How important is leverage? Investigating the link between consumption and leverage using two samples

Presenter: Peter Levell, Institute for Fiscal Studies

In this paper we estimate the degree to which leverage amplifies the effects of house price shocks on consumer spending. We do by using instrumental variable methods that are able to combine the information in two datasets. One is a panel with information on household balance sheets which does not include consumption spending. The other is a survey with detailed consumption data that does not include information on wealth. In addition, we show how existing two sample methods can be extended to estimate dynamic relationships between leverage and consumption changes despite only observing repeated cross-sectional data on consumption. We find evidence that each 10 percentage point increase in leverage increases the size of housing wealth effects by roughly 0.02. Effects are larger for spending on durable goods.