次貸風暴發生後，各國金融監理當局對於金融機構壓力測試益發重視，本文以香港金管局研究員Wong and Hui(2009)所建立的流動性風險壓力測試模型為主要架構，根據本國銀行的資產負債結構加以修改，以台灣上市銀行為主要研究對象，對其進行壓力測試。測試銀行在金融資產價格大跌的壓力情境下對於流動性風險的忍受能力。另外，本文嘗試以向量自我迴歸模型建立一個總體壓力情境，並以違約迴歸模型連結銀行放款對象的違約率與總體風險因子，再將其與此流動性風險壓力測試模型結合，觀察總體壓力情境下，各銀行的流動性風險忍受能力。

一、出事銀行的流動性創造 流動性創造是銀行的主要功能之一，但從傳統的存放款到近年來蓬勃發展的表外交易，銀行長期頻繁地使用以短支長的工具，流動性創造最後竟成了次貸危機銀行受傷的一個關鍵原因。本文以次貸危機期間美國和歐盟共18個國家的商業銀行為研究對象，檢視其流動性創造的效果，並將出事銀行區分為擠兌銀行、紓困銀行和倒閉銀行，呈現流動性創造帶來的不同衝擊。此外，本文還進一步地分析銀行盈餘結構轉變造成的影響，觀察是否會增加或抑制原先流動性創造的效果。 實證結果顯示，高度的流動性創造會增加銀行出事的可能性，且收入多角化程度的提高，會更增加流動性創造帶來的負面效果。而以市場為基礎的國家，因為銀行活動和市場相關性高，能夠創造出較多的流動性，但銀行出事時，也容易選擇紓困銀行而不讓其倒閉，避免對整體金融環境造成太大的負外部性。最後，限制銀行從事證券活動，則有助於銀行免於紓困。 二、成本管理的銀行效率 本文是第一篇結合成本管理和銀行效率的研究。在盈餘管理的文獻中，呆帳費用 (Provision for Loan Loss, PLL) 被認為是銀行最常用來操弄盈餘的科目。因為它是銀行最大的應計項目，經理人往往利用將其提前或遞延認列來進行盈餘管理。甚至，政府的寬容政策、管制改變和企業醜聞等因素，也都會讓呆帳費用產生大幅波動。然而，呆帳費用是總成本的一部份，所有造成呆帳費用波動的原因，都可能進一步影響總成本，使總成本也大幅波動，本文將這樣的影響稱之為「成本管理」。過往成本效率的研究，使用的都是經過成本管理的總成本，而偏離真正總成本的結果導致銀行的效率表現也容易錯估。 本文提出了一個降低財報上呆帳費用波動的方法，盡可能地讓其在當期認列，成為符合經濟意涵的呆帳費用 (即「經濟呆帳費用」)。然後，用經濟呆帳費用求得當期理論上的總成本 (即「經濟總成本」)，並重新對台灣29家商業銀行的成本效率進行估計。實證結果顯示，經濟總成本不但遠高於財報上的總成本，所估得的效率排名也和原先有很大的不同，但明顯較符合現實狀況。

當保險契約被發行時，保險公司必須被要求盡可能的具備承擔未來不可知的風險的能力。本文將死亡風險、利率風險、脫退風險以及流動性風險引入，並針對生死合險契約進行盈餘分析。在此以 Vasicek (1977) 所提出之隨機利率模型、根據被保險人理性行為作為基礎之脫退模型以及引入簡化後的 Longstaff、Mithal與Nies (2005)流動性風險債券價格來描述各種風險。根據上述模型假設下計算保費及準備金，遂以蒙地卡羅模擬法量化源於各種風險之盈餘。最後，本文計算保險公司之盈餘對各風險參數之敏感度分析，並計算各期破產與發生流動性問題之可能性。 / Once insurance contracts are issued, the insurers should be capable to deal with the unknown conditions in the future as possible. In this paper, we analyze the impact of mortality, interest rate, surrender and liquidity risks on the surplus of endowment contract. We model the interest rate risk by Vasicek model, the surrender rate based on the rational behavior of policyholders and introduce the discounted price of zero coupon bonds as the liquidity risk. Under such assumptions, we compute the premium and reserve, demonstrate the simulated insurance surplus, and finally exhibit the statistics of the surplus from different sources. The simulated results show the sensitivity of the surplus to the parameters of the risks. At the same time, we also show the probabilities of insolvency and illiquidity of the insurer before the maturity date of the contract due to the fluctuating surrender rate and liquidity risk resulting from the stochastic interest rate.

