The real effects of household debt in the short and long run

BIS Working Papers No 607
January 2017

Household debt levels relative to GDP have risen rapidly in many countries over the past decade. We investigate the macroeconomic impact of such increases by employing a novel estimation technique proposed by Chudik et al (2016), which tackles the problem of endogeneity present in traditional regressions. Using data on 54 economies over 1990-2015, we show that household debt boosts consumption and GDP growth in the short run, mostly within one year. By contrast, a 1 percentage point increase in the household debt-to-GDP ratio tends to lower growth in the long run by 0.1 percentage point. Our results suggest that the negative long-run effects on consumption tend to intensify as the household debt-to-GDP ratio exceeds 60%. For GDP growth, that intensification seems to occur when the ratio exceeds 80%. Finally, we find that the degree of legal protection of creditors is able to account for the cross-country variation in the long-run impact.

JEL classification: E21, E44, G21

Keywords: household debt, consumption, cross-sectional autoregressive distributed lag model, output growth, threshold effect