PrefBlog

The Boston Fed has released a Public Policy Discussion Paper by Daniel Cooper titled Did Easy Credit Lead to Economic Peril? Home Equity Borrowing and Household Behavior in the Early 2000s, which dispels at least part of the notion that Americans went insane earlier this decade and spent their home equity loans on beer and prostitutes:

Using data from the Panel Study of Income Dynamics, this paper examines how households’ home equity extraction during 2001‐to‐2003 and 2003‐to‐2005 affected their spending and saving behavior. The results show that a one‐dollar increase in equity extraction led to ninetyfive or ninety‐eight cents higher consumption expenditures. Nearly all of this spending increase was reversed in the subsequent period. A fair amount of these expenditures went toward home improvements and repairs. In addition, households used home equity to help finance their purchases of used cars. Equity extraction also led to some household balance sheet reshuffling. In particular, households who extracted equity were somewhat more likely than other households to pay down their higher‐cost credit card debt and to invest in other real estate and businesses. Overall, the results in this paper are consistent with households’ extracting equity during the first half of this decade to fund one‐time durable good consumption needs.

The author concludes, in part:

Overall, the results suggest that households’ reasons for borrowing against their homes have changed little over time. Households who have lower levels of financial wealth are more likely to extract equity as are households who experience strong local or regional house-price growth. In addition, the consumption analysis suggests that households borrowed against their homes in the early 2000s, to finance one-time consumption shocks. A one-dollar increase in equity extraction led to a roughly ninety-five cent increase in consumption between 2001 and 2003, which did not persist over time. In particular, consumption fell by roughly the same amount in the period ( 2003-to-2005) following the period when households extracted equity.

Additional analysis suggests that households extract equity for one-time, durable goods purchases. Roughly a quarter of each dollar of equity extraction goes toward home repairs and improvements, consistent with anecdotal evidence. Households also extract equity, for used car purchases. There is limited evidence, however, that households’ (non-durable) food purchases increase when they borrow against their homes. The findings also show that households have a roughly 10 percentage point higher predicted probability of paying down their non-collateralized debt when they extract equity than when they do not. Homeowners are also slightly more likely to invest in other real estate or personal businesses when they borrow against their houses. As a result, equity extraction has household balance sheet effects in addition to the effect of households’ borrowing to fund their one-time consumption needs.

The analysis in this paper does not cover the final years of the recent house-price boom, since the PSID data are available only through 2005.

So, while in hindsight it is obvious that American households became over-leveraged during the early part of the housing boom, it is something of a relief to learn that the money was spent on durable goods rather than immediate consumption.

Update, 2009-10-1: The Bank of Canada has released a related study by Ian Christensen, Paul Corrigan, Caterina Mendicino and Shin-Ichi Nishiyama titled Consumption, Housing Collateral, and the Canadian Business Cycle:

Using Bayesian methods, we estimate a small open economy model in which consumers face limits to credit determined by the value of their housing stock. The purpose of this paper is to quantify the role of collateralized household debt in the Canadian business cycle. Our findings show that the presence of borrowing constraints improves the performance of the model in terms of overall goodness of fit. In particular, the presence of housing collateral generates a positive correlation between consumption and house prices. Finally we find that housing collateral induced spillovers account for a large share of consumption growth during the housing market boom-bust cycle of the late 1980s.

This entry was posted on Tuesday, September 29th, 2009 at 1:46 pm and is filed under Interesting External Papers. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.