本文利用搜尋模型，為勞工流動與人口遷移提供一套理論的分析架構，並且找出一套衡量流動性與人口遷移的工具與指標。模型假設為封閉經濟體系內只存在兩個地區（高工資城市與低工資鄉村），而有 的同質個人，將居住遷移與工作流動視為單一的決策，加入房屋市場且兩地存在外生的工資差異，而且沒有失業的存在，個人在「工資差距」、「遷移成本」與「居住成本」的考量下做出遷移決策，當人口遷移達到均衡時可求出在steady-state下衡量遷移的指標，由比較靜態分析可知，工資差距與遷移呈現正向關係、而遷移成本與遷移存在負向關係，結果與客觀推論一致。 在進行福利分析時，若社會福利函數為簡單加總型態時，發現經濟體系各地區最適人口數與均衡人口數產生不一致的現象，其中一個與現實符合的情況為鄉村人口過度外移，在同時滿足「均衡與最適一致」、「收支平衡」與「最適不變」三個條件下，提供一個「差異性」的房屋稅制矯正均衡與最適偏離的問題，為相關政策制訂提供一項依據。

興櫃股票是指已經申報上市（櫃）之公開發行公司的普通股股票，一般而言興櫃股票的流動性小、股本小，因此短線投資人多以即將掛牌上櫃的公司為主，至於中長線的投資人則以長期價值的觀點來投資，投入資金後必須等待一段時間，待上市（櫃）後，才能獲取較好的投資報酬率。長期投資須以自有資金投入，進一步瞭解該公司產業狀況及基本面表現後，應用適合未上市公司的評價模式，來做為投資的依據是很重要的。 本研究以興櫃市場中的半導體產業中的廣穎電通股份有限公司作為研究對象，以自由現金流量法為評價模型，並針對未上市公司在參數估計困難的部分作調整，使評價未上市公司時有一套可依循的流程，並加入考量未上市公司股票因流動性不足，所造成之流動性折價，以提升在評價上的合理性及正確性。 實證結果顯示，以2011 年年報來估計相關參數，運用現金流量折現法的每股理論價值為39.24 元。若考量流動性折價，以上市股票市場資料推估所得之流動性折價，廣穎電通之每股理論價值將為39.05 元；以興櫃股票市場資料推估所得之流動性折價，廣穎電通之每股理論價值將為37.93 元。

京都大学 / 新制・課程博士 / 博士(工学) / 甲第23234号 / 工博第4878号 / 新制||工||1761(附属図書館) / 京都大学大学院工学研究科化学工学専攻 / (主査)教授 松坂 修二, 教授 宮原 稔, 教授 山本 量一 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

