企業借助專業承銷機構的居間服務，自股權發行市場中取得資金是其持續成長的重要因素之一，居間服務中非單一承銷商所能獨自完成者乃配銷的部分，因此承銷團是西方投資銀行間最常見的組織間合作形式，但是，和許多組織間正式合作關係不同，承銷團的每一次組成都僅是為了去執行單一發行的幾個為數不多的交易。然而，這種生命週期短暫的臨時組織卻有賴於投資銀行間的長期穩定的非正式關係。因此，近年來逐漸有學者以網絡鑲嵌觀點來看西方大型投資銀行在證券承銷活動上的組織連結關係。開發中國家如台灣的資本市場未若西方先進國家成熟與健全，承銷商在組織連結關係之特性上是否亦有其不同之處。本論文以量化和質化的方式，分別從主辦承銷商、發行公司及協辦承銷商的角度出發，嘗試驗證過去西方文獻提出的連結策略之適用性、國內承銷網絡核心成員間連結關係之獨特性以及所引申出承銷商在協辦市場中的主動性與策略意涵。 本論文計分為三個研究，在第一個研究中，本論文以西方在投資銀行上的研究結果觀察國內主辦承銷商與其合作夥伴們的關係連結如何影響其承銷績效，藉以驗證已開發國家實證結果是否得以在開發中國家複製。本論文發現主辦承銷商無論與其合作夥伴(即協辦承銷商)維持強連結或是弱連結關係，均對其主辦承銷績效無顯著影響。顯示西方對於承銷活動中的主辦承銷商與合作夥伴間的組織連結關係策略在國內的情境下並不能複製。 本論文第二個研究於是繼續探索導致研究一的可能原因，亦即主辦承銷商在籌組承銷團的主控地位是否受到承銷網絡核心中其他成員—發行公司和協辦承銷商—的影響。本論文發現國內主辦承銷商在發行公司與承銷團員之間之「關係代理」角色並不若西方大型投資銀行之明顯。國內承銷商藉由其自身集團資源與主辦能力，得以直接與發行公司連結。本論文進一步分析此連結之特性，發現協辦承銷商與發行公司之連結傾向藉由「廣」而達到「多」的目的；亦即，承銷商爭取協辦承銷案件的多並非參與發行公司所發行的大多數案件，而是參與許多發行公司發行的有限案件。 然而，協辦承銷商的「多」與「廣」所形成在產業中的結構位置和弱連結現象的網絡特性終將反映到其承銷績效。因此，本論文繼續在研究三以次級資料驗證網絡特性與承銷績效之關聯性。研究發現，承銷商在網絡中的結構地位與其協辦承銷績效呈現正向關係；同時，承銷商的弱連結關係在上述結構地位與協辦承銷績效間之正向關係上扮演了部分中介角色。 總結來說，本論文對於股權發行市場承銷網絡中未為文獻所提及之發行公司與協辦承銷商二者之間的連結關係提供了實證文獻上的補充以及實務上的建議。與西方文獻比較起來，在開發中國家之發行市場中，決策所需資訊之正式來源較少，必須高度依賴人際互動，各大小協辦承銷商均可能為重要資訊來源，發行公司因此不透過主辦承銷商的關係代理，而必須與眾多之協辦承銷商產生直接連結；而各協辦承銷商亦得以藉由與承銷網絡外的關係、利用自身集團的資源或逐漸培養主辦的能力等爭取較多連結，提高在網絡中的結構地位，進而提高其承銷績效。 / Securities underwriting syndicate is one of the most common forms of ties among the investment banks. Being different from the long-term nature of the formal alliances such as strategic alliances, each syndicate is organized only for a single offering. But its brief lifespan must be sustained by the stable informal relationships among underwriters. The relationship-intensive nature of the investment banking industry has therefore drawn attentions to researches from the viewpoint of social network in recent years. The focus has been the large-scale investment banks acing as lead managers and their tie strategies with syndicate members. Capital markets in emerging markets such as Taiwan are not as sophisticated as those of developed countries. There must exist some uniqueness in the ties among underwriters when forming syndicates. How and why do they differ from the findings of current empirical studies? This study takes views of so-called the core underwriting network –lead underwriters, issuing firms and syndicate underwriters—respectively to examine the effect of the tie strategies proposed by prior literature on lead underwriters’ performances in local market, to explore the uniqueness and the role of syndicate underwriters in syndicate formation, and finally to evaluate the effects of those findings on underwriting performances of syndicate underwriters, which have been considered to be peripheral actors in industry network. We conducted three studies to explore these issues. Using longitudinal industry data, the first study takes the rationale provided by the results from past research to empirically examine the influence of lead underwriters’ tie strength with syndicate partners on their underwriting performances. The study finds that the tie strength neither positively nor negatively associated with lead underwriters’ performances, showing that the tie strategies suggested by the empirical results from the sophisticated capital markets do not apply to the local market. Using a case study approach, the second study continues to explore how syndicate formation in local market differs from that of the western world. The study finds that the mediating role of lead managers in local market appears to be not as significant as that of lead managers in sophisticated markets if we view lead managers as the agents between issuing firms and syndicate underwriters. In most cases, the issuing firms are the decision makers of the syndicates instead of the lead managers. That is, the connections between issuing firms and co-lead underwriters, which have never been found in current literature, do exist in local market. The underwriters speak directly to the issuing firms for the syndicates. And the more resources, capabilities and social capital they have, the more central positions in the industry they occupy and so the more exchange relationships they can access. Furthermore, underwriters tend to connect with as many issuing firms as possible for few of their offerings instead of continuously transacting with a small group of issuing firm partners for most of their offerings. But how the central position and relationship expansion strategy affect their underwriting performances? Following the findings from the second study, the third study develops arguments that underwriters occupy more central positions in the industry network will have better underwriting performances. They also pursue an expansion strategy instead of a strengthening strategy in building up interorganizational linkages. That is, weak ties will mediate the positive effects of centrality of syndicate members on the performances. Using longitudinal industry data, the study empirically tests these hypotheses and the results support our arguments that underwriters’ central positions in the network are –directly or through the mechanism of relationship expansion strategy—positively associated with their performances.