Heterogeneity in Regional Consumption due to Current Income Shocks: Testing the Permanent Income Hypothesis for German Panel Data

Please do not quote without permission. Thanks! In this paper we analyse the role of current income shocks on regional consumption patterns for a panel of German regions (NUTS1 level) between 1970 and 2007. The basic workhorse model of the analysis is the Permanent Income Hypothesis, which has recently been subject to new methodological improvements regarding the specification of expectation formation. Using both a short-run as well as cointegration perspective we tackle earlier findings for German regional data, which find a significant degree of ”excess sensitivity ” of consumption to income shocks. However, our results do not support these findings. Instead, in the short-run approach the influence of past and current income changes turns out to be significant if we control for potentially ommitted variables. Using also long-run information in a Panel Error Correction Model (ECM) approach we indeed find a significant share of liquidity constrained households. However its share is below the fraction recently found for German regional data. These results may give new insights with respect to the effectiveness of policy measures for short-term consumption stabilization as being raised in the policy debate to combat the global fincancial and economic crisis (e.g. so-called consumption vouchers or tax rebates). Additionally, by testing for the asymmetry in the consumption patterns for German regions both in the short and long run, we are able to give an initial answer to the question dealing with the regional effects of the world economic and financial crisis. Here we apply different testing approaches for slope homogeneity of the regions in the panel and aim to identify different long- and short-run adjustment regimes in Germany. We finally also account for the likely role of spatial autocorrelation when dealing with territorial data