Posts by Collection

Individual heterogeneity in the returns to schooling: instrumental variable quantile regression

Author: Hayun Song

2018

Abstract: The main focus of this paper is to investigate whether people with varying levels of unobserved ability obtain different earnings based on their years of schooling. This paper's contribution to the literature is to use the instrumental quantile regression method to capture the heterogeneity of returns on the twins' sample while controlling for ability and measurement error biases. After controlling all covariates and biases, the range of estimates is between 9 percent and 15 percent. Although there is a weak identification problem, the results from both the levels and the proxy models are statistically significant. This paper shows the existence of heterogeneity across individuals through the general Wald-type location shift test. This indicates the complementary relationship between education and schooling in the generation of earnings. Furthermore, I check the positive ability bias, negative measurement error, linearity of schooling, and the heterogeneity of returns of other covariates, including age, race, gender, union membership, and tenure.

Bayesian dynamic factor augmented structure learning: cross-sectional dependence for residuals

Author: Hayun Song

2021

Abstract: We propose the Bayesian approach to estimate the dynamic factor-augmented VAR model. As a result, we can obtain the contemporaneous connectedness as a graphical model of the cross-sectional dependence. In this paper, we estimate unobserved factors as a principal component given the known number of factors. Then, we draw factors through the Gibbs sampler using the forward-filtering backward-sampling algorithm. For the transition matrix, we use a rescaled version of the spike and slab priors for our coefficients of lagged variables, which solves the matrices’ collinearity (or possible rank deficiency) when the number of variables is high-dimensional. We check the properties of the estimators derived from the rescaled spike and slab prior by converting the original Bayesian problem into the Frequentists’ ridge estimation problem. We show that the posterior mean asymptotically maximizes the posterior distribution by analyzing the sensitivity of the choice of priors of coefficients. Lastly, we use the fractional Bayes factor to implement the Bayesian graphical model selection based on the graphical VAR. MC simulation shows the performance of our estimation strategy, and we consider weak cross-sectional dependencies in U.S. house prices.

Female labor force participation and gender role attitudes

Authors: Hayun Song and Junghyuk Lee

2023

[Manuscript in preparation]

High-dimensional Bayesian nonparanormal dynamic conditional model with multivariate volatility applications

Author: Hayun Song

2023

Abstract: This paper proposes a Bayesian approach for the estimation of large conditional precision matrices instead of inverting conditional covariance matrices estimated, using, for example, the dynamic conditional correlations (DCC) approach. By adopting a Wishart distribution and horseshoe priors within a DCC–GARCH(1,1) model, our method imposes sparsity and circumvents the inversion of conditional covariance matrices. We also employ a nonparanormal method with rank transformation to allow for conditional dependence without estimating transformation functions to achieve Gaussianity. Monte Carlo simulations show that our approach is effective at estimating the conditional precision matrix, particularly when the number of variables $(N)$ exceeds the number of observations $(T)$. We investigate the utility of our proposed approach with two real-world applications. First, to study conditional partial correlations among international stock price indices. Second, to test for $\boldsymbol{\alpha}$ in the context of CAPM and Fama-French 5 factor models with a conditional precision matrix-based Wald-type test. The results indicate stable conditional partial correlations through market disruptions. When there are market disruptions, blue chip stocks chosen from S&P 500 daily returns provide statistically significant evidence against the CAPM and Fama-French five models.

Small Differences, Large Divisions: How Political Gaps Widen Over Time

Author: Hayun Song

KIEP Opinions No. 319, July 10, 2025

Abstract: This essay explains how modest political differences can compound into deep polarization through a feedback loop involving human psychology, digital platforms, and electoral incentives. Drawing on bounded-confidence models, filter-bubble economics, and models of local cultural convergence, it shows how people increasingly interact only with like-minded views while engagement-driven algorithms reinforce those boundaries. A simulation illustrates how an initially moderate opinion distribution can split into opposing clusters even without deliberate manipulation. The analysis argues that polarization can emerge from ordinary incentives operating at scale and that restoring cross-cutting exposure is essential to preserving a functional democratic center.

Cross-Country Economic Spillover Effects of Intangible Asset: Technology Diffusion and Implications

Korean title: 무형자산 기술확산의 국가 간 경제적 파급효과와 시사점

Authors: Jung Eun Yoon, Hayun Song, and Byungjun Lee

KIEP Policy Analyses 25-01, November 18, 2025

Abstract: This study analyzes cross-border investment in artificial intelligence and the international spillover effects of intangible assets, with a focus on how technology-following economies such as Korea can respond to the concentration of global AI investment in the United States and China. Combining historical evidence, a theoretical model, and GVAR analysis, it finds that technology partnerships and sustained R&D are both important and that overseas AI investment can be closely associated with domestic output and market expectations. The report recommends supporting overseas investment and joint ventures while bringing acquired technology and talent back to Korea, concentrating resources in areas that combine Korea's manufacturing and IT strengths, and balancing international cooperation with continued domestic innovation.

