The Drivers of Inflation in Korea: Insights from a Small Open DSGE Model

Kim, Kyunghun Article The Drivers of Inflation in Korea: Insights from a Small Open DSGE Model East Asian Economic Review (EAER) Provided in Cooperation with: Korea Institute for International Economic Policy (KIEP), Sejong-si Suggested Citation: Kim, Kyunghun (2025) : The Drivers of Inflation in Korea: Insights from a Small Open DSGE Model, East Asian Economic Review (EAER), ISSN 2508-1667, Korea Institute for International Economic Policy (KIEP), Sejong-si, Vol. 29, Iss. 1, pp. 41-76, https://doi.org/10.11644/KIEP.EAER.2025.29.1.444 This Version is available at: https://hdl.handle.net/10419/316640 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ PISSN 2508-1640 EISSN 2508-1667 an open access journal East Asian Economic Review vol. 29, no. 1 (March 2025) 41-76 https://dx.doi.org/10.11644/KIEP.EAER.2025.29.1.444 ⓒ Korea Institute for International Economic Policy The Drivers of Inflation in Korea: Insights from a Small Open DSGE Model* Kyunghun Kim† Hongik University [email protected] 1 This study analyzes the drivers of inflation in South Korea using a Dynamic Stochastic General Equilibrium (DSGE) model tailored to the characteristics of Korea as a small open economy. Employing quarterly data from 1999Q2 to 2023Q2, Bayesian estimation is used to estimate Korea-specific parameters. Based on the estimated parameters, impulse response analysis and historical decomposition are conducted. The results indicate that cost-push shocks tied to imported goods pricing have been the primary driver of recent inflation surges in Korea. Accordingly, policymakers need to adopt a comprehensive approach—including not only monetary policy but also macroprudential measures and raw material supply management—to mitigate supply-side inflationary pressures effectively. Keywords: Inflation, Dynamic Stochastic General Equilibrium, Bayesian Estimation, Cost-push Shocks JEL Classification: E31, E32, C11, F41 I. Introduction The annual growth rates of Korea’s Core Consumer Price Index (CPI) exceeded 3.6% and 3.4% in 2022 and 2023, respectively. Although the index declined to 2.2% in 2024, these elevated inflation rates represent unprecedented levels in the past * This work was supported by the Ministry of Education of the Republic of Korea and the National Research Foundation of Korea (NRF-2022S1A5A8052515). I would like to thank three anonymous referees for their insightful comments. † Associate Professor, School of Economics, Hongik University, 94 Wausan-ro, Mapo-gu, Seoul 04066 Republic of Korea. Tel: +82-2-320-1809. ID 42 Kyunghun Kim ⓒ Korea Institute for International Economic Policy decade. 1 Historically, such high inflation rates were observed in the early 2000s (2001– 2003) and during the global financial crisis in 2008, when inflation exceeded 3% (Figure 1). Figure 1. Core Consumer Price Index(CPI) in Korea: Year 2000~2024 Note: Core Consumer Price Index(CPI) represents a measure of inflation that excludes volatile prices like food and energy from the CPI. Source: Bank of Korea Economic Statistics System This study identifies the major shocks that have influenced inflation dynamics in Korea since the 2000s. It also investigates the key drivers behind the recent surge in inflation, providing a deeper understanding of the factors contributing to the sharp price increases observed in recent years. Finally, the study explores policy implications for managing interest rate policy effectively to achieve price stability in the context of Korea’s unique economic characteristics. This study employs the Dynamic Stochastic General Equilibrium (DSGE) model by Justiniano and Preston (2010), which incorporates the characteristics of Korea as a small open economy. Justiniano and Preston (2010) build on the framework of Galí and Monacelli (2005), which assumes imperfect competition and price rigidities, by 1 Global supply chain bottlenecks and heightened energy price volatility have been identified as common factors driving global inflation (International Monetary Fund, 2021). Similarly, South Korea has experienced a significant rise in consumer prices, with inflation rates exceeding 3% since October 2021 (Oh et al., 2022). 