金融海嘯提供我們一個深刻的教訓，因為危機前信貸過度增長伴隨著大量的系統風險，最後導致景氣反轉時銀行業龐大損失。而這些損失將動搖整個金融體系，並引發了一連串的惡性循環(Basel Committee on Banking Supervision , BCBS ,2010a, 2010b)；若依循過往個別審慎監理((Microprudential regulation)原則，將無法察覺背後隱藏的系統風險。因此目前趨勢是將以個別(Micro)與總體審慎監理原則(macro)並重，針對能夠影響整體市場金融穩定風險來源而詳加監管，同時透過規範與監理措施適度的降低系統風險，最終達到金融穩定的目的。IMF、BIS以及FSB(2009)針對G20制定的金融機構、市場與工具的指導文件(Guidance to Assess the Systemic Importance of Financial Institutions, Markets and Instruments)中，認為有效控制系統風險是現階段政策監理最重要的主軸之一。所謂系統風險是指能影響金融機構所持有的部位以及對於實體經濟存在嚴重負面影響的風險來源；此總體風險將存在負外部性而非個別審慎監理的風險因子。 因此本文由兩篇宏觀審慎監管框架文章所構成的研究，分別針對市場流動性風險和信用風險的因子。透過非流動性賣權與逆景氣資本緩衝（CCB）買權來分析和評價兩種新的監管框架。第一篇論文的主要概念是討論市場流動性風險因子，雖然當前銀行監管的重點是資金流動性風險，如新巴塞爾協議三 (Basel III)的流動性風險覆蓋率（LCR）和淨穩定資金比率（NSFR），但金融機構實際上也同時面臨資金和市場流動性之間的高度順週期效應，導致流動性螺旋，並威脅到金融穩定。因此，本文提出一個市場流動性，系統性風險和宏觀審慎監理分析框架來填補這一空白。 與Drehmann和Juselius（2013B）的實證研究結果比較，我們發現利用6個月歷史波動度建構的非流動性選擇權是最有效的提前預警指標（EWIS），且符合穩定政策結構和最小監管成本。此外在三個子樣本和嚴重危機時期亦能同樣保持預警的穩健性。因此如果金融機構能透過預警減少金融機構投資種類、行業、交易對手與大額暴險的集中度時，將可以由危機發生後被動式轉變成危機發生前主動式的風險管理，將符合總體審慎監理定義：能影響所有而非單一的金融機構，以及有效控制破壞總體市場產生的系統風險。 在第二篇文章中，我們專注於信用風險監管框架的避險，即Basel III的逆景氣資本緩衝(CCB)。這個新穎的監理視角將鼓勵銀行在危機前的信貸繁榮時期增加資本緩衝，而非在危機後接受援助或者增加昂貴的資本。據美國聯邦存款保險公司（FDIC）統計，2014年第1季全美的存款機構風險加權資產為10.27兆美金；如果最高的逆景氣資本緩衝被應用到這些銀行，將有2570億美金的資本不得不額外注資。因此本文設計了一個新的買權來符合CCB的監管框架，建立提前資本防禦措施來減輕系統性風險和整體銀行業不穩定。首先發現這款買權將能在順境時注入資本，即更低的潛在違約風險與信貸寬鬆時期，進而抵禦未來發生的金融危機。我們的建議也符合Basel III的目標，在危機前2至5年協助銀行取得資本保護。最重要的是，CCB買權可以透過提前取得資本形成一個“減震器”，舒緩隨後而來經濟衰退的壓力達到降低銀行資本順週期性目標；此外還提供了一個對於銀行過度冒險行為的抗衡力量，成為一個“自動穩定器”來達到宏觀審慎監理目標。 / Financial tsunami offered a profound lesson as the pre-crisis excessive credit growth was accompanied by huge systemic risks that ultimately led to the reversal of economy and huge losses of the banking sector. Such losses will shake the entire financial system and trigger a series of vicious cycles (Basel Committee on Banking Supervision, BCBS, 2010a , 2010b ); the hidden systemic risk may not be observed if we follow the previous principles of micro prudential regulation. The guidance formulated by G20 to assess the systemic importance of financial institutions, markets and instruments (IMF, BIS, and FSB, 2009) analyzes that the main issue of prior micro prudential regulation is that every financial institution’s incentive is to manage its own return-risk tradeoff but not necessarily manage the stability for the financial system as a whole. Consequently, the macroprudential regulation focusing on shocks originating outside the financial system can control the negative externalities of systemic risk rather than micro prudential regulation. This dissertation consists of two essays on the macro prudential framework of market liquidity risk and credit risk factor. We introduce, analyze, and value two new regulation frameworks via an illiquidity put option and a CCB call option respectively. The main concept of first essay is to discuss the macro prudential framework of market liquidity risk factor. Although the current banking regulation focuses on systemic funding liquidity risk such as Liquidity Coverage Ratio (LCR) and Net Stable Funding Ratio (NSFR) of Basel III, financial institutions would actually have highly procyclical effects between funding and market liquidity at the same time, leading to liquidity spirals and threatening to financial stability. We therefore propose a market liquidity, systemic risk and macroregulation analysis framework in Taiwan's capital market to fill this gap. Comparison with the Drehmann and Juselius' empirical study (2013b), we find that illiquidity options by using 6-month historical volatility and forecasting short-term stock declines are effective early warning indicators (EWIs) having most stable policy structures and minimal regulation costs. Applying AUC macroregulation criteria, we show this illiquidity measure is also maintained fairly robustness in different intervals, e.g. during three sub-samples and serious crisis periods. If financial institutions can diversify the concentration of portfolios varieties, industries, and counterparty before crises by using EWIs, the passive risk taking can be converted into the active risk management. It is necessary to prepare the market liquidity and macroregulation framework in advance. In the second essay, we focus the hedging product for credit risk factors, i.e. countercyclical capital buffer (CCB). This purpose of countercyclical capital buffer standards is to encourage banks to increase capital buffers in credit good times that can be used in the future stress. According to Federal Deposit Insurance Corporation (FDIC), the risk-weighted assets of U.S. depository institutions were $10.27trillion dollars in 2014:Q1. If the maximum CCB is applied to these banks, an additional US$257 billion of equity capital will have to be raised. Hence, we design a new option to establish the capital defenses meeting CCB framework and then mitigating systemic risk and banking instability in advance. We show this product injects capital in good times i.e., lower credit risk and more credit expansion, to weather the future financial crisis. Our proposal also complies with the goal of Basel III to obtain capital in 2 to 5 years prior to crises. Most importantly, the CCB option can provide protection with additional capital to act as a "shock absorber" reducing a procyclicality problem in the subsequent downturn. Besides, this type of option also offers a countervailing force to excessive risk-taking behaviors to act as an "automatic stabilizer" for reaching macroprudential goals.

