Much is made of economists’ inability to agree on anything. Yet, global antitrust regulation shows that convergence in antitrust economics is recent, remarkable, and widespread. In the courtroom, however, there is still room for genuine professional disagreement among economic experts.
For counsel, judges, and jurors interested in understanding how experts with often similar views can reach divergent conclusions, it is important to focus on the root causes of these disagreements.
Economists tasked with evaluating allegedly collusive or monopolistic conduct may provide expert insights into how a market would have evolved “but for” particular behaviors, based on shared economic beliefs that can often seem as natural as the laws of physics, including:
To identify anticompetitive behavior, economists typically rely on economic theory (the mathematical or graphical expression of underlying beliefs) and statistical analysis (data analysis and the construction of alternative, “but-for” scenarios). In antitrust litigation, economists often apply these tools to evaluate a coherent set of issues:
Despite consensus around behavioral assumptions and methodologies, attorneys, judges, and juries are often confronted with dueling economic opinions that can result in polar-opposite conclusions. This is only rarely related to the application of “unprofessional” economic arguments. More often, it is the result of subtle differences in the application of shared economic theories and statistical tools, the choice of data, or the consideration of external factors. These minor discrepancies in economists’ understanding of the basic building blocks of the economic world can have significant consequences in the courtroom.
Consider the litigation surrounding Best Buy in the matter of alleged cartel activity among manufacturers of thin-film-transistor liquid-crystal display (TFT-LCD) panels used in televisions, laptop computers, and monitors. This lawsuit stemmed from a long-running U.S. investigation of an alleged conspiracy to set artificially high prices for, and restrict the supply of, various sizes of TFT-LCD panels. The plaintiffs claimed that since at least 1998 the defendants had engaged in price-fixing behaviors, resulting in overcharges to direct purchasers of TFT-LCD panels and finished products containing those panels and to indirect purchasers of finished products.
In this litigation, two well-respected economic experts measured the alleged overcharges and proposed a statistical analysis to assess the effect of the conspiracy. Both agreed on the importance of various measures of costs and demand in determining prices, regardless of a conspiracy. In fact, the experts seemed to disagree primarily on whether the Producer Price Index (PPI) – a measure of price trends in wholesale markets – for microprocessors was an appropriate proxy for costs. This difference of opinion resulted in damages estimates for direct purchases that varied by more than $200 million. The plaintiff’s expert included the PPI for microprocessors in his analysis and found an overcharge resulting in damages of more than $230 million for direct purchases; the defendants’ expert used PPIs for other LCD panel inputs to show that the overcharge was close to zero.
Given the complexity of the analytics on both sides, what was counsel – or a jury – to do? In this matter, the statistical analysis was rigorous and differed only in subtle ways between the two parties. Ultimately, deciding on the right answer was a matter of belief on one key question: Was the PPI for CPUs a good proxy for the cost of building TFT-LCD panels? In other words, was the CPU essential to an analysis of TFT-LCD panel prices?
Ultimately, the California federal jury said “No” and awarded only $7.4 million in direct damages, based on figures calculated and presented by the defendant’s testifying experts. The Best Buy plaintiffs had asked for $770 million in total damages based on the plaintiffs’ economic experts’ analysis, and $230 million for direct purchases alone. The jury also found the defendant, Toshiba Corporation, not liable for conspiracy to fix the prices of TFT-LCD panels.
In this case, and others like it, top economic experts took similar roads and reached vastly different endpoints. Analyzing differences in statistical specification may seem challenging for both counsel and the jury, but understanding economists’ beliefs about human behavior, organizational activities, firm characteristics, market dynamics, and, in this case, determinants of costs is likely to be more intuitive and, ultimately, more helpful in assisting the trier of fact to evaluate the relative merits of two opposing claims. ■
Pierre Cremieux is President of Analysis Group and Aaron Yeater is a Managing Principal in the Boston office.
From Analysis Group Forum: 2015 Year in Review