A critical examination of collusion, the pricing behaviour of a multi-product cartel and the cartel enforcement record in South Africa

Muzata, Tapera Gilbert (2021-03)

Thesis (PhD)--Stellenbosch University, 2021.

Thesis

ENGLISH SUMMARY : In this dissertation three research questions relating to collusion and cartel enforcement in South Africa were examined. The first question entailed examining the characteristics of detected cartels, together with the institutional features of selected key South African cartels. The author found that South African cartels incorporated some of the institutional features reported in the literature, including compensation schemes, joint ventures, sub-contracting arrangements, and entry or expansion deterrence strategies. In some selected key cartels, firms participated in collusion at two successive levels of the value chain, giving the cartels greater control over pricing throughout the value chain. Communication and monitoring were found to often involve a mix of various forms, including firm-level mechanisms (notably sales infrastructure) that complemented the centralised communications mechanisms discussed in the literature. To help fully explain collusion under conditions of imperfect information, the theory should account for the complementarities among various forms of communication and monitoring used by cartels. In addition, the author found trade policy to be an inexpensive tool used by some cartels to weaken threats from imports. The second question concerned the pricing dynamics of a cartel involving multi-product firms and where the cartel faces periods of instability, producing distinct collusive phases. Like single product cartels, a multi-product cartel raises prices above competitive levels, but to varying degrees on different products. Cartel overcharges also vary over collusive phases, influenced by the demand and supply conditions in each phase. This suggests that a multi-product cartel maximises profits by imposing overcharges that vary by product and over collusive phases in response to changing market conditions. The assumptions about the nature of the transition between collusion and competition affects overcharge estimates. This dissertation provides arguments that penalties and damages estimates, reflecting overcharges, should consider product-level and phase-specific overcharges, rather than relying on averages. Finally, the dissertation examined the cartel enforcement record from a deterrence perspective, focusing on the drivers of cartel enforcement, the duration of cases from initiation to final decision, and the subsequent impact on the deterrence-effect of penalties. Leniency, settlements and penalties, supported by increased funding for the Competition Commission are considered to be the main drivers of cartel enforcement in South Africa. Contrary to expectation, the author found that these have not reduced the duration of cases. Instead, case duration increased progressively over the study period. Delays in penalising firms have resulted in firms paying significantly discounted penalties, weakening the deterrence effect of penalties. To preserve the deterrence effect of penalties, the author argues that an optimal cartel enforcement policy should account for these delays and should focus on higher present-value penalties for those firms that delay finalising cases. This dissertation, focusing on South Africa, contributes to the body of empirical literature on collusion and cartel enforcement. It provides suggestions for further advances in the theory on collusion under imperfect information and on overcharge estimation when dealing with multi-product and multi-period collusion. The dissertation also makes policy contributions that could enhance the efficacy of cartel enforcement in South Africa and in other jurisdictions.

AFRIKAANSE OPSOMMING : Hierdie proefskrif ondersoek drie vrae wat met samespanning en die toepassing van kartelwetgewinig in Suid-Afrika verband hou. Die eerste vraag behels die eienskappe van kartelle wat reeds ontbloot is, tesame met die institusionele kenmerke van uitgesoekte kartelle. Die proefskrif bevind dat Suid-Afrikaanse kartelle sommige van die institusionele eienskappe in die literatuur weerspieël, insluitend vergoedingskemas, gesamentlike ondernemings, subkontrakteringsooreenkomste, en strategieë om toetrede of uitbreiding te voorkom. In sommige sleutelgevalle word bevind dat firmas op twee opeenvolgende vlakke van ’n waardeketting saamspan, ten einde beter beheer oor prys te bekom. Die proefskrif bevind dat kommunikasie en monitering dikwels ’n kombinasie van vorme behels, insluitend firma-vlak meganismes (insluitend verkoopsinfrastruktuur) wat gesentraliseerde kommunikasie-meganismes, soos in die literatuur vervat, komplementeer. Die proefskrif bevind dus dat teorie samespanning onder toestande van onvolledige inligting beter kan beskryf indien dit vir hierdie komplementariteite voorsiening maak. Voorts bevind die proefskrif dat sommige kartelle handelsbeleid as ’n goedkoop afweermetode gebruik om die bedreiging van invoere die hoof te bied. Die tweede vraag hou met die prysdinamika van ’n kartel verband, waar multi-produk-firmas saamspan en waar die samespanning periodies onderbreek word. Soos vir samespanninig rondom ’n enkele produk, bevind die proefskrif dat ’n multi-produk-kartel pryse bo die mededingende vlak verhoog, maar in wisselende mate na gelang van die bepaalde produk. Voorts bevind die proefskrif dat pryse oor verskillende fases van samespanning wissel, na gelang van veranderende markomstandighede. Aannames omtrent die aard van die oorgang tussen fases van samespanning en van mededinging, en die impak daarvan op skattings van kartelskade, word ook ondersoek. Die proefskrif betoog dus vir kartelboetes en skadebepaling om op produk-vlak en fase-spesifieke-ontledings, eerder as gemiddeldes, staat te maak. Laastens ondersoek die proefskrif die karteltoepassingsrekord vanuit ’n ontmoedigingsoogpunt. Die soeklig val op die drie plekke, naamlik die vernaamste drywers van karteltoepassing, die duur van kartelsake vanaf inisiasie tot finale besluit, en die gevolglike impak van hierdie tydsduur op kartelboetes. Aansoeke om toegeeflikheid, skikkingsooreenkomste en boetes – sowel as beter befondsing – word as die vernaamste drywers van karteltoepassing beskou. Teen verwagting bevind die proefskrif dat hierdie drywers nie die duur van kartelsake verminder het nie. Trouens, die duur van kartelsake, oor die studieperiode, het stelselmatig toegeneem. Die proefskrif bevind voorts dat vertragings in die penalisering van firmas die voorkomende effek van boetes verlaag. Ten einde die impak van boetes te verbeter behoort karteltoepassing rekening met die kontantwaarde van boetes te hou, veral waar firmas die finalisering van kartelsake vertraag. Hierdie proefskrif, met ’n fokus op Suid-Afrika, lewer dus bydraes tot die empiriese literatuur rondom samespanning en karteltoepassing. Die proefskrif doen voorstelle aan die hand vir verdere ontwikkeling van die teorie van samespanning onder onvolledige inligting en die skatting van kartelskade waar dit multi-produk- en multi-periode-samespanning aangaan. Die proefskrif doen ook voorstelle aan die hand om die doeltreffendheid van kartelbeleid in Suid-Afrika en in ander jurisdiksies te verhoog.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/110135
This item appears in the following collections: