上櫃公司轉上市之事件研究--隨機係數模型的應用

1

行政院國家科學委員會補助專題研究計畫成果報告

※※※※※※※※※※※※※※※※※※※※※※※※

※ 上櫃公司轉上市之事件研究－隨機係數模型的應用 ※

※ Event Study of Exchange Listing: An Application of ※

※ Stochastic Coefficient Regression Model ※

※※※※※※※※※※※※※※※※※※※※※※※

計畫類別： ˇ個別型計畫

□整合型計畫

計畫編號：NSC 90－2416－H－004－024

執行期間：九十年八月一日至九十一年十月三十一日

計畫主持人：郭維裕

共同主持人：

本成果報告包括以下應繳交之附件：

□赴國外出差或研習心得報告一份

□赴大陸地區出差或研習心得報告一份

□出席國際學術會議心得報告及發表之論文各一份

□國際合作研究計畫國外研究報告書一份

執行單位：國立政治大學國際貿易學系

Event Study of Exchange Listing: An Application of Stochastic

Coefficient Regr ession Model

中文摘要

本研究利用事件研究法探討國內上櫃公司轉為上市公司交易後，其異常報酬與流動性的 變化情形。我們利用 Fama, Fisher, Jensen, and Roll (1969)的事件研究法計算異常報酬。 此外，我們透過文獻上常用的數個流動性指標來衡量各公司在轉上市後流動性的變化情 形。這些流動性指標包括 Elyasiani, Hauser, and Lauterbach (2000)的買賣價差替代指標、 Amivest 流動性指標、市場模型殘差變異數以及 Tkac（1999）所提出的異常流動性指標。 值得一提的是，本文的主要貢獻之一就是利用 Tkac（1999）的異常流動性指標來評估上 櫃公司轉上市事件的流動性變化程度，這是該指標第一次被應用於此領域。根據本研究 結果發現，平均而言，上櫃公司在轉上市前後的異常報酬率皆為負值，此結果與文獻的 發現有些不同。一般而言，文獻發現上櫃公司在正式宣告申請轉上市後，股價會呈現正 面的反應而出現正的異常報酬，但於正式上市交易後，股價便呈現相對於市場疲軟的走 勢而導致負的異常報酬。因此，總合而言，現有的文獻不認為上櫃公司可以透過轉上市 交易的方法來增加其公司的整體市場價值。雖然和文獻一致，本文發現上櫃公司的轉上 市後股票的異常報酬率是負的，但是轉上市前的結果卻與文獻的發縣大相逕庭。這個結 果顯示：國內的投資人並不認為上櫃公司可透過轉上市交易的方法獲得任何好處。至於 流動性的變化方面，雖然不同的指標呈現出些微不同的結果，但整體而言，上櫃公司在 轉上市後股票的流動性並未獲得明顯的改善，此結果亦和文獻的發現相異。因此，我們 認為上市市場並未提供足夠的誘因促使上櫃公司申請改變其交易場所。 關鍵詞：上櫃轉上市事件研究、異常報酬、異常流動性

3

Abstr act

We employ the event study methodology of Fama, Fisher, Jensen, and Roll (1969) to examine the behaviour of abnormal returns and abnormal liquidity after a firm whose stock is traded in the over-the-counter market switch its trading venue to an exchange market in Taiwan. The abnormal returns are calculated according to the market model. The abnormal liquidity is measured based on several commonly used liquidity measures including the proxy of spread of Elyasiani, Hauser, and Lauterbach (2000), Amivest liquidity measure, the variance of the residual of market model, and the abnormal turnover ratio proposed by Tkac (1999). We would like to stress that the abnormal turnover ratio has never been used to examine the liquidity effect of exchange switching event before. This is one of main contributions of this study. Our result shows that the abnormal returns are negative both before and after the announcement of exchange listing. Although the negative abnormal returns after exchange listing are consistent with the findings in the literature, those before exchange listing are different from what have been found in the literature. It implies that investors in Taiwan do not consider changing trading venue as a means of market value creation for over-the-counter firms. Regarding the behaviour of abnormal liquidity for exchange listing, we find that the liquidity improvement resulting from the event is generally disappointing although different liquidity measures reveal somewhat different results. This result is inconsistent with the liquidity gain hypothesis of exchange listing supported by the literature. Overall, the exchange market in Taiwan does not provide enough incentives for over-the-counter firms to change their trading venue from the OTC market to the exchange market.