0 0.5 1 1.5 2 2.5 3 3.5 4 0 20 40 60 80 100 120 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 Core CPI (left, index) Growth(right, %) The Drivers of Inflation in Korea: Insights from a Small Open DSGE Model 43 ⓒ 2025 East Asian Economic Review introducing additional features such as an incomplete asset market, habit formation, and price indexation to past inflation. For the analysis, the study utilizes calibration and Bayesian estimation to estimate Korea-specific parameters. Based on these estimated parameters, the study focuses on inflation, the primary variable of interest, to conduct an in-depth analysis. The variables used for parameter estimation include Korea’s industrial production, policy interest rate, real effective exchange rate, terms of trade, and core consumer price index (excluding food and energy), as well as the G-7 countries’ industrial production, core consumer price index (excluding food and energy), and average policy interest rate. Quarterly data were utilized, covering the period from 1999Q2 to 2023Q2, which corresponds to the post-IMF crisis era when Korea adopted inflation targeting and transitioned to a floating exchange rate regime. Based on the analysis of the estimated parameters, during periods of rising prices in the pre-COVID era—when overall inflation was relatively stable—both monetary policy shocks and technology shocks played a significant role in driving inflation in Korea. However, in the two years following 2021, when inflation accelerated sharply, technology shocks continued to contribute substantially, while cost-push shocks emerged as a major influence on price increases. This contrasts with the pre-COVID period, during which cost-push shocks largely helped stabilize inflation; more recently, however, they appear to have propelled inflation upward. These findings indicate that effectively managing supply-side pressures—such as raw material costs and domestic distribution margins—along with fostering productivity growth, is central to maintaining price stability. While many studies have utilized DSGE models to analyze Korea’s business cycles or optimal monetary policy (considering inflation targeting), few have focused specifically on inflation dynamics and the theoretical transmission channels, as proposed in this study. Analyzing inflation dynamics using a small open economy DSGE model offers the advantage of identifying not only domestic factors but also external factors, such as rising import prices, foreign interest rate shocks, or foreign output fluctuations, that contribute to inflation. This approach is particularly significant for understanding inflation in a highly trade-dependent, small open economy like Korea and represents a valuable contribution to the existing literature. The remainder of this paper is organized as follows: Section II reviews the existing literature related to this study. Section III describes the model, while Section IV presents the Bayesian estimation results. Section V discusses the findings of the analysis, and Section VI concludes the paper. 44 Kyunghun Kim ⓒ Korea Institute for International Economic Policy II. Literature Review There is extensive existing research on exchange rate pass-through in Korea. Cha (2007) analyzed the factors driving the increase in Korea's exchange rate pass-through and its impact on domestic inflation. Covering the period from 1980 to 2005, the study examined exchange rate pass-through by different exchange rate regimes. The findings indicated that both shortand long-term exchange rate pass-through to import prices have been increasing, not due to changes in trade composition but rather due to rising pass-through rates at the industry level. Zhu et al. (2010) used a structural VAR model (including variables such as oil prices, production, exchange rates, money supply, import prices, and consumer prices) to analyze the pass-through effects of won/dollar exchange rate changes. According to their study, the pass-through effects of exchange rate changes on import prices and consumer prices were very small and showed a declining trend over time. They argued