以知識經濟的觀點建構廠商垂直關係上的共演化聯盟模型

Co-evolution Alliances in Vertical Relationships: The Knowledge Economy Perspective

洪德芳(Der-Fang Hung)

指導教授 ： 湯明哲

國立臺灣大學/管理學院/國際企業學研究所/博士(2012年)

https://doi.org/10.6342/NTU.2012.01903

摘要

本研究源起於對知識密集產業(knowledge-intensive industries)的觀察。在這樣的產業中，上游的設備生產廠商會不斷地與下游的設備使用廠商合作開發新的製程設備，學者稱這樣的情況為共演化(co-evolution)現象(Koza & Lewin, 1998; Levinthal & Myatt, 1994; Sturgeon & Lee, 2001; Volberda & Lewin, 2003; Wilson & Hynes, 2009)。然而本研究卻關心一個基本的問題：為什麼共演化的現象會存在於垂直的關係上？依據經濟學家的觀察，共演化所形成的聯盟可能是經濟租(economic rent)的不同來源，如關係租(relational rent)(Dyer & Singh, 1998)。但對策略學者而言，一個共演化的聯盟可能是競爭優勢的不同來源，如鎖喉策略(foreclosure strategy)(Allen, 1971; Hart, Tirole, Carlton, & Williamson, 1990; Rey & Tirole, 2007)。而從知識經濟的觀點來說，共演化聯盟的出現是因為知識屬性所引起的交易成本使然。這些知識的屬性包含：知識的複雜程度(knowledge complexity)、知識的範疇經濟(scope economy of knowledge)、知識的不對稱性(knowledge asymmetry)及知識的不可締約性(knowledge noncontractibility)。由於本研究的興趣在於垂直關係上公司的策略思維，所以定義共演化聯盟為一種合作式的合約安排。在這個安排下，一家公司的知識改變，將會影響另一家公司的知識改變，進而回頭影響第一家公司知識的再改變，並隨著時間周而復始。
既然共演化聯盟是公司的策略選擇，必然也遵循利潤極大化的策略選擇原則。再者，共演化聯盟是一種混合式的治理模式(hybrid governance mode)。是故，要證明共演化聯盟在某些條件下是最適的治理模式，得比較不同治理模式下其治理成本的高低，愈低者，代表其治理的模式愈是最佳的選擇。Williamson (1991)認為治理的模式主要可區分成：市場(market)、組織(organization)與混合模式(hybrid)。而在混合模式中，本研究再進一步區分成：策略聯盟(strategic alliance)、聯盟組合(alliance portfolio)與共演化聯盟(co-evolution alliance)等三種治理模式。在治理成本上，本研究認為在知識密集的產業中，治理成本應包含：知識複雜程度所帶來的成本增加、知識範疇經濟所帶來的成本降低、與知識不對稱性和知識不可締約性所帶來的機會成本的增加。因此，本研究提出三個命題：1) 當知識複雜程度愈高時，市場的治理成本將高於組織的治理成本；2) 當知識複雜程度愈高時，組織的治理成本將高於混合模式的治理成本；3) 在考量機會成本下，當知識的複雜程度愈高，共演化聯盟的治理成本將低於策略聯盟與聯盟組合的治理成本。
在論述上，本研究先行探討垂直整合(如：組織治理模式)與垂直控制(如：市場或混合治理模式)的差異。其中，最大的差異就是交易成本的不同(Chen & Hylton, 1998; Mahoney, 1992)。若沒有交易成本，則垂直整合與垂直控制並無不同。但事實上，交易成本無所不在。而在知識密集的產業中，知識屬性所引起的交易成本，更是影響深遠。
知識，Fransman (1998)稱之為：經過處理的資訊，具有內隱性與情境相依的本質。它具有許多屬性，其中最重要的有：知識複雜程性、知識的範疇經濟性、知識的不對稱性與知識的不可締約性。知識的複雜性源自於知識的互動性，若知識的存取不需透過知識擁有者間的任何互動時，則稱此知識的複雜程度為零。知識的範疇經濟源自於知識不會減少的特性，運用於愈多的不同領域中，對知識的掌握與使用的成本會相對降低。知識的不對稱性源自於知識的多樣性，使得每個人所擁有的知識不必然一樣，而形成不對稱性。知識的不可締約性源自於知識的內隱性，無法於合約上將知識的運用細節描述清楚。本研究利用Cobb-Douglas生產函數，將這四個知識的屬性所引起的成本增減，整合成一個知識成本函數(如下)，再利用此函數，找出不同治理模式的成本。最後，再比較不同治理模式的成本高低，證明本研究先前所提出的三個命題。

