KOSPI200 추적 ETF의 추적오차

We analyze the tracking errors of ETFs that track KOSPI200 using daily data from October 15, 2002 to March 19, 2012. First, the over-performance of ETFs relative to KOSPI200 is statistically insignificant and economically negligible. Even though there is an ETF that shows over-performance ex post, we cannot anticipate the over-performance of the ETFs ex ante. However, the tracking errors of ETFs relative to KOSPI200 are shown to be statistically significant and economically meaningful. Also, the level of tracking errors among the ETFs differs significantly. The tracking errors of ETF relative to NAV turn out to be a major component in the tracking errors of ETFs relative to KOSPI200, and the tracking errors of NAV relative to KOSPI200 are negligible. Therefore, the selection of a liquid and efficient ETF is more important than the selection of an ETF managed by superior fund manager to get similar performance to KOSPI200. Lastly, ETFs tend to be under-priced relative to NAV, and the under-pricing persists for a long period of time. The erratic movements of ETF cause extreme discrepancy between ETF and NAV, which is reduced by the reversion of ETF on the next trading day. These results indicate that although fund managers track KOSPI200 effectively, inefficient ETF price-determination makes it difficult for ETF to track KOSPI200 effectively. The policy measures leading to efficient ETF price-determination are necessary.