that policymakers should prioritize managing inflation over focusing on exchange rates. Choi (2013) estimated the exchange rate pass-through for Korea’s manufacturing sector in response to changes in the won/dollar exchange rate. Using annual data from 1999 to 2011, a panel analysis was conducted on 30 manufacturing industries. The results showed that a 10% increase in the won/dollar exchange rate led to a 2.4% to 4.9% decrease in dollar-denominated export prices across all industries. The pass-through effect was particularly high in basic materials industries, while industries related to consumer goods—such as food, textiles and apparel, leather products, and miscellaneous manufactured goods—showed minimal responsiveness. Park and Cha (2014) analyzed the relationship between exchange rate pass-through to consumer prices and monetary policy using panel data from 11 emerging economies that had adopted inflation targeting policies. Their findings indicated that both the level and volatility of inflation decreased following the adoption of inflation targeting. They argued that this reduction in exchange rate pass-through was attributable to monetary authorities effectively managing inflation expectations under the inflation targeting framework. Lee (2017) estimated the exchange rate pass-through to domestic price levels using a time-varying parameter model with quarterly data. The analysis revealed a general declining trend in exchange rate pass-through since the 2000s. The study found that exchange rate volatility and inflation positively affected exchange rate pass-through, while trade dependency, inflation volatility, and production volatility had negative effects. The Drivers of Inflation in Korea: Insights from a Small Open DSGE Model 45 ⓒ 2025 East Asian Economic Review Chiang (2021) analyzed the asymmetry and nonlinearity of exchange rate pass-through to import prices across Korean industries. Using the pricing-to-exchange rate model proposed by Goldberg and Knetter (1997), the analysis demonstrated that the pass-through effect was greater during currency appreciations than depreciations, showing significant asymmetry. This effect became more pronounced after the foreign exchange crisis. Recent studies on DSGE models include Choi (2019), Rhee et al. (2020), Kim et al. (2021), and Hur and Oh (2024). Choi (2019) developed a small open economy DSGE model to analyze Korea’s business cycles and optimal monetary policy. Using Bayesian methods to estimate parameters, the study demonstrated that a positive technology shock leads to an increase in output, a decline in prices, and an appreciation of the exchange rate. In contrast, a positive foreign income shock causes increases in both production and prices. The analysis concluded that the optimal monetary policy rule is one that primarily responds to domestic inflation. Rhee et al. (2020) developed a small open economy DSGE model to analyze Korea’s business cycles and assess the forecasting performance of macroeconomic variables. Similar to Choi (2019), they estimated the model parameters using Bayesian methods and employed a DSGE-VAR approach. The analysis revealed that, compared to monetary or supply shocks, demand and technology shocks have a more significant impact on key economic variables. This effect was particularly pronounced for growth rates and inflation. Kim et al. (2021) analyzed the impact of the Federal Reserve’s monetary policy normalization on Korea’s external sector, including exchange rates, capital flows, and swap basis spreads. They conducted scenario analyses focusing on the effects of the Federal Reserve’s tapering and interest rate hikes. Notably, their study shares similarities with the current study in utilizing the model by Justiniano and Preston (2010) to assess the effects of monetary policy normalization on Korea’s exchange rate. The analysis showed that under a pessimistic scenario, the exchange rate could increase by up to 22%, and GDP could initially decline by 2.1% before recovering. Hur and Oh (2024) employ a DSGE model to examine how the Bank of Korea has conducted monetary policy since the 2000s, showing that it has partially responded