Business Cycles and Growth Regime Transitions in Morocco: Evidence from a Markov-Switching Model

Résumé

Economic growth in Morocco shows pronounced fluctuations over the long run. Understanding the structure of these fluctuations requires identifying shifts in the underlying growth process. This paper tests whether Moroccan growth is generated by a single stable data-generating process subject to symmetric shocks, or by structurally distinct regimes with different volatility and persistence properties. This paper examines the determinants and dynamics of Moroccan economic growth over the period 1966–2018 using a Markov-Switching Mean and Variance model with exogenous regressors (MSM(H)-X). A formal likelihood-ratio test strongly rejects the linear single-regime specification in favor of a two-regime model, revealing two distinct regimes. The first corresponds to a stable growth phase characterized by low volatility and high persistence (average duration of about 4.8 years). The second reflects a volatile regime marked by rapid fluctuations and stronger exposure to external shocks (average duration of about 3.5 years). Regime-specific volatilities differ by a factor of more than four (σ₂/σ₁ ≈ 4.15), with non-overlapping confidence intervals, providing strong evidence of regime-dependent dynamics. The estimated regimes broadly correspond to the cycles identified by the Hodrick–Prescott filter; the concordance, formally quantified through the hit ratio and Cohen's kappa coefficient, provides external validation of the probabilistic regime classification. Within the MSM(H)-X framework, agricultural output and total factor productivity are significantly associated with national growth. Deterioration in the trade balance exerts a negative influence on economic expansion, highlighting the economy's sensitivity to external imbalances. Estimated transition probabilities indicate strong persistence in the stable regime, whereas episodes of turbulence tend to be short-lived. The results provide evidence of structurally distinct growth regimes in Morocco's growth dynamics and offer a clearer probabilistic characterization of the country's cyclical trajectory over several decades.