Housing in the Business Cycle and the Life Cycle

Georgetown University, Washington DC

Investigators

Abstract

Housing is an unusual good, both in terms of the way it is produced and the way it is subsequently used. On the production side, residential investment is three times as volatile as non-residential investment, and leads the business cycle whereas non-residential investment lags. On the consumption side, housing is unlike non-residential structures or producer durables in that it is rarely an input to marketed production. Rather housing is perhaps best characterized as a form of capital used to produce at home. In addition, housing is an important form of saving for most households: in aggregate the stock of housing is equal in value to annual GDP and is three times larger than the stock of consumer durables. The goal of this project is to deepen our understanding of the role of housing in the economy. The project consists of two broad components. The first component focuses on the production of new housing and attempts to understand the aggregate dynamics for residential investment, housing prices, and the relation between these variables and standard macroeconomic aggregates. The second component focuses on the role of housing and production at home for understanding behavior through the life-cycle at the household level. These two subprojects can be integrated into a single coherent framework for studying many issues to housing. The analysis takes place within calibrated artificial economies, in which households make dynamic decisions and all prices are determined in general equilibrium. Housing is carefully distinguished from other goods in the economy. More specifically, the question to be ad-dressed in the first part is whether a neoclassical growth model, augmented to include a sector producing new housing, can account for the behavior over the business cycle of residential investment and house prices. A calibrated model of the economy with explicit microfoundations is simulated and its ability to account for the stylized facts about housing at the aggregate level is assessed. Industry-level data will be collected to as-sess the model's ability to account for observed time series for macro aggregates and house prices since WWII. Previous work exploring the empirical determinants of residential investment is investigated in light of the exact equilibrium model relationship between residential investment, house prices and the real interest rate. The second part of the project focuses on the role housing and other con-sumer durables may play in explaining observed consumption and savings be-havior through the life-cycle. Two interesting empirical facts are that a large fraction of households save very little in the form of financial assets, and that household consumption typically drops significantly at retirement. The hypothesis to be explored here is that the gap between observed behavior and the predictions of a standard life-cycle model would disappear once production at home is introduced. Suppose true household consumption exceeds measured consumption, because measured consumption ignores consumption of home pro-duced goods. This measurement issue might shed light on retirement behavior if households substitute working in the market for working at home when they retire.