UNRAVELLING THE MUSSA PUZZLE: REAL EXCHANGE RATE VOLATILITY AND REGIME CHANGE IN ASEAN-3, 1967–2024

Abstract

Evidence for the Mussa Puzzle in Indonesia, Malaysia and Thailand is weak and depends largely on the 1997–1999 Asian financial crisis. Mussa observed that real exchange rates became far more turbulent once the Bretton Woods parities were abandoned, a pattern that has scarcely been tested in Southeast Asia. This paper asks whether moving from pegged to more flexible arrangements changed real exchange rate movements in the three countries between 1967 and 2024. Real rates are constructed from World Bank nominal exchange rates and consumer price indices, and regimes are coded from documented policy chronologies. Annual logarithmic changes in the real exchange rate form the baseline measure, while absolute changes and a five-year rolling standard deviation capture the size of fluctuations; all models include country fixed effects. The baseline regime coefficient is positive but statistically insignificant, and country patterns diverge, with higher average real depreciation under flexible arrangements in Indonesia and Thailand but lower in Malaysia. The alternative measures give larger positive coefficients, significant only at the 10 per cent level for the rolling standard deviation, and these estimates shrink and lose significance once the crisis years are excluded.