Introduction and Empirical Results

In the literature of event study of exchange listing of stocks previously traded in the over-the-counter market, three common phenomena have been documented including the significant positive abnormal returns prior to the announcement date, the significant negative abnormal returns after the listing date, and the improvement of liquidity. Several explanations for the significant positive abnormal return and the improvement of liquidity have been proposed while the significant negative abnormal return still remains as a puzzle. Sanger and McConell(1986) propose two explanations for the significant positive abnormal stock returns before the announcement of listing. First, firms may decide to apply for exchange listing after experiencing a period of strong performance. Second, the abnormal returns may result from information leakage through insider trading in advance of the public announcement. Another possible explanation is from the literature of market microstructure that there exists a liquidity premium in capital markets, e.g., Amihud and Mendelson(1988) and Brennan, Chordia, and Subrahmanyam(1998). In other words, the significant positive abnormal stock returns may arise from the expected improvement of liquidity after the exchange listing. However, Barclay, Kandel, and Marx(1998) do not find evidence which supports this line of argument. In contrast, Elyasiani, Hauser, and Lauterbach(2000) do discover a significant relation between the abnormal stock return and two different kinds of liquidity measures. In addition, Clarkson and Thompson(1990) suggest that the expected reduction of estimation risk resulting from the exchange listing may cause the significant positive abnormal returns. As for the improved liquidity after listing, Barclay, Kandel, and Marx(1998) provide an explanation that the lower transaction cost in the exchange results in higher liquidity for individual stocks. We argue that the evidence on the significant positive abnormal returns and the improved liquidity should be reexamined within a more general framework before distinguishing between the aforementioned possible explanations. The framework we suggest can also shed some light on drawing a distinction between different explanations.

According to the event-study methodology suggested by Fama, Fisher, Jensen, and Roll(1969), abnormal return(AR) and cumulative abnormal return(CAR) are usually used to gauge the

5

the more general framework of event study suggested by De Jong, Kemna, and Kloek(1992) to calculate the abnormal stock returns during the period of event. This framework is based on Rosenberg’s(1973) “return to normalcy” model in which the beta of individual stocks follows an AR(1). It also considers the autocorrelation and heteroskedasticity of daily data. The Kalman filter technique is used to estimate the model. Apart from the estimated model parameters, we also have a complete time path of betas during the event period as a by-product of Kalman filter estimation. From the time path of beta, we are able to see whether beta actually decreases after listing as suggested by the hypothesis of estimation risk. Therefore, the first contribution of our paper is to reexamine the previous reported abnormal return within a more general framework and indirectly test the estimation risk hypothesis.

