THE VICIOUS CIRCLE HYPOTHESIS: FACT OR FANTASY?

After generalized floating was introduced in 1973, some countries underwent rapid exchange rate depreciations which were often translated into higher growth rates of prices and wages. The latter often led to further depreciations in the exchange rate and that would set off another round of price and wage hikes. It was said that this was a self-perpetuating cycle. The problem was particularly acute for countries which has some degree of formal indexation of wages to prices. Countries experiencing this cycle were dubbed "vicious circle countries.". Many economists have argued that such a self-perpetuating cycle cannot exist without monetary accommodation. We agree with this reasoning. This dissertation investigates the hypothesis that an economy adjusting from one steady-state to another in response to some exogenous shock may have the ability to display vicious cycle symptoms as a segment of the adjustment process. The symptoms, in the context of the present model, are a rising rate of unemployment, rising growth rate of prices and a depreciating exchange rate. The possibility of cyclical behavior is ruled out by assumption, consistent with the neoclassical, flexible price model that is used. Also, the money supply is exogenously determined so that monetary accommodation will have to be result of discretionary policies. A well known closed economy model is extended to the open economy in order to test this hypothesis. It has the ability to display vicious circle symptoms in the short-run but is assumed to be stable in the long-run. The reduced form macrodynamic system consists of the unemployment rate, growth rate of prices and expected prices as the state variables and the growth rate of the money supply, changes in real government expenditure and the ratio of private financial wealth to the stock of money as control variables. The model is estimated (using quarterly data) for the U.S. for the 1970's, after being converted into a set of empirically testable hypotheses. It is shown that the reduced form system is quite successful in tracking the actual data and that the model is indeed stable. The system is then brought to a steady-state and the exogenous shock is introduced via an increase in the growth rate of the money supply. It is shown that vicious circle symptoms are displayed by the model during a particular segment of the adjustment process just prior to reaching the new steady-state. Noteworthy features of this exercise are the long-lags involved in the adjustment cycle (about 9 years from the time of the shock to the steady-state) and also the prolonged period of time through which the vicious circle phase lasts (about 14 quarters). Also, recent developments in the adjustment theory literature such as exchange rate overshooting and asset markets adjusting faster than goods markets are evident from these exercises. Furthermore, they indicate that the model is very sensitive to changes in the growth rate of the money supply and that the exchange rate and unemployment rate are (in that order) the quickest to respond to the shock.