| 摘要: | 摘要
本論文由兩篇關於賣空限制對市場效率性以及對市場穩定性的影響之文章所構成。台灣股票市場於 2002 年至 2009 年間實施過多次賣空限制政令改變，提供很好的機會檢驗不同的賣空限制程度對於市場效率性和穩定性有何影響。在第一篇文章中，我們利用買賣權評價理論 （put-call parity） 所推衍的理論價格檢驗價格調整的效率性，發現越嚴格的賣空限制會造成價格調整所需的時間和速度不對稱，以及延遲價格調整的速度，即使控制市場狀況和流動性後仍發現一致的結果。
有趣的是，在擬真分析中，全面禁賣的政策宣告有助於防止價格崩跌，然而，我們發現嚴格賣空限制會引起較高的市場總風險與下方風險；更值得一提的是，我們偶然發現，先前的文獻在賣空限制對市場穩定性影響的討論中，明顯忽略內生性的問題，也就是政府監管單位通常在市場波動性特別高時限制賣空交易，如此難以判別賣空限制與市場波動性之間的相互關係。
為了解決內生性的問題，在第二篇文章中，我們利用Barndorff-Nielsen and Shephard (2004) 和 Huang and Tauchen (2005) 的跳躍偵測方法來控制內生性問題，進而探討不同賣空限制強度下的價格異常跳動。我們發現在嚴格賣空限制下，向下價格異常跳動明顯較大;然而，向上異常跳動在不同賣空限制強度沒有顯著的差異。
本論文的兩篇文章皆不支持政府對賣空限制的主張，實證發現限制賣空交易非但不能穩定市場價格，還令市場更波動且效率性降低。
第一篇文章是探討賣空限制對價格調整速度的影響以及檢驗賣空限制的施行是否符合主管機關所預期的市場穩定效果。台灣股票市場於 2002 年至 2009 年間曾有過多次賣空限制政令改變，提供很好的機會檢驗不同的賣空限制程度對於市場效率性和穩定性的影響。市場效率性定義為市場價格調整到選擇權市場隱含的現貨理論價格所需要的速度與時間長度，該理論價格由買賣權評價理論（put-call parity） 推衍，具有不須假設報酬分配型態和避免波動性估計誤差的優勢。此外，本文另一優勢，使用日內高頻資料分析，可觀察到更多價格發現的資訊內涵，在賣空限制相對嚴格的期間，被高估的市場價格下調到理論價格平均所需的日內調整時間，相較於寬鬆賣空限制期間的所需調整時間長，顯示賣空限制的嚴格程度與下調時間長度有顯著的關係；而低估的市場價格上調到理論價格平均所需的時間，在不同賣空限制程度下沒有顯著差異，此結果與賣空限制延遲負面消息反應在價格上的過去實證發現一致，故而嚴格賣空限制下所需的下調時間較長。進一步使用門檻誤差修正模型 (TECM），控制交易成本、市場微結構因素以及市場摩擦後，同樣發現價格調整到理論價格的速度在方向上有不對稱的現象，向下調整速度顯著比上調速度延遲許多，尤其嚴格賣空限制期間差異更大；也發現嚴格賣空限制相較於寬鬆期間的下調速度顯著減緩。我們亦透過政策擬真 (counter factual) 分析，雖然發現若不施行全面禁賣限制的價格會比實際實施全面禁賣期間的價格下跌地更劇烈，但嚴格賣空限制卻引起較高的總風險和下方風險，綜合以上結果顯示，賣空限制是犧牲市場流動性和市場效率性卻無法保證換取穩定市場。本文中釐清並佐證不同賣空限制對市場效率性與穩定性之各種實證效果探討，或可提供主管機關在實施賣空限制政策時的參考依據。
關鍵字：賣空限制、價格效率、市場穩定性、買賣權評價理論、門檻誤差修正模型、擬真分析。
第二篇文章經由跳躍活動的偵測（jump detection），探討賣空限制對市場穩定性的影響，本文是第一篇探討市場跳躍活動與賣空限制的關係，利用跳躍大小衡量市場穩定性，具有免除市場波動性與賣空限制的內生性問題。定義市場價格偏離買賣價評價理論推衍的理論價格之程度為賣空困難程度，市場價格偏離正值越大，顯示市場反應訊息的受阻程度越高。使用 2002 年至 2009 年的台灣股票市場加權指數日內資料，發現在高賣空限制強度下，市場價格異常向下跳動的程度相較於在低賣空限制強度下顯著大許多，但是對於向上價格異常跳動在不同賣空困難程度下沒有顯著的差異。實證結果沒有發現支持主管機關施行賣空限制能夠穩定市場的論點。
關鍵字：賣空限制、市場穩定性、買賣權評價理論、跳躍活動。
;Abstract
This dissertation contains two essays for discussing the effects of short sale constraints (SSC) on market efficiency and market stability. We take advantage of a natural experimental environment provided by the imposition of various short-sales policy regimes between 2002 and 2009 in the Taiwan stock market. In the first essay, we examine the put-call parity-defined price deviations in an attempt to assess the efficiency of price adjustments, and find that more extreme constraints lead to asymmetric and delay adjustment speed even when market conditions and liquidity are controlled.
