The regional effects of macroeconomic shocks in China

The extent and persistence of the inequality of regional output is an important policy issue in China and its sources have been the subject of considerable empirical research. Yet we have relatively little empirical knowledge of the effects on the regional distribution of output of shocks to national macroeconomic variables such as GDP and investment. This is an important gap in the empirical literature since much government macroeconomic policy seeks to influence GDP using instruments such as investment expenditure. It is likely that such national shocks will have differential regional impacts and so affect the regional output distribution. Policy-makers need to know the sign, size and timing of such effects before making policy decisions at the national level. We simulate the effects of aggregate shocks on individual provinces' GDP within the framework of a vector autoregressive (VAR) model restricted in a manner following Lastrapes (Economics Letters, 2005). We use annual data from 1980 to 2012 to estimate the model which includes 28 of China's provinces and simulate the effects on provincial outputs of shocks to aggregate output and investment. We find great diversity of effects across the provinces with discernible geographic patterns. There is evidence that output shocks benefit coastal provinces with developed industrial structure, export-exposure and less reliance on SOEs; the opposite is found for the effects of an investment shock and we conjecture that this is likely to have been the result of the strong bias in central government investment policy in favour of the interior provinces during a substantial part of our sample period.