Publication Details
Abstract
This paper investigates the asymmetric nexus between credit expansion and economic growth in Uzbekistan over the period 2010Q1–2024Q4 using the Nonlinear Autoregressive Distributed Lag (NARDL) framework of Shin, Yu, and Greenwood-Nimmo. Employing quarterly data on private sector credit, real GDP growth, inflation, exchange rate, and money supply (M2), we establish a robust long-run cointegrating relationship confirmed by the bounds test (F-statistic = 7.284, p < 0.01). The long-run elasticities reveal a significant positive asymmetry: a 10 percentage-point (pp) increase in the credit-to-GDP ratio raises real GDP by 0.284 pp (β⁺ = 0.284, p < 0.01), while an equivalent contraction reduces GDP by only 0.142 pp (β⁻ = 0.142, p < 0.05). Forecast Error Variance Decomposition (FEVD) indicates that credit shocks account for 35.6% of GDP forecast error variance at the 24-month horizon. Incorporating updated 2026 projections − credit-to-GDP reaching 37% (Kilde.sg, 2026); GDP growth at 7.7% in 2025 and projected at 6.7% in 2026 per ADB ADO April 2026; and IMF 2026 Article IV confirming 7.7% growth in 2025 − our out-of-sample forecast validates model fit. Granger causality runs bidirectionally (GDP → Credit: χ² = 12.84, p < 0.01; Credit → GDP: χ² = 9.62, p < 0.01), consistent with the «finance–growth feedback hypothesis». Policy implications emphasise gradual credit deepening toward the 40% threshold as a growth catalyst while managing NPL risks flagged by Fitch Ratings (2025).