to financial stability factors, such as household credit, as well as inflation and growth. Focusing on the perspective of optimizing a central bank’s loss function, they demonstrate that incorporating household debt variables into the interest rate rule leads to more effective mediumto long-term economic stabilization. Consequently, their 46 Kyunghun Kim ⓒ Korea Institute for International Economic Policy findings suggest that an Integrated Inflation Targeting framework could serve as a potential alternative for Korea’s monetary policy system. Previous studies closely related to this research include Choi (2019), Rhee et al. (2020), Kim et al. (2021), and Hur and Oh (2024). Like these studies, the current research is based on a DSGE model but focuses on identifying the theoretical mechanisms through which internal and external factors influence inflation dynamics. Notably, it distinguishes itself from prior research by analyzing the causes of Korea’s recent inflation surge following the COVID-19 pandemic using the DSGE framework. III. Model In Chapter III, the DSGE model is briefly described. The model is based on Justiniano and Preston (2010), which builds upon Galí and Monacelli (2005) by introducing assumptions of imperfect competition and price rigidities, along with additional features such as an incomplete asset market, habit formation, and price indexation to past inflation. For a concise explanation of the model, the economy is divided into three agents: households, producers, and retailers (who set prices). The optimal decision-making process for each agent is derived, and the equilibrium, where all markets clear, is defined. 1. Households Households are assumed to make decisions that maximize their lifetime expected utility. 𝐸∑𝛽   𝜀,󰇛󰇜  −  (1) Household consumption 𝐶 represents the consumption of final goods, influenced by habit formation ℎ𝐶, which reflects inertia in past consumption behavior. Both 𝐶 and ℎ𝐶 affect household utility. 𝑁 denotes labor supply, 𝛽 is the discount factor for utility, and 𝜎 is the coefficient of relative risk aversion, whose reciprocal represents the intertemporal elasticity of substitution. The reciprocal of 𝜑 corresponds to the Frisch elasticity of labor supply. 𝜀, represents the preference shock for households. The Drivers of Inflation in Korea: Insights from a Small Open DSGE Model 47 ⓒ 2025 East Asian Economic Review Consumption 𝐶 is defined as a final good, which is a Dixit–Stiglitz aggregator of domestically produced goods 𝐶, and goods produced abroad 𝐶,. The parameter 𝛼 indicates the degree of trade openness, 𝜂 represents the elasticity of substitution between 𝐶, and 𝐶,, and 𝜀 denotes the elasticity of substitution among the individual goods 𝑖 that compose 𝐶, and 𝐶,. 𝐶=󰇩󰇛1−𝛼󰇜 𝐶,  +𝛼 𝐶,  󰇪  (2) where 𝐶,=󰇣𝐶,(𝑖)   𝑑𝑖󰇤 , 𝐶,=󰇣𝐶,(𝑖)   𝑑𝑖󰇤  The household’s budget constraint is given by Equation (3). 𝑃𝐶+𝐷+𝑒𝐵=𝐷𝑖+𝑒𝐵𝑖 ∗𝜙(𝐴)+𝑊𝑁+𝛱,+𝛱,+𝑇 (3) 𝐷 and 𝐵 represent households' holdings of domestic and foreign bonds for one period, respectively, with the corresponding interest rates denoted by 𝑖 and 𝑖∗. 𝑒 represents the nominal exchange rate, while 𝑃, 𝑃,, 𝑃,, and 𝑃∗ denote, respectively, the overall domestic price level (or consumer price index), the price of domestically produced goods, the domestic currency-denominated price of imported foreign goods, and the foreign currency-denominated price of foreign goods. 𝑊 denotes wages, and 𝑇 represents lump-sum taxes and transfers. Π,and Π, are the profits of domestic goods producers and importers of foreign goods, respectively. It is assumed that all households receive equal shares of profits. Therefore, the nominal income of each household is 𝑊𝑁+Π,+Π,, which equals 𝑃,𝑌,+𝑃,−𝑒𝑃∗𝐶,. When issuing foreign bonds to borrow from abroad, a risk premium 𝜙(∙) is added to the foreign nominal interest rate.2 𝐴 represents the ratio of foreign bonds (denominated in domestic currency) to output in steady-state, and 𝜙 denotes the risk premium shock. 2 Refer to Kollmann (2002), Schmitt-Grohé and Uribe (2003), and Benigno (2009). 48 Kyunghun Kim ⓒ Korea Institute for International Economic Policy 𝜙=𝑒𝑥𝑝−𝜉𝐴+𝜙 (4) where 𝐴≡𝑒𝐵 𝑌 𝑃 The optimal expenditure allocation condition for domestic goods and foreign goods 𝑖, which maximizes household utility, is as follows. 