Regime-Dependent Exchange Rate Volatility in Nigeria: Persistence, Clustering and Implications for Growth Stability

This study investigates whether exchange-rate volatility in Nigeria is regime-dependent, and how its persistence and clustering bear on growth stability over the period 2000Q1–2023Q4. The sample is partitioned into a broad-based growth regime (2000Q1–2014Q4), when output expanded at roughly seven per cent annually, and a slowdown regime (2015Q1–2023Q4), when growth decelerated and per-capita output flattened. An Exponential Generalized Autoregressive Conditional Heteroscedasticity (E-GARCH) model is used to characterise volatility persistence and clustering in each regime; a Non-linear Autoregressive Distributed Lag (NARDL) model captures the asymmetric long- and short-run effects of positive and negative volatility shocks on growth; and a threshold regression identifies the lending-rate ceiling consistent with growth. Augmented Dickey–Fuller and Zivot–Andrews tests confirm a mixed I(0)/I(1) integration order and an endogenous break around 2015. The E-GARCH estimates reveal statistically significant ARCH and GARCH effects in both regimes, with volatility persistence markedly higher and covariance-non-stationary in the slowdown regime, indicating that shocks become near-permanent once growth weakens. The NARDL results show that the long-run impact of overall volatility on growth is negative and significant, while the partial-sum responses are asymmetric: positive volatility raises growth by about 0.51 per cent and negative volatility lowers it by about 0.10 per cent, though neither partial effect is individually significant at five per cent. Interest rate exerts a significant negative long-run effect on growth, and the optimal lending-rate threshold is estimated at 26.69 per cent. The findings imply that volatility, rather than its direction, is the binding constraint on growth stability. The study recommends export diversification, unification of the exchange-rate windows under a market-based regime, and anchoring the policy rate so that market lending rates remain at or below the estimated threshold.