THE IMPACT OF EXCHANGE RATE UNCERTAINTY ON FINANCIAL MARKET STABILITY IN PAKISTAN: EVIDENCE FROM EXCHANGE RATE AND STOCK MARKET DYNAMICS
Keywords:
exchange-rate uncertainty; PKR/USD; KSE-100; financial stability; GARCH; EGARCH; volatility spillovers; Pakistan JEL Classification: C32, C58, F31, G12, G15Abstract
This study examines whether exchange-rate uncertainty destabilizes Pakistan’s equity market using verified monthly observations for the PKR/USD exchange rate and KSE-100 Index from January 2010 to August 2026. Monthly log returns are constructed for both markets, exchange-rate and equity volatility are estimated with GARCH(1,1) and EGARCH models, and financial-market instability is measured through a composite index combining conditional stock volatility, downside returns and drawdowns. The empirical strategy further employs heteroskedasticity-robust regressions, VAR-based Granger-causality tests, subsample analysis, lagged-volatility robustness checks and structural-break analysis. Both return series exhibit significant ARCH effects. Exchange-rate volatility is highly persistent, and the EGARCH estimates show statistically significant asymmetry: positive PKR/USD innovations, which represent rupee depreciation, raise foreign-exchange volatility more strongly. In the full sample, however, conditional exchange-rate volatility does not significantly increase KSE-100 conditional volatility or the composite instability index. Rupee depreciation itself is associated with significantly lower monthly equity returns and greater composite instability. The strongest break in the instability relationship occurs around mid-2017, while the post-2018 regime displays a positive and significant exchange-rate-volatility coefficient in subsample analysis. Monthly VAR tests do not establish statistically significant bidirectional volatility causality. The results therefore distinguish the destabilizing effect of depreciation from the less robust full-sample effect of conditional exchange-rate uncertainty and show that the relationship is regime dependent.