𝐶,(𝑖)=,() ,𝐶,,𝑎𝑛𝑑 𝐶,(𝑖)=,() ,𝐶, (5) Furthermore, the optimal expenditure allocation condition for the composite of domestic and foreign goods that constitutes the final consumption good is as follows. 𝐶,=(1−𝛼)󰇡, 󰇢𝐶,𝑎𝑛𝑑 𝐶,=𝛼󰇡, 󰇢𝐶 (6) The overall domestic price level (or consumer price index) is defined as: 𝑃= (1−𝛼)𝑃, +𝛼𝑃,   , where 𝑃,and 𝑃, represent the price indices of domestic goods (composite) and foreign goods (composite), respectively. The optimal allocation of expenditure on final consumption goods and labor supply satisfies Equations (7) and (8), respectively. 𝜆 represents the Lagrangian multiplier, which reflects the marginal utility of nominal income. 𝜆=𝜀,(𝐶−ℎ𝐶) (7) 𝜆=𝜀,  (8) The allocation of domestic and foreign bonds satisfies the following optimality conditions. Figure 2. Time Series Data Used for Parameter Estimation a. Industrial p roduction b . Polic y interest rate c. Real effective exchange rate d. Terms of trade -20 -15 -10 -5 0 5 10 15 0 20 40 60 80 100 120 140 1999-05 2000-08 2001-11 2003-02 2004-05 2005-08 2006-11 2008-02 2009-05 2010-08 2011-11 2013-02 2014-05 2015-08 2016-11 2018-02 2019-05 2020-08 2021-11 2023-02 0 0.01 0.02 0.03 0.04 0.05 0.06 0 1 2 3 4 5 6 1999-05 2000-08 2001-11 2003-02 2004-05 2005-08 2006-11 2008-02 2009-05 2010-08 2011-11 2013-02 2014-05 2015-08 2016-11 2018-02 2019-05 2020-08 2021-11 2023-02 0 20 40 60 80 100 120 140 0 20 40 60 80 100 120 140 1999-05 2000-09 2002-01 2003-05 2004-09 2006-01 2007-05 2008-09 2010-01 2011-05 2012-09 2014-01 2015-05 2016-09 2018-01 2019-05 2020-09 2022-01 2023-05 -20 -15 -10 -5 0 5 10 15 20 0 50 100 150 200 250 1999-05 2000-09 2002-01 2003-05 2004-09 2006-01 2007-05 2008-09 2010-01 2011-05 2012-09 2014-01 2015-05 2016-09 2018-01 2019-05 2020-09 2022-01 2023-05 The Drivers of Inflation in Korea: Insights from a Small Open DSGE Model 55 ⓒ 2025 East Asian Economic Review Figure 2. Continued e. Consumer price index f. G-7 Industrial p roduction g . G-7 Consumer p rice index h. G-7 Polic y interest rate Note: The dashed lines represent the raw data(left y-axis), while the solid lines represent the preprocessed data(right y-axis). Consumer price indices for Korea and the G-7 exclude food and energy. Source: Author’s calculation based on data from the OECD, BIS, and the Bank of Korea Economic Statistics System. -3 -2 -1 0 1 2 3 4 0 20 40 60 80 100 120 140 1999-05 2000-09 2002-01 2003-05 2004-09 2006-01 2007-05 2008-09 2010-01 2011-05 2012-09 2014-01 2015-05 2016-09 2018-01 2019-05 2020-09 2022-01 2023-05 -20 -15 -10 -5 0 5 10 0 20 40 60 80 100 120 1999-05 2000-09 2002-01 2003-05 2004-09 2006-01 2007-05 2008-09 2010-01 2011-05 2012-09 2014-01 2015-05 2016-09 2018-01 2019-05 2020-09 2022-01 2023-05 -3 -2 -1 0 1 2 3 4 0 20 40 60 80 100 120 140 1999-05 2000-09 2002-01 2003-05 2004-09 2006-01 2007-05 2008-09 2010-01 2011-05 2012-09 2014-01 2015-05 2016-09 2018-01 2019-05 2020-09 2022-01 2023-05 0 0.01 0.02 0.03 0.04 0.05 0.06 0 1 2 3 4 5 6 1999-05 2000-08 2001-11 2003-02 2004-05 2005-08 2006-11 2008-02 2009-05 2010-08 2011-11 2013-02 2014-05 2015-08 2016-11 2018-02 2019-05 2020-08 2021-11 2023-02 56 Kyunghun Kim ⓒ Korea Institute for International Economic Policy Table 1. Prior and Posterior Distributions of Parameters Parameters and Descriptions Prior Distributions Posterior Distributions Distribution Mean S.D. Mean S.D. 10% 90% 𝜎 Inverse of Intertemporal Elasticity of Substitution Gamma 1.2 0.4 1.78 0.25 1.41 2.20 𝜑 Inverse of Frisch Elasticity of Labor Suppl y Gamma 1.5 0.75 1.66 0.41 1.06 2.28 𝜃 Domestic Calvo Paramete r Beta 0.5 0.1 0.82 0.04 0.76 0.89 𝜃 Foreign Calvo Paramete r Beta 0.5 0.1 0.95 0.001 0.95 0.95 𝜂 Elasticity of Substitution Between Domestic and Foreign Goods Gamma 1.5 0.75 0.83 0.01 0.82 0.83 ℎ habit formation Beta 0.5 0.25 0.78 0.04 0.72 0.84 𝛿 Indexation of Domestic Goods Price Inflation Beta 0.5 0.25 0.60 0.20 0.19 0.86 𝛿 Indexation of Foreign Goods Price Inflation Beta 0.5 0.25 0.63 0.07 0.52 0.73 𝜓 Taylor Rule Smoothing Paramete r Beta 0.5 0.25 0.98 0.01 0.97 1.00 𝜓 Taylor Rule Inflation Responsiveness Gamma 1.5 0.3 0.30 0.01 0.29 