寬恕政策與卡特爾運作的實證分析

Abstract

廠商隨意操縱市場價格，會破壞市場效率，使社會剩餘減少，因此，對於這類的行為通常是不會被政府所允許的。但是，因為此類行為可以為廠商帶來相當可觀的獲利，因此相當多的廠商仍冒著可能的裁罰風險，試圖對市場價格進行操控。而在市場中，如非獨占市場而是如多家廠商寡佔的市場中，廠商通常會試圖組成卡特爾組織，聯合各家廠商共同操控市場。而這類的組織的組成，就會給社會帶來相當大的傷害，因此政府通常會去打擊卡特爾組織，而寬恕政策就是近年深受各國政府歡迎用以打擊卡特爾組織所採用的方法之一。因此，本文就特別研究寬恕政策與卡特爾組織之間的關聯，觀察哪些因素是會影響卡特爾組織申請寬恕政策，並進一步觀察，寬恕政策是如何影響卡特爾組織的存續時間。 本研究所選用的資料是由美國普渡大學所收集的卡特爾組織資料。此份資料所包含的資訊有如，是否有參加寬恕政策、卡特爾組織開始及結束時間以及所屬產業等。此外，本次研究也收集了世界銀行所整理的相關之總體經濟資料，其中包含經濟成長率、通貨膨脹率以及貿易依存度。而由以上資料，本次研究使用Probit模型分析哪些因素會影響卡特爾組織使用寬恕政策之機率，並透過不同時間長度或不同產業的子樣本分析來觀察其中的差異，而研究卡特爾組織的存續時間，則是透過工具變數(Instrumental Variables)法的二階最小平方法(Two-stage least-squares)進行分析，並以分量迴歸(Quantile regression)模型對觀察不同存續時間長度下，政策對於卡特爾組織的影響是否有差異。 而在本次的研究結果中，本文發現整體而言，寬恕政策是一個很好打擊卡特爾組織的政策工具，寬恕政策可以有效的使卡特爾組織存續的時間顯著下降，而在子樣本分析中，則發現寬恕政策對於三級產業的效果優於二級產業，政策對於二級產業中的卡特爾組織，雖亦會使其存續時間下降，但效果並不顯著。

Manufacturers arbitrarily manipulating market prices will destroy market efficiency and reduce the social surplus, so such behaviors are usually not allowed by the government, but because such behaviors can bring considerable profits to manufacturers, there are quite a few Manufacturers still risk possible penalties to manipulate market prices. In the market, if the market is not monopolized but is oligopolies by multiple manufacturers, manufacturers usually try to form a cartel organization to jointly control the market with various manufacturers, and the composition of such an organization will bring serious harm to the society, so the government usually cracks down on cartels, and the leniency policy is one of the methods that governments have welcomed in recent years to crack down on cartels. Therefore, this paper specifically studies the relationship between leniency policy and cartel organization, observes which factors will affect cartel organization's application for leniency policy, and further observes how leniency policy affects cartel organization's existence time. The data selected for this study are cartel organization data collected by Purdue University in the United States. This data contains information such as whether to participate in the leniency policy, the start and end date of the cartel organization, and the industry to which the cartel organization belongs. In addition, this research also collects relevant Macroeconomic data compiled by the World Bank, including economic growth rate, inflation rate, and dependence upon trade. Based on the above data, this study uses the Probit model to analyze which factors will affect the probability of cartel organizations using leniency policies, and at the meantime observes the differences through sub-sample analysis of different duration and different industries. The duration is analyzed by the Two-stage least-squares method of the Instrumental Variables method, and the Quantile regression model is used to observe whether the effect of policies on cartel organizations under different durations is different or not. In the results of this study, this paper finds that leniency policy is a good policy tool to combat cartel organizations. leniency policy can effectively reduce the duration of cartel organization. In the sub-sample analysis, it is found that the effectiveness of the leniency policy in the tertiary industry is better than that of the secondary sector, although the leniency policy still reduces the duration of the cartel organization in the secondary sector the effect is not significant.