Policy

Why antitrust regulators underestimate competition

By David Jon Phillips     
March 1, 2016

From: Magazine

Photo by Associated Press.

When evaluating a potential merger or bailout, antitrust regulators consider the competitive landscape. If they see an industry that’s expensive to enter, or one where one or two companies will dominate segments of the market, regulators may be more likely to block the merger or bail out the troubled company.

The problem is that regulators use economic models that tend to underestimate the level of competition that would result from a merger or bankruptcy, argues Chicago Booth’s Thomas Wollmann, who says those models overlook the fact that an existing company could shake up a market by introducing new products.

“Traditional, workhorse economic models tend to ignore this fact,” he says.

Wollmann considered the US government’s $85 billion rescue of General Motors and Chrysler in 2009. “The policy they chose was to bail them out,” says Wollmann. “The alternatives were to do nothing—which probably would have resulted in at least one of them being liquidated—or to facilitate a merger with an existing firm in the market.”

To illustrate how new products could have affected the need for a bailout, Wollmann focused on the commercial-vehicle market, which represents roughly 5–10 percent of the $400 billion US automotive industry. In 2009, some segments of the commercial-vehicle market only included offerings by GM, Chrysler, Ford, and International. If two of those companies were to have failed, the remaining two firms would have had much more market power in those segments. It’s understandable that policy makers would have worried about that possibility: the commercial-vehicle market had seen virtually no companies enter or leave in close to three decades, so it seemed unlikely that one would choose to enter in the wake of General Motors’ and Chrysler’s collapse.

The traditional approach to evaluating the competitive landscape was developed in 1995 by Steven T. Berry and James A. Levinsohn of Yale, and Harvard’s Ariel Pakes (who was Wollmann’s PhD advisor). That model allows for prices to adjust after a company leaves a market, but does not predict whether another company would enter or exit. According to this model, if General Motors and Chrysler had been liquidated, companies that remained would have faced less competition and could have raised markups significantly—markup increases could have exceeded 70 percent on the most-affected products, such as flatbed construction trucks, according to Wollmann’s calculations. The remaining companies would have expanded their market shares, although total sales of commercial vehicles would have fallen.

But according to Wollmann, the traditional approach doesn’t consider that if GM and Chrysler were to have exited, other commercial-vehicle producers could have adjusted their product offerings and launched new models. Volvo and PACCAR, for example, manufactured commercial vehicles in the United States but tended to offer big trucks. They did not offer flatbed construction trucks as GM and Chrysler did. But would they have, if GM and Chrysler had disappeared? “This comes down to a decision for Volvo that weighs their sunk costs of entering against the added profits of doing so,” says Wollmann.

He finds that the sunk costs involved with introducing a new product model were low enough that, had GM and Chrysler liquidated, rival firms would have entered the market and had a dramatic effect on prices. He calculates that when a model allows for this kind of product entry, markups on the most affected products show a rise of only 10–20 percent, which is in sharp contrast to the 70 percent increase predicted from models that ignore entry.

In fact, he finds that when entry is considered, all three policy options he tests—bailing out GM and Chrysler, letting them fail, or facilitating a merger—“result in more-or-less similar effects to the market outcomes. The reason is straightforward: it is precisely those policies that would otherwise force markups to go up the most where entry goes to work in bringing them back down.”

While his study focuses on automakers, Wollmann says the findings also apply to other industries with high costs of entry and a few big incumbents, whether they make high-end watches, beer, or aircraft. For example, London-based SABMiller and Belgium’s AB InBev have sought to merge. If the merger were to go through, the combined company could raise prices. But then again, those higher prices could induce other rivals to enter and drive prices back down. In the beer market, Wollmann says, the question regulators should focus on is “whether any midsize breweries have the ability to expand production and marketing capacity and enter the mass-market segment.”

Steven T. Berry, James A. Levinsohn, and Ariel Pakes, “Automobile Prices in Market Equilibrium,” Econometrica, July 1995.

Thomas Wollmann, “Trucks without Bailouts: Equilibrium Product Characteristics for Commercial Vehicles,” Working paper, November

More from the Magazine

An error has occurred

An error has occurred