0.30 𝜓  Taylor Rule Output Responsiveness Gamma 0.25 0.13 0.02 0.01 0.01 0.03 𝜓  Taylor Rule Output Growth Responsiveness Gamma 0.25 0.13 0.03 0.01 0.01 0.04 𝜓 Taylor Rule Exchange Rate Growth Responsiveness Gamma 0.25 0.13 1.54 0.08 1.40 1.65 𝜌 Technology Shock AR(1) Beta 0.8 0.1 0.55 0.07 0.47 0.65 𝜌  Preference Shock AR(1) Beta 0.8 0.1 0.58 0.05 0.50 0.67 𝜌 Risk Premium Shock AR(1) Beta 0.8 0.1 0.96 0.02 0.94 0.99 𝜌 Cost-Push Shock AR(1) Beta 0.5 0.25 0.04 0.28 0.04 0.78 𝑠𝑑∗ Foreign Inflation Shoc k S.D. Inv Gamma 0.5 10 0.47 0.03 0.41 0.52 𝑠𝑑  ∗ Foreign Output Shock S.D. Inv Gamma 0.5 10 2.18 0.15 1.92 2.42 𝑠𝑑∗ Foreign Interest Rate Shock S.D. Inv Gamma 0.5 10 0.06 0.002 0.06 0.06 𝑠𝑑 Technology Shock S.D. Inv Gamma 0.5 10 12.57 4.20 7.14 19.50 𝑠𝑑 Monetary Policy Shock S.D. Inv Gamma 0.5 10 0.27 0.02 0.24 0.31 𝑠𝑑  Preference Shock S.D. Inv Gamma 0.5 10 29.00 3.27 24.30 34.96 𝑠𝑑 Risk Premium Shock S.D. Inv Gamma 0.5 10 8.21 0.65 7.14 9.27 𝑠𝑑 Cost-Push Shock S.D. Inv Gamma 0.5 10 3.79 0.30 3.29 4.27 Source: Author’s calculation based on Justiniano and Preston (2010). S.D. denotes standard deviation. The Drivers of Inflation in Korea: Insights from a Small Open DSGE Model 57 ⓒ 2025 East Asian Economic Review 58 Kyunghun Kim ⓒ Korea Institute for International Economic Policy V. Results In Chapter V, impulse response analysis and historical decomposition are conducted based on the estimated parameters. The impulse response analysis provides insights into the effects of each shock on Korea’s consumer price index, while the historical decomposition identifies the factors contributing to inflation dynamics in Korea since 2000. Additionally, the analysis reveals the primary drivers of recent inflation increases and derives relevant policy implications. 1. Impulse Response Analysis The first set of findings comes from the impulse response functions. In Figure 3 below, we present these functions, calibrated using our estimated parameters, to show how inflation responds to each of the model’s eight shocks. Each panel illustrates the effect on consumer prices (CPI) of a (+) 1 standard deviation shock. First, the technology shock lowers inflation by nearly 0.7% in the current period, then starts to rebound after about three quarters, eventually fading out. In contrast, the preference shock stimulates demand and raises inflation by around 0.15% during the first one or two quarters, but after that, inflation falls below the baseline and later returns to zero in the medium to long run. For the monetary policy (interest rate hike) and risk premium shocks, the domestic currency appreciates (i.e., the exchange rate declines), consistent with UIP. Due to this currency appreciation, inflation shows a short-term decline—around 0.12% for monetary policy shocks and about 0.06% for risk premium shocks, occurring either in the current quarter or soon afterward. The cost-push shock is directly linked to higher prices and exerts a sizable quantitative impact: it raises inflation by more than 0.9% immediately following the shock. Under a one-standard-deviation shock of the same size, the cost-push shock exerts the largest influence on inflation, followed by the technology shock. Finally, foreign output, inflation, and interest rate shocks have a smaller impact relative to domestic shocks, but their effects appear right away and continue to influence inflation over a relatively extended period compared to the other shocks.8 8 While the impact of the foreign output shock on inflation is relatively larger than that of foreign interest rate or foreign inflation shocks, it remains relatively modest compared to the non-foreign shocks. For additional responses of other variables to each shock, please refer to Appendix Figures 2–9. The Drivers of Inflation in Korea: Insights from a Small Open DSGE Model 59 ⓒ 2025 East Asian Economic Review Figure 3. Impulse Response Functions of Inflation to Eight Shocks a. Technology Shoc k b . Monetary Policy Shoc k c. Preference Shoc k d. Risk Premium Shoc k e. Cost-Push Shoc k f. Foreign Output Shock g . Forei g n Inflation Shoc k h. Forei g n Interest Rate Shoc k Source: Autho r ’s calculation. 