美日歐等先進國家在傳統貨幣政策提振經濟的效果逐漸失靈後，開始大膽啓用量化寛鬆(Quantitative Easing, QE)等非傳統貨幣政策(unconventional monetary policy)。本研究以此為背景，透過大量數據蒐集、整理與分析，比較國際金融市場在量化寛鬆貨幣政策前後的變化。 2007年美國次貸風暴引發全球金融危機後，FED於2008年開始實施密集、快速的QE政策，全面利用央行資產負債表與獨特而絕對在貨幣價格及數量的無限權力。本研究試圖從債券市場、股票市場、外匯市場、信貸與投資市場、各部門負債與去槓桿化程度與全球通貨膨脹現況等不同面向切入，嘗試以較長時間的統計資料比較並說明QE的有效性與侷限性。 在美日歐相繼實行規模程度不一的QE政策後，研究發現美國市場的各個層面已有顯著的改進，特別是在就業市場部分，其中失業率已逼近自然失業率的充分就業狀況，代表美國將逼近升息的時間點，並且不是只有單次調高基準利率(Fed Fund Rate，FFR)，而是一個升息循環的開始，預期FFR將在未來數年內逐漸調高到正常經濟的水平。 與此同時，開發中國家則在資金外流回到美元體系的大環境下，呈現匯率貶值、股市表現不佳與主權債利差變大的金融現象，反應出國際金融市場風險正移轉至新興市場；尤其令人不安的是新興市場持續累積相對龐大的負債，以及國際商品大跌，嚴重衝擊以出口這些商品為主的新興經濟體與生產製造商。在各國央行貨幣政策趨於分岐，特別是美國啟動升息周期將成為美元持續走強的驅動力，龐大國際資金的流動亦會顯著帶來市場風險的移轉。 本研究蒐集各種不同領域的報告及資料，進行分析，主要結果整理如下： 一、量化寬鬆政策(Quantitative Easing)將在一開始造成實施國的貨幣明顯貶值。 二、量化寬鬆政策將造成實施國的股票市場持續走揚。 三、量化寬鬆政策將造成實施國的主權債券殖利率明顯下跌，特別是短天期的部分(short-end)。若實施國進一步採行負利率政策，短天期主權債券收益率亦會由正轉負。 四、量化寬鬆政策帶動全球通貨膨脹脫離偏低水準的效果不明顯，無論已開發或開發中國家都仍深陷通貨緊縮的壓力。 五、在美日歐相繼實施量化寬鬆政策後，全球各商品(市場)的波動趨於一致，呈現越來越高的關聯性。 六、全球金融市場流動性有逐漸降低的趨勢(受到各國管理金融業法規趨嚴影響)，對照市場波動性時大時小，流動性風險影響國際資金的資產配置與流動成為一個重要議題。

