Academic journal article Federal Reserve Bank of Atlanta, Working Paper Series

Housing, Portfolio Choice, and the Macroeconomy

Academic journal article Federal Reserve Bank of Atlanta, Working Paper Series

Housing, Portfolio Choice, and the Macroeconomy

Article excerpt

Abstract: Much of the macroeconomics literature dealing with wealth distribution has become abstracted from modeling housing explicitly. This paper investigates the properties of the wealth distribution and the portfolio composition regarding housing and equity holdings and their relationship to macroeconomic shocks. To this end, I construct a business cycle model in which agents differ in age, income, and wealth and derive utility from housing services. The model is consistent with several facts such as the life-cycle pattern of housing-to-wealth ratios, the larger degree of concentration for nonhousing wealth, and the smaller weight of housing in richer households' portfolios as well as the larger housing-to-wealth ratios in recessions. In addition, the model delivers the familiar business-cycle moments regarding relative standard deviations and procyclicality of consumption, investment, and employment.

JEL classification: E21, E32, G11

Key words: heterogeneity, business cycles, life cycle

1 Introduction

Housing accounts for a substantial fraction of wealth in developed economies. In the US, residential structures, with a nominal value of 11 trillion dollars, account for half of the entire capital stock. Residential investment accounts for a third of total investment and about 9% of output. Moreover, from an individual's point of view a house is usually both an investment good and a durable good that provides a flow of "shelter" services. In the data, renters spend about 20% of their total expenditures on housing.

There exists an extensive literature in macroeconomics that compares the wealth distributions implied by equilibrium models with those observed in the data. Following the tradition of one-sector models, these studies have mostly treated capital as a monolith, effectively subsuming the housing stock into an overall measure of capital. Thus, despite the importance of housing in the US economy, housing investment and the consumption of housing services have been largely absent (with very few exceptions) from studies of the wealth distribution.

Previous studies that do consider wealth distribution properties of real estate holdings include Gruber and Martin (2003) and Diaz and Luengo-Prado (2003) (1). They introduce housing (or durable goods in general) into Aiyagari's (1994) framework to evaluate the effects on the level of precautionary savings as well as the ability of the model to deliver the wealth composition and concentration observed in the data. Both studies feature dynastic agents. Yet, as I argue below, the data show a clear life-cycle pattern in the composition of personal portfolios between housing and financial wealth, making the dynastic framework inappropriate for studying this issue.

In this paper I construct a business cycle model in which agents differ in age, income and wealth and where housing is explicitly modelled. The purpose is to investigate the model's ability to describe properties of the portfolio composition and the wealth distribution and their relationship to macroeconomic shocks.

Age heterogeneity is desirable in any model that deals with housing investment (2). One of the most salient features of a representative individual's wealth portfolio is the age-dependent pattern{young people accumulate home equity before they start accumulating

financial assets. Thus, early in people's lives, housing-to-wealth ratios are large, declining as people accumulate more non-real estate assets, and increasing slightly at the end due to depletion of financial assets during retirement. This pattern is depicted in Figure 1.


To study this feature and others, I construct a standard overlapping generations economy with incomplete markets and macroeconomic shocks, where agents derive utility from non-housing consumption, housing services and leisure. Because of heterogeneity across agents, it is the entire distribution of wealth that determines aggregate non-residential capital and labor and hence, interest rates and wages. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.