주가지수 선물과 옵션의 만기일이 주식시장에 미치는 영향: 개별 종목 분석을 중심으로

Expiration-Day Effects: The Korean Evidence

본 논문은 주가지수 선물과 옵션의 만기일이 주식시장에 어떠한 영향을 미치는가에 대한 분석을 통해 한국주식시장에서 만기일 효과가 존재하는지를 검증한다. 주가지수를 이용한 기존의 논문과는 달리 만기일에 현물 주식시장의 움직임을 개별 종목별로 분석했다는 점에서 본 논문은 차별성을 지닌다. 주가지수는 시장 움직임의 평균으로 개별 종목의 고유한 특성을 반영하지 못하기 때문에 주가지수를 이용한 분석은 만기일 효과를 해석하고 그 원인을 분석하는데 한계를 지니고 있다. 분석 결과 한국주식시장에서 선물 만기일 효과는 분명히 존재하지만, 옵션 만기일 효과는 뚜렷하지 않은 것으로 드러났다. 선물 만기일에 KOSPI 200 지수와 개별 종목은 가격상승 압력이 존재하고, 변동성과 거래량이 증가하며, 만기일 다음날 수익률은 반전하는 경향이 있는 것으로 나타났다. 그러나 비교표본인 NON-KOSPI 200 지수와 개별 종목에서 만기일 효과가 존재한다고 할 만한 증거를 찾지 못했다. 만기일 효과가 시장 전체적으로 나타나는 것처럼 보이지만 KOSPI 200에 속하는 대규모 기업에 한정되며, 장 후반에 집중적으로 나타난다는 사실은 프로그램 매매와 만기일의 결제제도가 만기일 효과의 간접적 원인임을 시사한다. 또한 만기일 다음날 가격이 반전하는 현상이 KOSPI 200에 속하는 대규모 기업에 한정되어 나타나는 사실은 만기일 효과가 새로운 정보의 반영에 의한 정보 효과(information effects)가 아니라 일시적 거래불균형에 의한 유동성 효과(liquidity effects)임을 보여주는 증거가 된다.

This study examines the expiration-day effects of stock index futures and options in the Korean stock market. The so-called 'expiration-day effects', which are the abnormal stock price movements on derivatives expiration days, arise mainly from cash settlement. Index arbitragers have to bear the risk of their positions unless they liquidate their index stocks on the expiration day. If many arbitragers execute large buy or sell orders on the expiration day, abnormal trading volumes are likely to be observed. If a lot of arbitragers unwind positions in the same direction, temporary trading imbalances induce abnormal stock market volatility. By contrast, if some information arrives at market, the abnormal trading activity must be considered a normal process of price discovery. Stoll and Whaley(1987) investigated the aggregate price and volume effects of the S&P 500 index on the expiration day. In a related study, Stoll and Whaley(1990) found a similarity between the price behavior of stocks that are subject to program trading and of the stocks that are not. Thus far, there have been few studies about the expiration-day effects in the Korean stock market. While previous Korean studies use the KOSPI 200 index data, we analyze the price and trading volume behavior of individual stocks as well as the index. Analyzing individual stocks is important for two reasons. First, stock index is a market average. Consequently, it cannot reflect the behavior of many individual stocks. For example, if the expiration-day effects are mainly related to a specific group, it cannot be said that the expiration of derivatives itself destabilizes the stock market. Analyzing individual stocks enables us to investigate the scope of the expiration-day effects. Second, we can find the relationship between the firm characteristics and the expiration-day effects. For example, if the expiration-day effects exist in large stocks not belonging to the KOSPI 200 index, program trading may not be related to the expiration-day effects. The examination of individual stocks has led us to the cause of the expiration-day effects. Using the intraday data during the period May 3, 1996 through December 30, 2003, we first examine the price and volume effects of the KOSPI 200 and NON-KOSPI 200 index following the Stoll and Whaley(1987) methodology. We calculate the NON-KOSPI 200 index by using the returns and market capitalization of the KOSPI and KOSPI 200 index. In individual stocks, we divide KOSPI 200 stocks by size into three groups and match NON-KOSPI 200 stocks with KOSPI 200 stocks having the closest firm characteristics. We compare KOSPI 200 stocks with NON-KOSPI 200 stocks. To test whether the expiration-day effects are related to order imbalances or new information, we check price reversals on the next day. Finally, we perform a cross-sectional regression analysis to elaborate on the impact of the firm characteristics on price reversals. The main results seem to support the expiration-day effects, especially on stock index futures expiration days. The price behavior of stocks that are subject to program trading is shown to have price effects, abnormal return volatility, and large volumes during the last half hour of trading on the expiration day. Return reversals are also found in the KOSPI 200 index and stocks. However, there is no evidence of abnormal trading volume, or price reversals in the NON-KOSPI 200 index and stocks. The expiration-day effects are proportional to the size of stocks and the nearness to the settlement time. Since program trading is often said to be concentrated in high capitalization stocks, these results imply that the expiration-day effects seem to be associated with program trading and the settlement price determination procedure. In summary, the expiration-day effects in the Korean stock market do not exist in all stocks, but in large capitalization stocks belonging to the KOSPI 200 index. Additionally, the expiration-day effects in the Korean stock market are generally due, not to information, but to trading imbalances.