This paper investigates both the information contents of recommendations disseminated by foreign security firms and the interaction of foreign security firms’ trading activities with their recommendations in Taiwan’s stock market. Using event study, correlation test, and regression analysis, we find negative average abcdrmal returns(AARs) and average cumulative abcdrmal returns(CARs) for negative and neutral foreign analysts’ recommendations levels and recommendation changes in the pre-recommendation period. AARs and CARs for positive recommendations in pre-recommendation period are positive, but reverse to negative three days after the event day. Our results also show that correlation coefficients of recommendations (both in recommendation levels and recommendation changes) and holding period returns are significantly positive in the pre-recommendation period, but insignificantly negative in the post-recommendation period. In the regression analyses, we find that price momentum factor is significantly related to foreign analysts’ recommendation, but the incremental contribution of this factor to foreign analysts’ recommendations are marginal and not significant. We also find that foreign security firms respond more rigorously to stocks receiving recommendation above buy recommendations and stocks being downgraded. These results show that foreign security firms are more conservative toward trading stocks in Taiwan’s stock market. They only buy stocks above buy recommendations (in a delay pattern), but immediately sell downgraded stocks.

This study examines the long-run accuracy of analyst recommendations issued at the expiration of the initial public offering (IPO) quiet period and examines the relation between the Global Settlement, NYSE Rule 472, NASD Rule 2711, and analyst recommendations. It is expected that firms which receive positive recommendations will outperform the market and firms with neutral recommendations. In addition, it is expected that banks named in the Global Settlement will become more selective when issuing recommendations. This study examines firms engaging in IPOs from July 9, 2002 through December 31, 2005 and finds that analyst ratings have become more balanced following the Global Settlement, NYSE Rule 472, and NASD Rule 2711. When controlling for firm size, underpricing, rating heterogeneity, and analyst affiliation, firms which receive positive analyst ratings experience greater buy-and-hold abnormal returns than firms which do not. Furthermore, firms which receive multiple “buy” ratings outperform firms which receive only one “buy” rating when controlling for underpricing firm size, and the number of neutral ratings. Banks named in the Global Settlement appear to be more selective when issuing positive recommendations. Firms which receive a positive rating from a bank named in the Global Settlement outperform firms which receive a positive rating from a bank not named in the Global Settlement. Lastly, prior to the Global Settlement, it appears that sanctioned banks issued ratings one level higher than they should have. Firms which received positive ratings experienced neutral performance and firms which received natural ratings experienced negative performance. Since the Global Settlement, sanctioned banks appear to issue accurate ratings since positive ratings are associated with increased buy-and-hold abnormal returns and neutral ratings do not significantly impact firm performance.

過去國內文獻大致指出投資人難以依靠券商的投資建議獲利，此與大部份國外文獻的發現相異。本文參考Barber et al. (2001)，建構一個適用於台灣股票市場的研究方法，再以四因子模型做實證。本文以2007年3月至2015年12月，共48987筆卷商個股報告為研究樣本，來探討券商報告的投資建議能否獲利。本文研究結果發現，台灣的券商報告擁有額外的資訊價值，此與Barber et al. (2001)及其他國外文獻大致相同。 本研究依券商的推薦強度建構四個投資組合。發現推薦程度高的投資組合平均月報酬為正，且高於大盤；而推薦程度低的投資組合平均月報酬顯著低於大盤，且擁有顯著的負超額報酬。本文進一步建構買進賣出策略，即買進推薦股票高的投資組合並賣出推薦程度低的投資組合，發現此策略報酬顯著高於零及大盤，且存在顯著的正超額報酬。另外在台股多頭期間，本研究的實證結果更加顯著，推薦程度高的投資組合平均月報酬增加至顯著高於大盤，且超額報酬顯著為正；推薦程度低的投資組合之大盤調整報酬及負超額報酬的顯著程度提高；而買進賣出策略獲得超額報酬的顯著程度也大幅提高。 / Past Taiwanese literatures generally indicated that it is difficult to obtain profit from Taiwanese stock recommendations of brokerage, which is different from most of foreign literatures. Referring to Barber et al. (2001), we improve and build a research methodology applied to Taiwanese stock market, conducting empirical analysis with four-factor model. From March 2007 to December 2015, we use total 48987 brokers’ stock recommendations as sample to investigate whether inventors could earn profit from the broker recommendations. Our empirical results show that Taiwanese broker reports hold additional information, which is consistent with Barber et al. (2001) and most of foreign literatures. According to the strength of recommendation, we construct four portfolios and find that the return of the most favorable portfolio is higher than market, while the return of the least favorable portfolio is significantly smaller than market and holds significantly negative access return. We further construct a long-short strategy, which buys the most favorable portfolios and shorts the least favorable portfolios. The return of this strategy is significantly higher than market, and excess return is significantly positive. During Taiwanese bull market, the significance of our empirical result improves. The significance level of market-adjusted return and access return for both the most favorable and least favorable portfolio is higher. In addition, the significance level of excess return for long-short strategy also greatly improves.