首先，本研究將市場的治理成本除以公司的治理成本，若比值大於1，則代表市場的治理成本大於公司的治理成本；反之，則市場的治理成本小於公司的治理成本。結果，市場的治理成本大於公司的治理成本，支持本研究的第一個命題。

再者，本研究將公司的治理成本除以混合模式的治理成本，若比值大於1，則代表公司的治理成本大於混合模式的治理成本；反之，則公司的治理成本小於混合模式的治理成本。結果，公司的治理成本大於混合模式的治理成本，支持本研究的第二個命題。

最後，本研究將策略聯盟或聯盟組合的治理成本除以共演化聯盟的治理成本，若比值大於1，則代表策略聯盟或聯盟組合的治理成本大於共演化聯盟的治理成本；反之，則策略聯盟或聯盟組合的治理成本小於共演化聯盟的治理成本。結果，策略聯盟與聯盟組合的治理成本皆大於共演化聯盟的治理成本，支持本研究的第三個命題。


透過三個命題的支持，本研究證明了在知識密集的產業中，隨著知識複雜程度的增加，共演化聯盟將會是最適的治理模式。

關鍵字

共演化聯盟 ； 策略聯盟 ； 聯盟組合 ； 治理成本 ； 知識複雜度 ； 知識範疇經濟 ； 機會成本

並列摘要

This research emerges from a phenomenon which occurs in knowledge-intensive industries where equipment companies collaborate unceasingly with their customers. Scholars refer to it as co-evolution (Koza & Lewin, 1998; Levinthal & Myatt, 1994; Sturgeon & Lee, 2001; Volberda & Lewin, 2003; Wilson & Hynes, 2009). I ask a fundmental question, why the co-evolution phenomenon exists in a vertical relationship. From the viewpoint of economists, a co-evolution alliance could be a different source of economic rent, such as relational rent (Dyer & Singh, 1998). For strategists, a co-evolution alliance could be a different source of competitive advantage, such as foreclosure (Allen, 1971; Hart et al., 1990; Rey & Tirole, 2007). However, from the viewpoint of knowledge economy, the reason why the co-evolution alliance emerges is the transaction cost derived from knowledge characteristics, that is, knowledge complexity, knowledge specialization, knowledge asymmetry and knowledge noncontractibility. This thesis is concerned with the firms’ strategies in a vertical relationship and defines a co-evolution alliance as a cooperative arrangement in which the knowledge change of one firm will influence the knowledge change of its counterparts over time. According to the definition, it is obviously belong to a firm-level study. In addition, this research regards a co-evolution alliance as a strategic choice for a firm pursuing its best performance according to the rule of maximizing profit. Moreover, governance cost in this thesis is viewed as the aggregation of the cost induced by knowledge complexity, the effect contributed by knowledge specialization, and opportunistic cost. Accordingly, how to prove the governance cost of a co-evolution alliance is less than the cost of market and the cost of a firm becomes the core research question. Hence there are three hypotheses which need to be supported. The first hypothesis is that the governance cost of the market is greater than that within a firm under high knowledge complexity. The second hypothesis is that the governance cost within a firm is greater than that of a hybrid mode under high knowledge complexity. The third hypothesis is that considering opportunism cost, the co-evolution alliance is the optimal hybrid mode under high knowledge complexity.
In reasoning, I argue that the relational rent which is proposed by Dyer and Singh (1998) is one of the sources for maximizing profit. But that raises another question: why does a firm not integrate the counterpart if their cooperation is frequent and profitable? The core of this question could be viewed as an issue of vertical control, that is, the decision of trade-off between vertical restraints and vertical integration.
The distinction between vertical restraints and vertical integration is the consideration of transaction costs, in particular contracting costs. Without transaction costs, the advantages that result from vertical integration can be replaced by vertical restraints (Chen & Hylton, 1998; Mahoney, 1992). In fact, transaction costs can be discovered in most of exchange relations and can vary with the characteristics of transacted objects. In knowledge-intensive industries, knowledge is the important resource and the core transacted object. Hence, knowledge characteristics influence the transaction cost and the choice of vertical control mode.
According to Fransman’s (1998) research, knowledge can be defined as processed information with a tacit and context-dependent nature. It has two important characteristics: knowledge complexity and knowledge specialization. Knowledge complexity originates from the interactivity of knowledge (Goldwasser, Micali, & Rackoff, 1985). In other words, knowledge that can be accessed without any interaction has zero complexity (Goldreich & Petrank, 1991). Knowledge specialization, derived from the undiminished feature of knowledge as it is used (Foss, 2005), could be defined as the applied cost of knowledge that is reduced by deploying frequently. Hence, the transaction cost of a firm is related to knowledge complexity and knowledge specialization. The transaction cost will increase as the knowledge complexity is growing; contrarily, the transaction cost will reduce as the knowledge specialization is deepening.

