Abstract: This study examines the relationship between foreign aid and the real exchange rate in Rwanda using annual data for 1990–2023 and a five-variable Vector Error Correction Model (VECM). The principal system contains the real effective exchange rate (REER), aid relative to gross national income (AID/GNI), terms of trade (TOT), trade openness (OPEN), and government expenditure relative to GDP (GOV/GDP). The empirical strategy uses Johansen maximum-likelihood cointegration with one lag in first differences and a restricted constant in the cointegrating relation. The trace test identifies one cointegrating relationship. In the normalized long-run vector, AID/GNI is 0.895 (SE 0.069; p<0.001), while TOT and GOV/GDP are positive and statistically significant and OPEN is negative and statistically significant. The REER adjustment coefficient is −0.230 (p<0.001). Short-run aid changes are negative but not statistically significant at the 5 percent level. Sensitivity specifications using intervention pulses for 1994 and 2002 show that the estimated aid relationship is sensitive to treatment of the exceptional 1994 disturbance, whereas the 2002 intervention has little effect on the principal coefficients. The findings document a statistically significant long-run association between aid and the REER but do not by themselves establish a causal Dutch-disease effect.