STABILITY OF MONEY DEMAND FUNCTION IN TUNISIA

International audienceThe aim of this paper is to give an answer to the question that remains wide open; Is Central Bank of Tunisia, capable to transmit all the information on the evolution of prices to economic agents through targeting monetary aggregates. To answer this question, a way that focuses on the stability of the money demand function through the test of Cumulative Sum of residues (SUMCU) and Chow's Forecast test is followed. The results of this study indicate that there is no significant and important relationship between money and prices either in short term or long term. In addition, by estimating the money demand function and its long-term stability, the results show that this relationship is not stable in the case of the Tunisian economy. Introduction The instability of money demand can be explained by the instability of the flow velocity. More frequently, the instability of the application is shown on the factors included in the demand function. Anderson (1985) identified three sources of instability in the demand for money, (i) the change in velocity in response to changes in interest rates as well as movements in other variables the demand function of the currency other than real income, (ii) The demand function of the currency itself may change. For example, financial innovations and releasing the interest rate may change the demand for money, and (iii) short-term monetary stocks actually provided may not correspond to the desired equilibrium. In other words, if the speed of adjustment is small compared to unexpected shocks, this can lead to unexpected changes in the velocity of circulation of money. 1. The relationship between money and prices in the Tunisian economy 1.1. Variables and data As Central Bank of Tunisia (CBT) targets the rate of increase in the money supply within the meaning of M3, then M3 is the variable used to measure changes in the money supply in the Tunisian economy. Concerning the measurement of inflation, changes in the price index (CPI) is the variable most suitable for our study. The adoption of the CPI instead of other measures of inflation is explained by two considerations. The first is the data. Indeed, the monthly data on the CPI are available for longer periods. The second consideration is the wide use of the CPI in the economic literature worldwide. While the CPI may involve some bais, it represents the most widely used measure as an indicator of inflation in empirical studies and analyzes of monetary policy. Regarding the sources of monthly data M2, M3 and CPI, they are issued by the IFC, and the CD-R 2008 IMF