60 Kyunghun Kim ⓒ Korea Institute for International Economic Policy 2. Historical Decomposition Analysis The second main result comes from the historical decomposition shown in Figure 4(a). This analysis reveals the extent to which each shock—technology (epsa), monetary policy (epsm), preference (epsg), risk premium (epsp), cost-push (epsc), foreign output (epys), foreign inflation (epps), and foreign interest rate (eprs)—has contributed to Korean inflation fluctuations since 2000. According to our historical decomposition, apart from the most recent five years, Korea’s inflation had been relatively stable. During periods of rising prices (beginning around the 40th quarter of the sample, i.e., following the global financial crisis), monetary policy shocks (epsm) and technology shocks (epsa) played a major role in driving inflation higher. However, in the last two years—roughly from 2022 through the final six quarters of the sample—technology shocks (epsa) continued to have a significant impact on inflation, while cost-push shocks (epsc) emerged as a central driver of rising prices. Notably, cost-push shocks had previously helped stabilize inflation, but they now contribute in the opposite direction, representing a new finding. Figures 4(b) and 4(c) illustrate historical decompositions for domestic goods inflation and imported goods inflation, respectively. As shown in (c), cost-push shocks have been fueling the recent surge in import prices. Consequently, the inflation upturn does not stem solely from domestic producer prices but also from higher inflation in imported items. Through cost-push shocks, import price increases raise domestic inflation overall. Looking at the variance decomposition confirms that cost-push shocks (epsc) account for about 63% of inflation (pie) fluctuations, followed by technology shocks (epsa) at around 30%. By contrast, monetary policy (epsm) and preference (epsg) shocks contribute only about 3%, while foreign inflation (epps) and foreign interest rate (eprs) shocks explain almost 0%. This suggests that, rather than directly capturing external pressures through shocks like foreign inflation (epps), the rise in external costs is strongly passed through to Korean prices via retailers’ pricing decisions (the cost-push channel). In other words, domestic inflation is heavily influenced by supply-side factors (e.g., import raw materials and distribution margins) and productivity shifts, whereas the model’s separate foreign The Drivers of Inflation in Korea: Insights from a Small Open DSGE Model 61 ⓒ 2025 East Asian Economic Review inflation and interest rate shocks appear to exert a relatively small direct effect.9 Since 2022, cost-push shocks have come to the forefront in driving inflation higher, due to soaring costs of commodities and supply chains under COVID-19, the Russia–Ukraine war, and heightened global economic uncertainty. In particular, geopolitical tensions and pandemic-related bottlenecks have pushed up energy and raw material prices, placing strong upward pressure on domestic inflation. Thus, what had been relatively moderate cost-side pressures pre-COVID can become amplified amid significant global instability, making cost-push shocks a key factor in recent changes to inflation. Bringing together these historical decomposition findings, the policy implication is that monetary policymakers should closely monitor and respond to external supply costs (e.g., import prices), domestic wage and distribution costs, and technological factors. Previously, cost-push shocks contributed to stable inflation or had a limited effect under relatively moderate price pressures; however, they have recently shifted to driving price increases through surging import prices and other external supply-side fluctuations. Therefore, it is crucial to manage and predict global raw material and energy prices as well as domestic distribution costs in a more proactive way. Furthermore, as the contribution of technology shocks (epsa) expands, there is a need to support long-term sustainable growth in the real sector—such as by improving productivity—and also to minimize the inflationary impact of unexpected technological changes through policy measures (for example, linking R&D investment to restructuring policies). Finally, examining the results of the historical decomposition reveals that monetary policy has exerted a relatively larger influence during periods of rising prices than when prices were declining. In light of this, relying solely on policy rate adjustments may not be sufficient during inflationary phases. Therefore, to prevent supply-side pressures and surging import prices from spilling over into broader inflation, it is advisable to employ a range of tools in tandem, including macroprudential and exchange-rate policies. 