Weryfikacja hipotez neutralności i superneutralności pieniądza w Polsce w środowiskach procesów I(1) oraz I(2)

Abstract

The doctoral dissertation verifies the hypothesis of long-run neutrality (LRN) and long-run superneutrality (LRSN) of money using a new research tool in the form of a cointegrated VAR (CVAR) model for second-order integrated processes - I(2). The LRN and LRSN hypotheses address the effects of changes in the money supply and its rate on the money market itself, the goods and services market, the labour market and the foreign exchange market. They indicate what short- and long-term effect the (mainly) expansionary monetary policy of the central bank has on the economy. The effects of changes in the money supply and its rate are viewed differently by schools of economics. So far, the LRN and LRSN hypotheses have been verified using mainly the SVAR model. The analysis of the LRN hypothesis was carried out, on the one hand, by imposing exclusion restrictions on the parameters in the cointegration space and, on the other hand, by applying exclusionary restrictions on the parameters in the weight matrix of common stochastic trends I(2) B ̃_(2⊥), in which the import of I(2) shocks is determined. Using polynomial cointegration analysis in the CVAR model for I(2) processes, it is shown that the detailed hypotheses of money neutrality and super-neutrality can be analysed simultaneously. Polynomial cointegration also made it possible to analyse medium-term equilibrium relationships that perpetuate over the long term. Using a matrix of coefficients defining common underlying stochastic trends I(1) A_(1⊥), the main sources of cyclical behaviour in the Polish economy were identified.