기간스프레드는 서로 다른 만기의 이자율의 차이를 의미하며, 관련 연구들은 기간스프레드가 미래의 경제상황 및 주식수익률에 대한 예측력을 보유하고 있음을 보였다. 본 연구에서는 장기기간스프레드(10년-3개월)와 단기기간스프레드(3개월-1일)의 정보력을 비교하고, 정보력 차이의 원인에 대해서 분석한다. 이자율기간구조를 활용한 주성분분석(Principal Component Analysis)에서 도출한 주성분(Principal Component)과의 비교를 통해, 장기기간스프레드와 단기기간스프레드의 정보력이 이자율 기간구조의 어떤 요소와 관련되어 있는지 확인하였다. 연구 결과에 의하면, 장기기간스프레드에 비해 단기기간스프레드가 미래 주가지수수익률에 소폭 우수한 정보력을 보유하는 것으로 나타났으나, 단기기간스프레드 역시 제한된 설명력만을 보유하는 것을 확인하였다. 반면, 모든 만기의 이자율정보를 활용하는 주성분분석의 기울기주성분이 가장 우수한 수익률예측력을 가지는 것으로 나타났다. 한편, 평균적으로 높은 금리수준에서는 주성분분석의 곡률도주성분 역시 수익률에 대한 예측력을 보유하는 것으로 나타났다.

Term spreads are the difference in interest rates across different maturities of bonds and are known to have predictive power for future economic conditions and stock returns. This study compares the informational power of long-term spreads (10 years-3 months) and short-term spreads (3 months-1 day) and analyzes the sources of the difference in informational power. By comparing the principal components derived from a principal component analysis (PCA) using the term structure of interest rates, we identify which elements of the term structure are related to the informativeness of long-term and short-term spreads. The results showed that short-term spreads are somewhat more informative than long-term spreads, but also have limited explanatory power. However, the slope component (PC2) of the PCA using interest rate information from all maturities had significant explanatory power for future returns. On the other hand, the curvature component of the PCA (PC3) had significant explanatory power for future returns, on average, during periods of high interest rates. The authors found that the explanatory power of the curvature component (PC3), which together with the slope component holds information about future returns, varies significantly with the level of interest rates: during low interest rate periods, the predictive power of the curvature component is not significant, while as interest rates rise, the predictive power of the curvature component changes significantly. The contributions of our study is that we identify differences in the informativeness of term spreads by maturity and the sources of these differences. In particular, we find that the informativeness of short-term spreads, which have received little attention, is significantly higher than that of long-term spreads. This informational difference can be explained by the possibility that different maturities in the term structure of interest rates may contain independent information. This suggests that in order to extract richer information from the interest rate curve, it is necessary to simultaneously utilize various information from the entire term structure in addition to term spreads of specific maturities.