A parametric, model-based, assessment of the post-floating behaviour of the Indonesian currency - October 2000 to March 2003

A parametric, model-based, assessment of the post-floating behaviour of the Indonesian currency - October 2000 to March 2003

The study, in the absence of previous research, tests the post-1997 floating conformity of the Australian/Indonesian (AUD/IDR) exchange rate to international financial parities. That is, the study seeks to provide an initial, three-year, analysis of the post-floating behaviour of the Indonesian currency of early interest to commercial agents. The commencement date of the sample July 2000 was dictated by a required three-year post-floating adjustment period of the Indonesian exchange rate. A modified version of the Guin and Maxwell (1996) model was used to first test for interest rate parities (IRP). Model accuracy was then tested using conventional, scale-dependent, measures based on squared or absolute errors and Theil’s U statistic. The sample period divides into a clear phase of instability of the IDR followed by a period of more sustainable equilibriums. Uncovered interest rate parity (UIRP) suggests that during a declining phase of Indonesian interest rates, with foreign interest rates remaining the same, the rupiah showed a marked tendency to divert from its implied UIRP path. If empirically confirmed, these findings may have significant implications for speculative transactions. However, the interest rate parity null, that mean errors for the sample exceed transaction costs, could not be rejected. The purchasing power parity (PPP) was also tested. The impact of the money supply on Indonesian inflation rates using the quantity theory of money was initially examined. Money supplies could however not be shown to be the primary cause of inflation in Indonesia during the research period. Two distinct and separate methodologies were then used to test for PPP. First, time series were computed and analysed for PPP deviations in terms of their length and magnitude. Standard autoregressive tests followed by unit root tests were then employed to further test findings. First-order differences were also analysed for confirmation of stationarity using conventional Dickey-Fuller tests. As a result, the null hypothesis, that the real exchange rate is non-stationary, could not be convincingly rejected. Policy implications for commercial agents arise from the expectation, given approximate IRP and continued interest differentials between Australia and Indonesia, that an appreciating IDR vs. the AUD is unlikely. It is in fact more likely that a gradual depreciation of the IDR vs. the AUD will occur. Moreover, deviations from PPP throughout the sample period signalled changes in international competitiveness, that is, commercial implications derived from PPP deviations depend on whether the foreign entity in question can be classified as a net exporter.