9 Although the overall quantitative contribution of monetary policy shocks to inflation appears relatively small over the entire period, their effect during episodes of rising inflation seems significant. In other words, monetary policy tends to have an asymmetrical impact on inflation, exerting a stronger influence when prices are on the upswing compared to when they are declining. Figure 4. Historical Decomposition of Korea’s Inflation (a) Inflation 62 Kyunghun Kim ⓒ Korea Institute for International Economic Policy Figure 4. Continued (b) Home goods price inflation The Drivers of Inflation in Korea: Insights from a Small Open DSGE Model 63 ⓒ 2025 East Asian Economic Review Figure 4. Continued (c) Imported goods price inflation Note: epsa = technology shock, epsm = monetary policy shock, epsg = preference shock, epsp = risk premium shock, epsc = cost-push shock, epys = foreign output shock, epps = foreign inflation shock, eprs = foreign interest rate shock. The solid line represents Korea’s consumer price index with trends and seasonality removed. Source: Author's calculation. 64 Kyunghun Kim ⓒ Korea Institute for International Economic Policy The Drivers of Inflation in Korea: Insights from a Small Open DSGE Model 71 ⓒ 2025 East Asian Economic Review Appendix Figure 4. Impulse Response Functions to Preference Shocks Note: c = consumption, y = GDP, r = policy interest rate, s = terms of trade, pie = CPI inflation rate, lop = law of one price gap, a = net foreign asset position, pif = imported goods price, er = exchange rate. Source: Author’s calculation. 72 Kyunghun Kim ⓒ Korea Institute for International Economic Policy Appendix Figure 5. Impulse Response Functions to Risk Premium Shocks Note: c = consumption, y = GDP, r = policy interest rate, s = terms of trade, pie = CPI inflation rate, lop = law of one price gap, a = net foreign asset position, pif = imported goods price, er = exchange rate. Source: Author’s calculation. The Drivers of Inflation in Korea: Insights from a Small Open DSGE Model 73 ⓒ 2025 East Asian Economic Review Appendix Figure 6. Impulse Response Functions to Cost-push Shocks Note: c = consumption, y = GDP, r = policy interest rate, s = terms of trade, pie = CPI inflation rate, lop = law of one price gap, a = net foreign asset position, pif = imported goods price, er = exchange rate. Source: Author’s calculation. 74 Kyunghun Kim ⓒ Korea Institute for International Economic Policy Appendix Figure 7. Impulse Response Functions to Foreign Output Shocks Note: c = consumption, y = GDP, r = policy interest rate, s = terms of trade, pie = CPI inflation rate, lop = law of one price gap, a = net foreign asset position, pif = imported goods price, er = exchange rate. Source: Author’s calculation. The Drivers of Inflation in Korea: Insights from a Small Open DSGE Model 75 ⓒ 2025 East Asian Economic Review Appendix Figure 8. Impulse Response Functions to Foreign Inflation Shocks Note: c = consumption, y = GDP, r = policy interest rate, s = terms of trade, pie = CPI inflation rate, lop = law of one price gap, a = net foreign asset position, pif = imported goods price, er = exchange rate. Source: Author’s calculation. 76 Kyunghun Kim ⓒ Korea Institute for International Economic Policy Appendix Figure 9. Impulse Response Functions to Foreign Interest Rate Shocks Note: c = consumption, y = GDP, r = policy interest rate, s = terms of trade, pie = CPI inflation rate, lop = law of one price gap, a = net foreign asset position, pif = imported goods price, er = exchange rate. Source: Author’s calculation.