The demand for money: evidence from the Tanzanian money market (1965-1976)

The Tanzanian money market has distinctive characteristics which distinguish it from money markets in developed capitalist countries. In Tanzania, interest rates are institutionally pegged and are therefore not allowed to fluctuate. In contrast, in developed capitalist countries, interest rates. are determined by the interplay of demand and supply forces. In addition, in Tanzania, the range of alternative financial assets is very narrow. There is very little in between money on the one hand and physical goods on the other. The aforementioned characteristics of the Tanzanian money market do imply that (1) interest rates do not determine the demand for money balances in Tanzania. Other variables like income and the price level are likely to be significant determinants of money holdings.(2) The linkage between the real and monetary sector is not via interest rates. (3) The impact of changes in the quantity of money would not percolate among various money substitutes but would be transmitted directly to the market for physical goods, The monetary policy implication of this assertion is that in Tanzania, changes in the quantity of money would have a direct impact on planned expenditures on physical goods. The effects of changes in the quantity of money would be swift and strong. This study investigates the roles of real income and prices changes in determining real money holdings. Three definitions of money are used in the specified money demand functions. The first definition of money refers to only currency outside banks. The second definition of money includes currency outside banks plus privately held demand deposits. The third definition of money refers to currency outside banks plus privately held demand deposits plus time and savings deposits at commercial banks. The period of study is 1965-76, Money demand functions are cast in real terms. Annual data are used. The results reveal that both income and price changes are significant determinants of real cash holdings. The non-linear money demand model is found to be more appropriate than the linear model; on the basis of goodness of fit and the size of the standard errors of estimates; non-linear money demand function (Without logged variables) utilizing a definition of money short of privately held time and demand deposits gives the best degree of fit and the lowest standard errors of estimates of the parameters. A partial adjustment model, which incorporates the assumption that the money market is not in equilibrium was found to apply to the Tanzanian monetary experience.