Interestingly, in our counterfactual analysis, short-sales bans help to prevent price collapse, nonetheless, we find that the tightening of short-sales restrictions leads to increases in both market volatility and downside risk. We also occasionally found that there is an overlooked endogeneity among previous literature on the stabilization effect of short sale constraints in that regulator usually restrict short-selling to prevent market abnormal movements when the market volatility is particularly high. Therefore, a period of high price variation per se (as it can be the cause instead of the consequence) does not preclude the stabilizing effectiveness of the SSC.
To resolve the picture, in our second essay, we thus control for the endogeneity concerns by probing the jump activities under SSC using a jump identification approach by Barndorff-Nielsen and Shephard (2004) and Huang and Tauchen (2005). We observed that the abnormal downturns under tighter short sale constraints are significantly larger whereas there is no difference for abnormal upturns across different short-sale regimes.
Both of the two essays do not support the claims by regulators that restraining short-sales can stabilize prices; instead, SSC has led to a more volatile and less efficient market.
First Essay
Market Efficiency, Stability and Short Sales Constraints: Evidence from Taiwan
Market efficiency under short-sales constraints has long been an issue of debate within the related literature. In this study, we take advantage of a natural experimental environment provided by the imposition of various short-sales policy regimes between 2002 and 2009 in the Taiwan stock market. We examine the put-call parity-defined price deviations in an attempt to assess the efficiency of price adjustments, and find that more extreme constraints lead to asymmetric adjustment rates in upward and downward convergence to theoretical price; although short-sales constraints are generally found to suppress the effects of negative information on prices, our application of a ‘threshold error correction model’ (TECM) reveals more rapid convergence to theoretical price from upward adjustments than downward adjustments. Furthermore, price efficiency is found to be improved when such constraints are relaxed. These results hold even when market conditions and liquidity are controlled. Interestingly, although evidence from our counter-factual analysis suggests that tighter constraints help to prevent price collapse, despite the fact that prices may fall more sharply without short-sales bans, our findings refute the claim that tighter constraints can help to stabilize the market. Instead, we find that the tightening of short-sales restrictions leads to increases in both market volatility and downside risk.
Keywords: Short Sale Constraints, Price Efficiency, Market Stabilization, Put-call Parity, Threshold Error Correction Model, Counterfactual Analysis
Second Essay
Stabilizing the Market with Short Sale Constraint?
New Evidence from Price Jump Activities
This essay re-examine the impact of short-sale constraints (SSC) on market stabilization via realized jump activities during 2002~2009 to circumvent the reverse causality in identifying the policy effects of SSC. We observed that the abnormal downturns under tighter short sale constraints are significantly larger whereas there is no difference for abnormal upturns across different short-sale regimes. The findings do not support the claims by regulators that restraining short-sales can stabilize prices; instead, SSC has led to a more volatile and less efficient market
Keywords: short sale constraint, market stabilization, put-call-parity, jump activity |