The knowledge cost function can be derived by combining the two transaction costs and the cost of possessed resource into the knowledge production function. As a result, the knowledge cost function has maximum cost when a firm owns two knowledge varieties simultaneously. Furthermore, if the rate of coordination cost of the market to governance cost of a firm is greater than 1, it means that the governance cost of the market is greater than that within a firm. Otherwise, it means the governance cost of the market is less than that within a firm. As a result, the rate is greater than 1 when knowledge complexity gradually increases. Accordingly, this result supports hypothesis 1, which suggested that the coordination cost in the market is greater than that within an organization as knowledge complexity gradually increases.

Similarly, if the rate of coordination cost of a firm to governance cost of a hybrid mode is greater than 1, it means that the governance cost of a firm is greater than that within a hybrid mode. Otherwise, it means the governance cost of a firm is less than that within a hybrid mode. As a result, the governance cost within a firm is greater than that of a hybrid mode as knowledge complexity gradually increases. In other words, a firm is not an optimal governance type when a new production process needs to integrate two different knowledge sets. This implies that cooperative co-specialization has its economic value. Hence, this result supports hypothesis 2, which stated that the governance cost within a firm is greater than that within a hybrid mode as knowledge complexity gradually increases.

Considering only knowledge complexity and knowledge specialization, I directly prove that the hybrid mode is an optimal choice in the case of co-specialization. In the next step, I consider opportunism costs, which stems from knowledge asymmetry and knowledge noncontractibility, and put it into the knowledge cost function to prove that the co-evolution alliance is the optimal governance mode among three hybrid modes: strategic alliance, alliance portfolio and co-evolution alliance. If the same calculative technique is applied, the rate of governance cost of strategic alliance (or alliance portfolio) to governance cost of co-evolution alliance shows that the governance cost of the co-evolution alliance is the least among the three hybrid modes. This result implies that a co-evolution alliance is the optimal governance mode when considering opportunism costs under high knowledge complexity, which supports hypothesis 3.



並列關鍵字

Co-evolution Alliance ； Strategic Alliance ； Alliance Portfolio ； Knowledge Complexity ； Scope Economy of Knowledge ； Opportunism Cost

參考文獻

Riordan, M. H., & Williamson, O. E. 1985. Asset Specificity and Economic Organization. International Journal of Industrial Organization, 3(4): 365-378.

Abramowitz, M., & Stegun, I. A. 1970. Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables

Aiello, F., & Cardamone, P. 2009. R&D Spillovers and Firms’ Performance in Italy: Evidence from a Flexible Production Function. In G. Arbia, & B. H. Baltagi (Eds.), Spatial Econometrics - Methods and Applications: 143-167. Heidelberg, Germany: Physica-Verlag.

Alchian, A. 1963. Reliability of Progress Curves in Airframe Production. Econometrica, 31(4): 679-693.

Alchian, A. A., & Demsetz, H. 1972. Production, Information Costs, and Economic Organization. The American Economic Review, 62(5): 777-795.