本文主要應用代理理論來評估建議目前全權委任契約的規劃草案。為深入瞭解全權委託契約的法律及經濟特性，本研究參考日本全權委託業務規範、及研究我國目前全權委託法令內容。藉實地訪談投顧業者以窺視業界對全權委託法令的真正看法，最後乃利用代理理論做為草案規範體系之基礎及進行評估。 在研究日本全權委託業務規範時，本研究發現日本的做法與目前我國草案的內容與精神大致上相同，因此難以日本作法來給我國全權委託規劃一些建議；至於規範台灣全權委託業務主要法規為「證券投資顧問事業管理規則」修正草案（重點內容請看第三章第二節），至本研究結束時還未經立法院三讀通過，為了要充分了解草案內容及業界意見，本研究在慎選二家投顧業者訪談後，發現相關法條中有許多不太合理的地方，例如為什麼在受託投資金額超過二十億元後，業者只要提存五千萬元的保證金就可以了，這樣是不是有圖利大型業者？以及對委託人而，開放業者投資國內開放型受益憑證似乎多此一舉了！（請看第三章第三節） 至於在應用 Holmstrom (1979) 代理理論方面，本研究認為此規範體系應具備以下特徵：1.在全權委託契約中不得採用固定佣金結構，否則主理人（投資者）將承擔所有全權委託契約的風險，如此代理人的道德危險會益加嚴重；2.應適當允許以績效為基礎的變動佣金結構，惟這種代客操作契約設計中的 F(x;a) 必具備一階強勢優勢的條件；3.要令 F(x;a) 的一階強勢優勢的條件在績效基礎的代客操作契約中能有效存在，則提出六個方法乃是必要條件（非充分條件）。（請看第四章第三節） 最後本文乃以先前所做的研究做為基礎，針對相關法條提出問題及建議方案，惟這部分的建議純粹是以本研究出發，完整的建議內容有賴各方面專家的共同參與才行。本文也希望對於此方面有興趣的研究者能參考本研究對後續研究者之建議部分，例如公會自律制度研究及歐美國家全權委託制度的研究均是十分值得繼續採所的領域。 / This study applies agency theory to assess the appropriateness of the regulatory framework of discretion account proposed by the Taiwan SFC (Securities and Futures Commission) which mainly imitates from Japan. This study considers investor as the principal and delegated investment company as the agent. Since the effort level of agent is unobservable, plus the interest conflict existing between the principal and agent, a rational agent would have moral hazard. If we plan to have an effective regulation of discretion account, it is necessary to design an efficient contract for eliminating or reducing agent's moral hazard. This study hypothesizes that Holmstrom's second-best agency contract could be the one for establishing an efficient regulatory framework of discretion account. Therefore, this study uses the propositions developed by Holmstrom (1979) to establish a proposed efficient regulatory framework for the business of discretion accout, including disclosure system and an incentive mechanism.

全權委託投資業務(俗稱代客操作),在證券市場中提供專業投資管理服務。我國證期會於民國81年，政策性地決定將開放此項業務。雖然整個規範體系已在民國87年建立，但它忽略掉經濟層面之考量，勢必造成全權委託投資契約無效率。 目前規範體系所允許之委託報酬的計算方法，稱之為「固定佣金制度」，由於此制度無法對業者提供經濟誘因，它會引發業者有嚴重之道德危險發生。因此，為解決誘因問題，應該使用「績效佣金制度」，不過在目前股市環境下，此一制度則引入過高之風險。因此，為解決兩種制度所帶來之問題，本文提出「改良式績效佣金制度」，在此制度下，委託報酬除基於傳統財務績效衡量值外，並基於非財務性績效衡量值，本文稱之為前瞻性績效衡量值（簡稱F-LPM）。 本文使用兩期之「主理人-代理人」模型，推導出最適委託報酬制度以及相關之比較靜態。在基本模式之下，本文進一步探討F-LPM之相關屬性及功能。所得結果與相關文獻一致，首先是，在最適委託報酬制度下，納入F-LPM可降低業者過度地重視短期，因而犧牲了委任人之報酬，這表示F-LPM具有增額資訊內涵。其次，無成本地提高F-LPM之精確度，可增加委任人之報酬。最後，改良式績效佣金制度優於固定佣金制度與績效佣金制度，因為它可降低風險，因而帶給委任人之報酬增加。 第壹章 緒論 第一節 / Discretion account business provides professional investment management service in the securities market. The Securities and Futures Commission decides in policy to approve the business in 1992. Althrough the regulation system of the discretion account business has been established, it will make the advisory contract inefficient with the negligence of economic consideration. The form of advisory fees permitted by the regulation system is 'constant fee system'. Under the system, advisory contract provides no incentive and serious moral harazd will occur. In order to solve the motivation problem, 'performance fee system' should be used. Under the circumstance of Taipei securities market, however, the performance fee system will bring in high risk. Hence, this paper suggests the 'improved performance fee system' to overcome the dilemma born by the performance fee system. In the improved system, advisory fees are based on the traditional performance measure and the fordward-looking performance measure (hereafter F-LPM). With a two-period principal-agent model, this paper develops the optimal reward system and related comparative static. Following the basic model, the paper examines the attributes and function of F-LPM. The findings are consistent with the related literature. First, the inclusion of a F-LPM in reward system can mitigate the potential for the advisor to focus excessively on the short-term at the expense of the investor. That is, F-LPM is incremental informative. Second, costless increases in precision of F-LPM will lead to increase in payoffs for the inveator. Finally, the improved performance fee system is superior to the constant fee system and the performance fee system in that it can reduce risk and results in higher payoffs to the investor.