- 學位論文
Three Essays on the Economics of Contracts
Three Essays on the Economics of Contracts
摘要
本論文由廠商於不對稱訊息下所面臨的三個不同的契約問題所組成，分別標題為: ”Why do Amusement Parks only Charge a Fixed Admission Fee?”, “Will a Small Firm Reject Good Agents More Often than a Large Firm?”, “Does a Nonverifiable Imperfect Informative Binary Signal Always Create a Strictly Positive Value?” 從簡單的觀察中，時下主題遊樂園並不如個體理論所預測，採取複雜的非線性訂價(nonlinear pricing)，而多採用簡單的一票到底的訂價方式？文獻中多採取因使用非線性訂價會產生交易成本來做為解釋廠商選擇一票到底的理由，但本論文第一部分認為即便沒有交易成本，單就非線性訂價的考量，主題樂園已有充分的理由採取一票到底的訂價方式。當消費者入園前須負擔運輸成本越高、廠商遊樂設施所帶來的歡樂氣氛越大、營運成本越低，越易採用一票到底的簡單訂價模式。 第二部份主要針對公司規模與員工薪資呈現正向關係，提出新的模型與解釋。由於小公司於不對稱訊息的環境中較無法有效的消弭因countervailing incentive所造成的訊息租(information rent)，將會比大公司易於拒絕僱用有能力的員工，不同公司規模雇用的員工素質將有所差別，因此公司規模與員工薪資將呈現正向關係。 第三部份則對於富含訊息(informativeness)的訊號(signal)是否對於面對不對稱訊息的一方永遠帶來價值，提出初步的理論探討。當訊號擁有的訊息並不完全時，由於不一定改變不對稱訊息一方生產效率與訊息租取捨的方式，因此有著「富含訊息的訊號不一定永遠帶來價值」的結論。
關鍵字
契約理論並列摘要
My dissertation is a collection of three essays which each considers a different contracting problem facing a firm under asymmetric information. In the first essay, I propose a model of asymmetric information to delineate an amusement park’s pricing plan. In the real world, it is common to observe that many amusement parks adopt pass pricing (i.e., charging only a fixed admission fee). One might wonder why an amusement park does not adopt non-linear pricing or two-part tariffs (to improve its profit) as theories of non-linear pricing or two-part tariffs claim. To justify the real world observation, some recent literature argues that, in the presence of nonzero additional transaction (monitor) costs for implementing non-linear pricing schemes, charging only a fixed admission fee can be more profitable than other pricing strategies. As such transaction costs have been drastically reduced due to rapid technology progress, there might exist some other reasons causing an amusement park to adopt pass pricing. We come up with some explanations on why amusement parks adopt pass pricing even when no additional transaction costs exist for implementing nonlinear pricing schemes: The related factors include much fun generated by any ride and a higher transportation cost. The second essay contributes some new explanations for the positive relationship between firm size and wages. It is theoretically claimed and shown that a small firm may reject good (efficient) type agents more often (than a large firm) under asymmetric information. The condition that only the different sizes of endowments of a capital-like resource draw a distinction between a large firm and a small firm is assumed. As the difference between both types’ status quo utility levels is growing, a small firm will reject good type agents more often. The reason especially is not that the surplus generated by a small firm and a good type agent can not afford the good type’s status quo utility level, but the capital-like resource restriction interferes with the ability of a small firm to mitigate the information rent caused by the countervailing incentive (caused by a big difference between both types’ status quo utility levels). The combined impact of the capital-like resource restriction and the countervailing incentive on the (employment) decision of whether to reject to delegate a task to good type agents or not is brought into focus. Moreover, while the countervailing incentive exits and when both the large firm and the small firm decide to delegate the task to both types, good agents are more productive in large firms and therefore paid higher wages. These findings support the labor quality explanation in the economic literature that larger firms employ higher-quality workers, and the productivity hypothesis that workers are more productive in large firms and therefore ask for higher wages, but in a kind of markedly different reasoning. The third essay relates to the value of information. Even ignoring the costs for acquiring and adopting the signal, does an (ex ante) nonverifiable imperfect informative binary signal always create a strictly positive value for a firm under asymmetric information? Because of a difference between both agent types’ status quo utility levels, the firm adopting a nonverifiable imperfect informative binary signal is not guaranteed to choose a different way (from the way the firm not adopting the signal does) to optimally trade off output efficiency against rent extraction under asymmetric information; hence the signal does not always create a strictly positive value for the firm. It is unlike the well-known result about the ex post verifiable signal that, under mild assumptions, the agent receives no excess rent whatever his type, and moreover the complete information optimal output levels can be implemented; hence the signal does create a strictly positive value for the firm whenever the second-best contract (without adopting that signal) entails some output distortion or/and some excess information rent given up for some type(s).