Money, Credit and Spending: Drawing Causal Inferences

It is widely accepted that loans cause deposits. Hitherto, though, the empirical evidence has come from bivariate causality tests which we now know can give rise to invalid inference if either of the two variables is causally influenced by some third, omitted, variable. In this paper we have used tests developed by Toda and Yamamoto to investigate the possibility that earlier inferences were incorrect because of the omission of a third relevant variable, total transactions in this particular case. Including the third variable requires us to revise some of the earlier inferences reported here, Howells and Hussein (1998). The most striking result, however, is that while deposits appear to be caused by total transactions (which could have invalidated the fundamental inference that loans cause deposits) our tests show that even in the presence of a third variable, the core of the endogeneity thesis prevails. Loans do cause deposits.     

Common stochastic trends and inflation convergence in the EMS

Common Stochastic Trends and Inflation Convergence in the EMS. — This article seeks to clarify the relationship between the concept of convergence and univariate and cointegration analysis by looking at inflation convergence within the EMS. We take issue with the view put forward by Artis and Nachane, who hold that cointegration of inflation in Germany with inflation in the other EMS countries is a necessary condition for the “german leadership” thesis to stand. We think that, on the contrary, as long as convergence is still in the process of being achieved, inflation differentials are likely to be non-stationary and, if so, to exhibit common stochastic trends. However, our empirical results, based on Phillips-Perron tests and on the maximum likelihood framework developed by Johansen to test for common trends and cointegration, confirm the validity of the German leadership hypothesis.     

Fractional integration and cointegration in US financial time series data

This paper examines several US monthly financial time series using fractional integration and cointegration techniques. The univariate analysis based on fractional integration aims to determine whether the series are I(1) (in which case markets might be efficient) or alternatively I(d) with \(d < 1\) , which implies mean reversion. The multivariate framework exploiting recent developments in fractional cointegration allows to investigate in greater depth the relationships between financial series. We show that there might exist many (fractionally) cointegrated bivariate relationships among the variables examined, for some of which only standard cointegration tests had previously been carried out.