The liquidity of the stocks that apply for exchange listing is found to increase a great deal after listing. Elyasiani, Hauser, and Lauterbach(2000) use five different measures as proxies for liquidity and report significant liquidity improvement no matter which measure of liquidity is used. Barclay, Kandel, and Marx(1998) discover similar liquidity increase after listing although they use trading volume as a proxy for liquidity. Christie and Huang(1994) present similar evidence using transaction data. We argue that these measures all fail to disentangle the market-wide liquidity from the unique liquidity that solely belongs to individual stocks. Failure to segregate the market and unique components of liquidity measure may mistakenly attribute the liquidity improvement to exchange listing. This is because the observed improvement may probably come from the market-wide liquidity improvement rather than from the unique liquidity that we aim to gauge. Hence, it is very important to filter out the market-wide liquidity and use the unique liquidity in the event study of exchange listing. We use the equilibrium liquidity model of Tkac(1999) to accomplish this task. The proxy of liquidity we adopt is the share turnover as suggested by Lo and Wang(2000). Since the model of Tkac is very similar to the market model, we also extend it within the stochastic coefficient regression model alike to that proposed by De Jong, Kemna, and Kloek(1992). In other words, we assume the sensitivity of the liquidity of individual stock to the market-wide liquidity to follow an AR(1) model. Because the unique liquidity is frequently used to quantify how much information incorporated into trades, we can test the information leakage hypothesis of positive abnormal return based on the behavior of the unique liquidity. This is another contribution of our paper. Through the more general procedures discussed above, we expect to provide some insights into the behavior and further the possible causes of the abnormal stock returns and abnormal liquidity of the event study of exchange listing.

Our result shows that the abnormal returns are negative both before and after the announcement of exchange listing. Although the negative abnormal returns after exchange listing are consistent with the findings in the literature, those before exchange listing are different from what have been found in the literature. It implies that investors in Taiwan do not consider changing trading venue as a means of market value creation for over-the-counter

firms. Regarding the behaviour of abnormal liquidity for exchange listing, we find that the liquidity improvement resulting from the event is generally disappointing although different liquidity measures reveal somewhat different results. This result is inconsistent with the liquidity gain hypothesis of exchange listing supported by the literature. Overall, the exchange market in Taiwan does not provide enough incentives for over-the-counter firms to change their trading venue from the OTC market to the exchange market.

Refer ence

Amihud, Y., and H. Mendelson, 1988, Liquidity and Asset Prices: Financial Management Implications, Financial Management, 5-15.

Bhandari, A., T. Grammatikos, A.K. Makhija, and G. Papaioannou, 1989, Risk and Return on Newly Listed Stocks: The Post-Listing Experience, Journal of Financial Research 12, 93-102.

Brennan, M.J., T. Chordia, and A. Subrahmanyam, 1998, Alternative Factor Specifications, Security Characteristics, and the Cross-Section of Expected Stock Returns, Journal of Financial Economics.

Brown, M.J., E. Kandel, and L.M. Marx, 1998, The Effects of Transaction Costs on Stock Prices and Trading Volume, Journal of Financial Intermediation 7, 130-150.

Brown, S.J., and J.B. Warner, 1980, Measuring Security Price Performance, Journal of Financial Economics 8, 205-258.

7

with an Application to the Dutch Stock Market, Journal of Banking and Finance 16, 11-36.

Elyasiani, E., S. Hauser, and B. Lauterbach, 2000, Market Response to Liquidity Improvements: Evidence from Exchange Listings, Financial Review 41, 1-14.

Fama, E.F., L. Fisher, M.C. Jensen, and R. Roll, 1969, The Adjustment of Stock Prices to New Information, International Economic Review 10, 1-21.

Grammatikos, T., and G. Papaionnou, 1986, Market Reaction to NYSE Listings: Tests of the Marketability Gains Hypothesis, Journal of Financial Research 9, 215-227.

Lo, A.W., and J. Wang, 2000, Trading Volume: Definitions, Data Analysis, and Implications of Portfolio Theory, Journal of Finance.

MacKinlay, A.C., 1997, Event Studies in Economics and Finance, Journal of Economic Literature 35, 13-39.

Rosenberg, B., 1973, The Analysis of a Cross-Section of Time Series by Stochastically Convergent Parameter Regression, Annals of Economics and Social Measurement 2, 399-428.

Sanger, G.C., and J.J. McConnell, 1986, Stock Exchange Listings, Firm Value, and Security Market Efficiency: The Impact of NASDAQ, Journal of Financial and Quantitative Analysis 21, `1-25.

Tkac, P.A., 1999, A Trading Volume Benchmark: Theory and Evidence, Journal of Financial and Quantitative Analysis 34, 89-114.