A recent study we conducted as part of the Competition Commission’s work on the impact of anti-competitive behaviour illustrates what happens after cartels are busted, and has implications for the determination of penalties for firms engaged in collusion.
We examined the concrete products cartel uncovered by the commission towards the end of 2007 to understand how competition had developed. This had been a long-running cartel, established in the early 1970s, and only ended thanks to the commission’s corporate leniency policy, which led to Rocla (a subsidiary of Murray & Roberts) coming forward and telling all.
Following the commission’s focus on collusion in construction, Rocla took legal advice and contacted the commission, with its application for corporate leniency being filed on December 7, 2007.
Under the commission’s corporate leniency policy, a participant in a cartel is not subject to a penalty under the Competition Act in exchange for revealing the extent of the cartel and co-operating fully in prosecuting the remaining members.
It should be noted that this does not exempt the firm from any claims for damages. Thus, while other firms appear to believe that the cartel was ending of its own accord in late 2007, Rocla was ending its role after more than three decades.
The cartel was mainly focused on precast concrete pipes and culverts. These are products used in various construction applications such as road construction and earthworks, and are important for the government’s infrastructure development drive. According to Rocla, it and nine other firms had engaged in anti-competitive conduct involving market allocation, price fixing and collusive tendering. The cartel was tightly managed on the basis of very sophisticated rules contained in a document referred to as the “modus operandi”.
Cartel members agreed market shares along regional lines, along with the types of products each was allowed to produce. Market shares by product were allocated in defined areas in Johannesburg, Durban and Cape Town. Firms agreed that only Rocla would supply in the rest of the country.
For ease of implementation of the arrangement, cartel members agreed to charge similar prices and to increase these prices by the same percentage twice a year. There was a system of allocating contracts, where firms agreed who would make the winning bid, while others bid higher to give the impression of competition.
Regular meetings were held in which tonnages were reported and contracts allocated. The meetings in Gauteng usually followed those of the industry association.
In 2010 the Competition Tribunal imposed administrative penalties on all the nine implicated firms who had admitted guilt. However, two of these firms, Southern Pipeline Contractors (SPC) and Conrite Walls lodged appeals at the Competition Appeal Court (CAC) contesting the level of the fines imposed on them (R16.9 million and R6.2m, respectively).
The CAC reduced the fines to R8.7m and R2m. SPC argued that the fine was unfair as its participation in the cartel was limited to Gauteng and only to the sale of concrete pipes. SPC argued that it did not make culverts during the cartel period due to high costs of setting up production and not because of the cartel arrangement.
Furthermore, SPC argued that it did not earn excessive profits from the cartel because its price increases between 2002 and 2007 were below cost increases.
The CAC concurred, stating that “there was very little evidence to suggest that there had been significant consumer losses pursuant to the first appellant’s (SPC’s) activity”. The CAC implied that the commission or tribunal should have linked the level of the SPC fine to the amount of consumer harm caused by the firm’s conduct.
Our study highlights the challenges that will be faced by such an approach. The picture that emerges contradicts the assertions made by SPC in the court hearing.
The study found that former cartel members, including SPC, are now supplying wider product ranges over wider areas. During the cartel period SPC and Cobro (around Durban) did not make culverts. Similarly, Concrete Units agreed not to supply concrete pipes in the Western Cape. These firms have all extended their range since the end of the cartel. SPC now supplies the whole product range that was covered by the cartel in a 150km radius around Johannesburg within which it agreed to stay under the cartel. Cobro is delivering in the northern parts of the Eastern Cape. Concrete Units, which was limited to the regions around Johannesburg and Cape Town, now reaches areas as far as Limpopo, Mpumalanga and the Free State on a regular basis, and has added concrete pipes to its Western Cape range.
It is also notable that since the demise of the cartel, five new players have entered various markets that were previously the reserve of the cartel. This points to the fact that the stability of any cartel lies in its ability to keep out new entrants.
Using pricing data from the Johannesburg and Durban regions, the study shows that the cartel caused substantial consumer harm by charging prices above the competitive level. Concrete pipe prices in both areas continued to increase for 18 months after the cartel was uncovered. However, from mid-2009 to June 2011, we estimate prices declined by 37 percent in Durban and 27 percent in Johannesburg, mainly due to the ending of the cartel. This is consistent with submissions by a former cartelist, who said price pressures resulted from increased competition rather than a lack of demand. Despite booms and slumps, concrete pipe prices had only ever increased.
We estimate the amount of overcharging by calculating the difference between the actual prices charged and the price that would have been charged (the counterfactual price). Our estimates indicate a 51 percent to 57 percent overcharge in the Durban area, and 16.5 percent to 28 percent for Johannesburg, which are very high by international comparison. Global surveys find that only 5 percent of cartels have resulted in overcharge estimates that exceed 50 percent, and most overcharge estimates are in the order of 15 percent to 25 percent.
The challenge in estimating a consumer overcharge is that such a calculation can only be done with a reasonable level of accuracy after prices have fallen towards competitive levels. But normally prices do not fall immediately after the uncovering of the cartel but rather take some time as the colluding firms intimately understand each other’s businesses and how to prevent competition.
In this case, prices only fell towards competitive levels some time after the cartel’s end. The main cartelists continued to share monthly volume data. It took time to make and implement the decisions that saw them expanding into previously reserved products and geographic markets.
The original penalty imposed on SPC by the tribunal – the maximum allowable penalty under the act – only covered at most 43 percent of SPC’s overcharge for six years. The penalty was too low if one assesses it against additional profits earned.
What does this mean? The Competition Act prohibits cartel conduct outright. Some cartel arrangements harm consumer welfare not only through high price mark-ups, but also through firms forgoing competing, creating an artificial market structure, as was the case with this cartel.
More work needs to be done to better understand the deterrent effect of cartel fines. Given the historical prevalence of cartel conduct in the economy, such work will only be done as more cartels are ended.
Junior Khumalo and Jeffrey Mashiane are economists at the Competition Commission and associates of the Centre for Competition Economics at the University of Johannesburg. They write in their personal capacity. The working paper can be downloaded from www.uj.ac.za/cce.