Expenditure Cascades, Low Interest Rates or Property Booms? Determinants of Household Debt in OECD Countries

Abstract

The past decades have witnessed a strong increase in household debt and high growth of private consumption expenditures in many countries. This paper empirically investigates four explanations: First, the expenditure cascades hypothesis argues that an increase in inequality induced lower income groups to copy the spending behaviour of richer peer groups and thereby drove them into debt ('keeping up with the Joneses'). Second, the housing boom hypothesis argues that increasing property prices encourage household spending and household borrowing due to wealth effects, eased credit constraints and the prospect of future capital gains. Third, the low interest hypothesis argues that low interest rates encouraged households to take on more debt. Fourth, the financial deregulation hypothesis argues that deregulation of the financial sector boosted credit supply. The paper tests these hypotheses by estimating the determinants of household borrowing using a panel of 11 OECD countries (198011 OECD countries ( -2011. Results indicate that real estate prices and low interest rates were the most important drivers of household debt. In contrast the data does not support the expenditure cascades hypothesis as a general explanation of debt accumulation across OECD countries. Our results are consistent with the financial deregulation hypothesis, but its explanatory power for the 1995-2007 period is low.

The drivers of household indebtedness reconsidered: An empirical evaluation of competing arguments on the macroeconomic determinants of household indebtedness in OECD countries

“…The other standard measures of income inequality, the Gini coefficient is less useful in this context because it is more sensitive to the middle rather than the upper tail of the distribution, which is key for the expenditure cascades hypothesis. Therefore, the income-share of the top 1% is standard in the empirical literature as a proxy for the expenditure cascades hypothesis (see, for example, Malinen, 2014;Klein, 2015 andWildauer, 2017). In addition to this norm in the related which is in line with the low interest rate hypothesis.…”

“…The house price hypothesis is tested by using house prices as an explanatory variable for household debt both in panel (Rubaszek and Serwa, 2014;Stockhammer and Wildauer, 2017) and time series work (Kohn and Dynan, 2007;Oikarinen, 2009;Gimeno and Martinez-Carrascal, 2010;Valverde and Fernandez, 2010;Anundsen and Jansen, 2013;Meng et al, 2013). The econometric approaches used are panel co-integration (Rubaszek and Serwa, 2014), error correction models (ECMs) (Stockhammer and Wildauer, 2017), in the panel econometric studies, and a vector error correction model (VECM) and Granger causality tests (Oikarinen, 2009), co-integrated vector autoregressive models (CVAR) (Gimeno and Martinez-Carrascal, 2010;Anundsen and Jansen, 2013;Meng et al, 2013) and OLS regressions (Kohn and Dynan, 2007), in the time-series studies. Despite the differences in sample and method used, all the papers find a positive and significant impact of house prices on household indebtedness.…”

“…Nonetheless, one can interpret this as evidence in support of the low interest rate hypothesis. Chrystal and Mizen (2005), Oikarinen (2009), Gimeno and Martinez-Carrascal (2010), Valverde and Fernandez (2010), Anundsen and Jansen (2013), Rubaszek and Serwa (2014) and Stockhammer and Wildauer (2017) also control for interest rates in their econometric testing of household indebtedness. 5…”

“…That is, we allow for the building up and deleveraging of debt that may be influenced by the boom and bust periods of household debt and house price cycles. Only Stockhammer and Wildauer (2017) have explicitly catered for asymmetric behaviour in household debt accumulation thus far.…”

“…He tests the effects of the top 1% income share, which corresponds to a testing of Frank et al's argument, as well as the effects of the inverted Pareto-Lorenz coefficient, the wage share and the Gini index on household debt accumulation, but he does not consider any of the other existing arguments. More recently, Stockhammer and Wildauer (2017) test the arguments that the top 1% income share, house prices, financial deregulation, and the age structure a country's population drive household debt.…”