Housing and consumption: the home-purchase connection
Research linking the housing market to household consumption—and, by implication, to cyclical movements in the nation’s economy—has typically focused on the mechanisms through which housing prices affect consumer spending. At the same time, much less attention has been paid to whether home purchases affect consumer spending. Does buying a home spur household consumption independently from movements in home prices? If so, what is the magnitude, composition, and timing of that spending increase?
These are the questions addressed by economists Efraim Benmelech, Adam Guren, and Brian T. Melzer in their article “Making the house a home: the stimulative effect of home purchases on consumption and investment” (National Bureau of Economic Research, working paper no. 23570, July 2017). The authors argue that buying a home induces greater household consumption by increasing demand for durable goods related to the home. This effect is attributed to frictional effects whereby homebuyers unable to find a house that meets their specific preferences boost spending on home improvements and other durables, such as furnishings and appliances.
To assess the empirical plausibility of this argument, the authors use expenditure data from the U.S. Bureau of Labor Statistics Consumer Expenditure Survey and data on housing permits from BuildFax. The analysis focuses on the 2001–13 period, which encompasses the Great Recession of 2007–09. To estimate the timing and magnitude of spending changes due to a home purchase, the authors employ an event-study methodology, controlling for household fixed effects. Because house prices and home purchases tend to be correlated, a fixed-effects model allows for separating their independent effects.
The authors’ empirical tests lend credence to the hypothesized relationship, indicating that a home purchase is associated with an average spending increase of $3,700 over the period from 3 months before the purchase to 12 months after it. Consistent with expectations, this money is spent on home-related durables, home improvements, and maintenance. Consumption increases the most within the first 3 months after the purchase and levels off after the first year.
There is no evidence that the increased spending on home-related durables is met by a comparable decline in purchases of nondurables. According to the authors, this result suggests that (1) the effects revealed by the empirical tests are causal and (2) concerns about omitted variables (e.g., a sudden increase in wealth or a job promotion) would be unfounded because if such variables were to affect consumption, they would do so across both durable and nondurable categories.
In discussing aggregate impacts, the authors also provide some noteworthy insights into how the home-purchase–consumption channel played out during the Great Recession. In theory, a broad-based decline in home purchases would have translated into a decline in aggregate consumption. Indeed, the authors find that this channel operated strongly during the recession. Between 2005 and 2010, the number of home purchases dropped by a whopping 50 percent, and this slump in real estate transactions was paralleled by a 12-percent fall in spending on home durables and a 28-percent fall in spending on home improvement and maintenance. Overall, the authors assess that this downward spiral cost the economy $14.3 billion in lost spending during the recession.