Abstract In this paper, we examine the relation among different information asymmetry measures in Taiwan Stock Exchange and exploit the ability of the microstructure measures to measure asymmetric information. We also investigate the role of information asymmetry measures in affecting stock returns. With a random sample of 180 firms, we find that the market microstructure measure is significantly correlated with most of the corporate finance measures that should shed lights on the level of information asymmetry in advance. We also find that the analysts’ forecast measures have no relation with the microstructure measure. One main result is that the adverse selection risk does affect the stock returns. For the whole sample period, the adverse selection component has a significant impact on the stock returns and dominates all other variables except for the number of analysts following. Other significant measures include the volatility, firm size, leverage, and market to book ratio of equity. Although these information asymmetry measures act as competent determinants in the whole- period regression, they do not have consistent performance across quarters. The inconsistent result suggests that these measures may have diverse performance with regard to different periods.

This paper investigates the liquidity and the bid-ask spread components of the Taiwan Stock Exchange Capitalization Weighted Stock Index futures contracts, Taiwan Stock Exchange Electronic Sector Index futures contracts, and Taiwan Stock Exchange Banking and Insurance Sector Index futures contracts traded on the Taiwan Futures Exchange, which switched from an electronic periodic call auction market to an electronic continuous auction market on July 29th 2002. It is a rare opportunity to deeply examine the liquidity and transaction cost components of financial derivatives under different trading mechanisms. Using intraday transaction data of transaction and quotes covering from March 2002 to May 2002 for the old trading mechanism and from October 2002 to December 2002 for the new trading mechanism, liquidity measures and bid-ask spread components are examined before and after the enforcement of the electronic continuous auction mechanism. First, for each type of futures contracts, liquidity measures including bid-ask spread, trading volume, trade number, trade size, volatility, and liquidity ratio are explored to show the multifacet of liquidity. Next, the model of Lin, Sanger, and Booth (1995) is used to decompose the spreads of each product in the two periods. The empirical results show that quote spreads, effective spreads, percentage effective spreads, and dollar-weighted percentage effective spreads of the new system are all significantly lower than those measures in the old system for all of the three types of futures contracts. However, other liquidity measures do not show the same patterns. Overall, improvement of liquidity is found for futures contracts but not very consistent though. Multifacet of liquidity is showed by different measures, although two of these measures, including trade size and trade number, may not be suitable for this study. Moreover, the adverse selection is the most important component in the call auction market, which decreases in the continuous auction market. However, the change of other components, including order processing cost and order persistence, does not demonstrate the same pattern. The results indicate that the electronic continuous auction system protects uninformed traders from being hurt by informed traders. However, we also show that each type of futures contracts has its own specific component structure.