Geopolitical Risk, Digital Trust And Exchange Rate Management Under Inflation Targeting: Evidence From Romania

The intensification of geopolitical tensions over the past decade has fundamentally altered the macro financial environment facing central banks in emerging economies, challenging the adequacy of conventional inflation targeting frameworks and their underlying analytical infrastructures. This paper investigates the dynamic relationship between geopolitical risk, exchange rate volatility, and central bank credibility in Romania, a small open economy operating under formal inflation targeting since 2005, over the period February 2015 to December 2025. Using monthly data on exchange rate returns, sovereign spreads, central bank reserves, policy rates, and the Geopolitical Risk Index of Caldara and Iacoviello (2022), we estimate a GARCH (1,1) model with exogenous regressors to assess volatility dynamics, complemented by a VAR (2) framework with impulse response functions and forecast error variance decomposition. The results reveal three findings of analytical significance. First, the conditional volatility of the EUR/RON exchange rate follows an Integrated GARCH process (α + β = 1.000), indicating that geopolitical shocks permanently alter the exchange rate variance regime rather than generating transient perturbations. Second, the impulse response analysis documents a three-month lag in the NBR monetary policy rate response to geopolitical shocks, confirming the reactive character of the inflation targeting framework under conditions of structurally elevated uncertainty. Third, geopolitical risk shocks explain 12.5 percent of the forecast error variance of the policy rate and 24.0 percent of reserve variation at the twelve-month horizon, revealing that the NBR's intervention and rate-setting decisions are substantially conditioned by the geopolitical environment. Building on these results, the paper introduces the concept of digital trust in monetary policy understood as the capacity of central banks to strengthen institutional credibility through the transparency, verifiability, and real-time responsiveness of their analytical processes. By linking the empirically documented analytical lag to the broader TrustTech paradigm, the study argues that the integration of artificial intelligence, machine learning, and high-frequency digital tools into central bank operations is not a discretionary upgrade but an institutional necessity in an era of persistent geopolitical uncertainty. The findings carry direct policy relevance for emerging market central banks at early stages of digital transformation and contribute to the nascent literature connecting TrustTech infrastructures to macroeconomic governance.