On Testing the Exogeneity Specification Underlying the Monetary Approach to Balance of Payments: The Indian Case

The Monetary Approach to Balance of Payments (MABP) has triggered off a controversy over fundamental proposition of exogeneity of price level, real income, interest rate, money multiplier and changes in domestic credit with respect to reserve flows. The present study investigates into the validity of the exogeneity assumptions under-lying the monetary model in three different ways. Firstly, a simultaneous estimation of the reserve flow and sterilization equations is made to confirm the absence of sterilization. Secondly, using three causality tests viz., the Granger, Sims and Multiple Rank F tests, bi-variate causality is examined. Finally, a multivariate systems test of causality within the framework of a Complete Dynamic Simultaneous Equation Model (CDSEM) is applied to test the joint exogeneity of the right hands side variables in the reserve flow equation. The findings of the study which are based on Indian annual data relating to the period 1950-90, by and large, do not support the exogeneity specification. While the causality tests collectively fail to provide any definite pattern of pair-wise causality, the OLS and 3SLS estimates of the reserve flow and sterilization equations, and Wald and likelihood ratio tests of the CDSEM, clearly reject the exogeneity assumptions.