The full text will be available at the end of the embargo period: 31st December 2026.

Diese Arbeit umfasst drei unabhängige Aufsätze, welche die Auswirkungen verschiedener regulatorischer Maßnahmen auf das Bankenrisiko und/oder die Finanzstabilität untersuchen. Zunächst wird der Einfluss von Eigenkapitalanforderungen auf den Zusammenhang zwischen Bankgröße und Volatilität analysiert. Unsere Panel-Datenanalyse zeigt, dass strengere Eigenkapitalanforderungen den Nexus zwischen Größe und Volatilität schwächt. Große Banken haben, ceteris paribus, einen weniger volatilen Kreditbestand, wenn sie strengerer Kapitalregulierung ausgesetzt sind. Gemäß dem Granularitätskonzept kann dies ebenfalls die makroökonomische Stabilität erhöhen. Als Nächstes untersuche ich, ob MiFID II die frühzeitige Informationsweitergabe über Änderungen von Analystenempfehlungen an einzelne Anleger, genannt Tipping, reduziert hat. Die Ergebnisse zeigen, dass die absoluten Renditen und Handelsvolumina einen Tag vor Veröffentlichung einer Hoch- oder Herabstufung vor und nach Inkrafttreten von MiFID II signifikant ansteigen. Da die Aktienkurse am Veröffentlichungstag weiter steigen bzw. fallen, profitieren ausgewählte Anleger trotz der regulatorischen Änderung weiterhin von einem Informationsvorteil. Dies hat vermutlich negative Auswirkungen auf den Finanzmarkt insgesamt. Zuletzt untersuche ich wie sich die Ausgabe von Contingent Convertible (CoCo) Anleihen, die als regulatorisches zusätzliches Kernkapital (AT1) geltend gemacht werden können, auf das Bankenrisiko auswirkt. Meine Analyse zeigt, dass AT1-CoCo-Anleihen ein bis drei Jahre nach Ausgabe zu einem signifikant höheren Bankenrisiko führen. Übereinstimmend mit theoretischen Studien deutet dies darauf hin, dass CoCo-Anleihen ihr Potenzial zur Stärkung der Eigenkapitalbasis der Banken durch die regulatorischen Anforderungen genommen wurde. / This thesis comprises three independent essays evaluating the impact of different regulatory policies on bank risk and/or financial stability. First, we examine the effects of capital regulation on the link between bank size and volatility. Our panel data analysis reveals that more stringent capital regulation weakens the size-volatility nexus. Hence, large banks show, ceteris paribus, lower loan portfolio volatility when facing more stringent capital regulation. According to the granularity concept, that can increase macroeconomic stability. Next, I evaluate if MiFID II reduced the early information disclosure on analyst recommendation changes to selected investors - so-called tipping. I find absolute returns and turnover rise significantly on the day preceding the up- or downgrade release before and after MiFID II became law. Given that stock prices move further in the revision direction on publication day, selected investors continue to profit from an informational advantage, notwithstanding the regulatory change. That is likely harmful to the financial market overall. Lastly, I examine the impact of issuing contingent convertible (CoCo) bonds that qualify as regulatory additional tier 1 (AT1) capital on bank risk. My treatment effects analysis reveals that issuing AT1 CoCo bonds results in significantly higher risk-taking one to three years after the issuance. That is in line with previous theoretical studies suggesting that regulators have stripped CoCo bonds of their potential to strengthen the banks’ capital bases.