Divisia monetary aggregates: An empirical investigation of their usefulness for Turkey

Money plays an important role in the economic health of a nation. The amount of money in the economy matters because it affects real variables (output, employment) and/or nominal variables (price level). Monetarist view suggests that a target growth rate for a narrow definition of money be set so that a stable (and low) rate of inflation could be achieved. This is possible if the demand for money is stable and changes in the monetary base have a predictable impact on money supply. Therefore, the performance of a monetary aggregate is examined through a money demand specification and/or as an effective monetary target in the conduct of monetary policy. Empirical work in the 1970s showed that broad monetary aggregates like M2 were closely correlated with such potential policy target variables as output, unemployment and inflation. Several countries such as Germany, the United Kingdom and the United States declared monetary aggregates as intermediate targets. By mid-1980s however, the stability of money demand relationship was questioned and the target growth rates of monetary aggregates became unpredictable. Monetary authorities were forced to abandon monetary targeting or monitor monetary aggregates as economic policy indicators along with other variables like inflation and interest rates. The poor performance of monetary aggregates and the instability of money demand functions were attributed to financial innovations, inadequate econometric methodology and restrictive conventional framework of the analysis. Under such circumstances, Barnett (1980) suggested the microfoundations approach and his subsequent work has shown that weighted monetary aggregates could outperform simple-sum monetary aggregates. The theoretical framework of the microfoundations approach derives from monetary aggregation theory and statistical index numbers. This approach constructs new monetary aggregates and compares their performance with simple-sum monetary aggregates. This dissertation empirically investigates the microfoundations approach for four of the G-7 countries (Germany, Japan, United Kingdom and United States) and a developing country (Turkey), using, cointegration analysis, stability test and Granger-causality. My results indicate the superiority of microfoundations approach in all of the countries