China’s Medium- and Long-Term Growth Outlook and Changes in Its Growth Structure

Korean title: 중국경제 중장기 성장 전망과 성장구조 변화에 대한 연구

Authors: Jiyoung Moon, Hayun Song, Hong Won Kim, JiWon Choi, and Ko Un Cho

KIEP Basic Research Report 25-12, December 30, 2025

Abstract: This report examines the structural slowdown of China's economy from both supply- and demand-side perspectives and evaluates its medium- and long-term growth prospects using time-series analysis and a GVAR model. It documents weakening returns to capital, demographic pressures, slower productivity growth, and constraints on consumption-led expansion, while also considering policy efforts centered on technological innovation and new quality productive forces. The baseline projection points to a gradual decline in average growth over successive five-year periods, with a widening gap between optimistic and pessimistic scenarios depending on the success of structural reform and productivity gains. The report discusses implications for Korea, including intensified competition in strategic industries, new opportunities from China's structural transition, and the need to monitor macroeconomic and supply-chain risks.

Global Trends in High Debt Levels and Their Macroeconomic Implications

Korean title: 글로벌 고부채 동향 및 거시경제적 함의

Authors: Hongseok Choi, Jee Won Park, Hayun Song, ByeongJun Lee, and Kotbee Shin

KIEP Basic Research Report 25-11, December 30, 2025

Abstract: This report examines the post-pandemic high-debt environment across emerging markets, advanced economies, and Korea using threshold-based macroeconometric models. It finds that advanced-economy government debt ratios have risen persistently since 2008 and that the growth effects of debt depend on debt levels and economic regimes. For Korea, private-debt growth has a statistically significant negative effect on GDP when leverage is high, while household and corporate borrowing have different implications for growth and financial stress. The report calls for debt policy that distinguishes across sectors and business-cycle conditions, dynamic macroprudential monitoring, and stronger regional financial-safety arrangements.

Economic Spillovers of Major External Shocks: Trade Policy Uncertainty and Oil Prices

Korean title: 주요 대외충격의 경제적 파급효과: 무역정책 불확실성과 유가를 중심으로

Authors: Hayun Song, Sang-Ha Yoon, and Youngdoo Cho

KIEP Today's World Economy 26-03, January 14, 2026

Abstract: This report uses a GVAR model covering 33 economies from 1985Q1 to 2025Q2 to compare the effects of U.S. trade policy uncertainty and oil supply news shocks on Korea. A trade policy uncertainty shock first affects financial markets and then the real economy, producing statistically significant declines in Korean GDP and stock prices together with currency depreciation. By contrast, a 10% oil-price increase has a significant negative effect on stock prices but no statistically significant effect on Korean GDP or inflation, indicating substantial uncertainty around real-economy spillovers. The findings support differentiated policy responses: broad real and financial risk management for trade-policy shocks and flexible, monitoring-centered responses to oil shocks.

AI Usage in Major Economies and Korea

Korean title: 주요국의 AI 사용 실태와 한국의 현황

Authors: Hayun Song and ByeongJun Lee

KIEP World Economy Focus 26-04, January 29, 2026

Abstract: Using the Anthropic Economic Index based on Claude usage data, this report compares how artificial intelligence is used across major economies and examines Korea's position. Global usage remains concentrated in coding tasks, but the pattern is shifting from automation toward augmentation, while adoption across countries is strongly associated with income levels. Korea accounts for 3.06% of observed global usage and has the highest work-related usage share among the four East Asian economies examined, although its average degree of task delegation to AI remains below the global mean. The report argues that AI can raise productivity while producing different skill effects across tasks and recommends AI literacy programs, human-capital investment, and support for adoption by small and medium-sized enterprises.

Oil Price Shocks from the U.S.-Iran War: Spillover Effects on Major Economies

Korean title: 미-이란 전쟁에 따른 유가 충격의 주요국 파급효과

Authors: Hayun Song and Munsu Kang

KIEP Today's World Economy 26-09, April 1, 2026

Abstract: This report uses a GVAR model covering 33 economies to quantify how the U.S.-Iran war could affect oil prices and major economies under three scenarios: an early ceasefire, a prolonged conflict involving a blockade of the Strait of Hormuz, and escalation involving attacks on energy facilities. Conditional forecasts indicate that oil prices remain above their prewar level through 2027 even under an early end to hostilities, with larger and more persistent increases in the other scenarios. Because Korea depends heavily on Middle Eastern energy, the report highlights risks to import costs and the supply of naphtha and LNG. It calls for immediate supply diversification, contingency planning, and a longer-term strengthening of energy security.