한국 통화정책의 비대칭적 반응과 중앙은행 손실함수 가중치 추정

Asymmetric Reaction of Monetary Policy and Estimation of Central Bank Loss Function Weights in Korea

초록

This paper empirically investigates the conduct and evolution of Korean monetary policy since the early 2000s within an augmented New Keynesian framework. Going beyond the conventional estimation of a Taylor-type reaction function, this research implements an inverse-optimization approach to identify the time-varying weights in the Bank of Korea’s (BOK) loss function. By internalizing the structural parameters of the Investment-Savings (IS) and Aggregate Supply (AS) relations, the study reveals how the central bank’s latent preferences have evolved alongside structural changes in the domestic and global economy. The theoretical model is grounded in a small open economy (SOE) framework that incorporates households’ dynamic optimization, firms’ forward-looking price-setting, and explicit interactions with the foreign sector. Given Korea’s structural characteristics, the model accounts for policy sensitivity to the U.S. federal funds rate and exchange rate fluctuations. Furthermore, the inclusion of a Financial Stress Index (FSI) in the IS relation allows the model to capture the impact of financial frictions and risk premiums on aggregate demand. The empirical analysis yields several critical findings. First, consistent with Korea’s status as a small open economy, the BOK demonstrates a high degree of interest rate smoothing (above 0.9) and significant synchronization with U.S. monetary policy. The persistent significance of the FSI and external interest rates suggests that financial stability mandates and global financial cycles operate as fundamental constraints on domestic policy discretion. Second, the study uncovers pronounced regime-dependent asymmetries. Through GMM-based threshold analysis, the BOK is found to exhibit “opportunistic” behavior in high-inflation episodes, increasing its response intensity to price gaps. Simultaneously, a “recession-avoidance” bias is evident during downturns, where the central bank adjusts interest rates more aggressively (long-run coefficient of 0.687) to mitigate real-side weakness compared to expansionary periods. Third, policy weights recovered from a Time-Varying Parameter (TVP) state-space model show a gradual rise in the relative emphasis on output stabilization, from 0.62 in the early 2000s to approximately 0.72 in recent years. Robustness checks using survey-based inflation expectations and weak-instrument robust inference confirm that this upward trend is a consistent feature of the data rather than a result of statistical noise. Fourth, a structural decomposition reveals that this shift in policy weights reflects an optimal response to a structural decline in the interest-rate sensitivity of output (), rather than a shift in the central bank’s intrinsic preferences. As the transmission mechanism has weakened due to demographic shifts and credit channel changes, the monetary authority has assigned a higher weight to the output gap to maintain macroeconomic stability. Overall, the results highlight the endogenous flexibility of Korean monetary policy amid an evolving economic structure. The identified decline in interest-rate sensitivity suggests that monetary policy alone may face limitations in steering the real economy. Consequently, the study underscores the necessity of a coherent policy mix, emphasizing closer coordination with fiscal policy and the refinement of macroprudential tools to manage structural risks without overburdening the interest rate path.

키워드

통화정책 반응함수손실함수 가중치시변모수(TVP) 모형역최적화정책 비대칭성Taylor RuleTime-Varying Parameter (TVP) ModelLoss Function WeightAsymmetryMonetary Policy Reaction Function
제목
한국 통화정책의 비대칭적 반응과 중앙은행 손실함수 가중치 추정
제목 (타언어)
Asymmetric Reaction of Monetary Policy and Estimation of Central Bank Loss Function Weights in Korea
저자
곽노선
DOI
10.21023/JMF.40.1.3
발행일
2026-03
유형
Y
저널명
금융연구
40
1
페이지
71 ~ 113