The paper studies the behavior of the household saving rate during the period from 1985 to 2016. We examine the variables considered by the literature to determine the long-term saving rate: Income, wealth, unemployment, credit and indebtedness, public savings, population structure, social benefits, interest rates and prices. In the framework of the error correction model, we contrast the cointegration of the variables by means of four specifications for the saving rate according to its long-term determinants: Permanent income, wealth, unemployment, credit, dependency rate and social benefits. The results show the special long-term relevance of the dependency rate, which, as a structural demographic variable, allows us to predict the evolution of the saving rate. Equally important, social benefits show the relevance of social legislation and its impact on the survival of the welfare state. In the background, credit and unemployment determine the long-term saving rate at any stage of the economic cycle. The growth of labor income affects the saving rate especially in cycle changes when expectations on permanent income are modified. We find that wealth is less relevant. Finally, we find unconsolidated evidence about the fact that imbalances on the long-term rate are corrected in the short term through changes in labor income, wealth and credit.
error correction model
EconStor: OA server of the German National Library of Economics - Leibniz Information Centre for Economics