownership and portfolio decisions. The data used to test these hypotheses come from the first wave of the Health and Retirement Study. Among 3475 households with more than $100,000 net worth, 40 percent owned some real estate. Use of a two-stage selection model appears to be appropriate, in which a household chooses first whether to own real estate and then what share of wealth to hold in real estate. I use a logit analysis of the decision to hold real estate, combined with a regression of the share of real estate in the portfolio among real estate owners. Wealth is found to have a powerful effect on whether a household owns real estate, but an unclear effect on the amount held as a share of net worth. Sensitivity to inflation, earnings, schooling, region, marginal tax rates, and race all play significant roles in either the decision to hold real estate or what proportion of wealth should be in real estate, or both. By comparison, Ordinary Least Squares results have striking biases. When viewed in context with broad trends in the U.S. economy, especially the rapid rise in household wealth in the stock market and the early retirement trend, the implications of these results for the supply of rental housing appear generally positive. Opportunities for further research are extensive.
JEL Classification: G11, R21, R22
Suggested Citation: Suggested Citation