Anti-Trust & Mergers: A Couple of Uninformed Views

Suppose your company is considering purchasing another company and wishes to convince the Department of Justice (DOJ) that the benefits to shareholders come from efficiency-enhancing effects and not competition-reducing effects. One could hire a team of economists from a host of consulting firms such as these two. Or supposed you simply wished to learned how these issues have been addressed more and more by good economic analysis. You might start with the various articles collected by John E. Kwoka and Lawrence J. White. As Brad DeLong recently noted, the National Review writers often opine on matters that they don’t take the time to understand. Forgive me for the following long post, but the attack on economists and the DOJ from this crowd has my dander up.

A little over a year ago, Judge Walker ruled in favor of Oracle’s bid to acquire PeopleSoft. Stuart Anderson decides now to attack the DOJ and Kenneth G. Elzinga:

While the Soviet Union is a distant memory, the notion that a small group of people can figure out what is best for a nation’s economy is alive and well in an unlikely place — the Antitrust Division of the U.S. Department of Justice (DOJ) … A year can be a lifetime in the world of software and high technology. Yet based on the preferences of lawyers in the DOJ’s antitrust division, companies may be forced to wait years before moving ahead with an acquisition. Every purchase or merger involving real money prompts a review by antitrust enforcers, whose power is immense. Under the current procedures, companies must first submit documents explaining an acquisition and implicitly answering any possible objections from the DOJ. If the antitrust division is not satisfied in the “first round,” it can ask for more information. To comply with this second, more expansive request is generally considered so time-consuming and disruptive that most companies withdraw an acquisition at this stage rather than endure the costly process, according to industry sources … The policy question for Republicans is this: Why after so roundly criticizing the Clinton administration for its antitrust policies – including the case against Microsoft – has the Bush administration allowed career lawyers at the DOJ to promulgate essentially the same policies? In attempting to block Oracle’s acquisition of PeopleSoft, SEC filings show that the DOJ ended up costing Oracle close to $100 million, not including the indirect costs of forcing the chairman, president, and other executives to spend weeks in depositions and, more generally, distracting them from their primary duties of running a major U.S. corporation. The judge in the case disclosed that the DOJ paid one of its expert witnesses, an economist, over $1 million. No doubt it would have been front-page news if this economist had worked for Halliburton. Ultimately, in an embarrassing rebuke, the judge ruled squarely against the DOJ, but much of the damage had already been done.

In his attack on Dr. Elzinga and DOJ, Anderson displays little to no understanding of the issues involved in this litigation. Shearman & Sterling’s Antitrust group provides a nice summary of the Court’s reasoning:

In particular, Judge Walker rejected DOJ’s product market definition that the relevant market consisted only of high function (sales in excess of $500,000) financial and human resource management enterprise resource planning software (“ERP”), sold only by Oracle, Peoplesoft and SAP. Judge Walker found instead that Microsoft and Lawson participated in the market, as well as mid-market vendors, best of breed solution companies, and other companies providing outsourcing. In addition, Judge Walker rejected DOJ’s geographic market definition, finding the market for business software was a global one, and not limited to the United States. Consequently, Judge Walker concluded that DOJ had failed to establish that the merged firm could exercise market power and thus, that the merger would substantially lessen competition.

Elzinga had originally argued that the relevant definition of the market included U.S. providers of high-function financial and human-resources enterprise applications software, which was essentially a duopoly before the merger of Oracle and PeopleSoft. Jerry Hausman of MIT testified on behalf of Oracle and noted that SAP of Germany was also part of the world market for such software. Oracle also argued that the market should include other entities such as Lawson Software, which the Court saw as providing mid-market products.

Clearly, the definition of the relevant market is central to whether a proposed merger will reduce the degree of competition. Alas, the National Review defines “damage” in terms of not being able to extract wealth from the stock market with little concern about the possible damage to consumers or overall economic efficiency if a merger does create market power.

But Anderson went further in his insulting the economists who worked on this litigation by questioning why experts were paid in the first place. Compare the resumes of Dr. Hausman and Dr. Elzinga to the guru responsible for this:

The antitrust division of the DOJ continues to be guided by poor precedent. In 1911, the U.S. government separated Standard Oil into 33 entities — not because the company raised prices but because a judge decided it had concentrated various parts of the production process under one umbrella. Even before the case was settled, Standard Oil’s market share fell from 90 percent to 60 percent. The case of Alcoa Aluminum in 1945 was “equally absurd,” notes economist Jim Cox. Alcoa was the dominant producer of primary aluminum in the United States. Yet similar to the Standard Oil case, there is no evidence that the company either restricted its output or raised prices. Yet here, too, the decision went against Alcoa. These decisions reflect the mindset by which DOJ antitrust lawyers today evaluate acquisitions that cross their desks. Jim Cox points out that in the case of Alcoa, a judge had to restrict the definition of the market to “primary” aluminum, ignoring the competition represented by reprocessed aluminum. In the Standard Oil case, the judge did not address or take seriously imported oil as a competitive factor. “In other words, the courts had to first artificially narrow the market in order to find these companies guilty,” says Cox. Narrowing the market and ignoring the potential entry of new competitors are common methods used by the DOJ to block acquisitions.

Not to go after Jim Cox, but I would more faith in Dr. Hausman or Dr. Elzinga on these issues. Anderson also seems to be upset that the economists who provided testimony in this litigation were well compensated. Economists who are well trained in economics put a lot of effort and their own reputations on the line when they testify for clients – whether the client is the government or the corporations who wish to merge. Oracle was well served by Dr. Hausman.

But let’s go back a couple of years for perhaps the dumbest op-ed written by Donald Luskin. Here are my favorite parts of his attack on the DOJ’s anti-trust division in regards a proposed merger between Dreyer’s and Nestle:

Let’s take a look at what was behind the FTC’s decision – a decision which inserts the power of government to disrupt a voluntary and mutually sought combination of two private businesses, a decision that slashed a billion dollars in market value from the stock of Dreyer’s last night as soon as it was announced. The essence of it as that the FTC believes that “that the elimination of Dreyer’s would likely lead to anticompetitive effects in the market for superpremium ice cream.” Did you know, before this, that there even was something called “the market for superpremium ice cream”? Well, there is now. Imagine some incredibly complex diagram covering the wall in the office of some Ph. D. at the FTC’s offices in Washington showing “the market for food.” Now erase everything that isn’t “the market for deserts,” and then erase everything that isn’t “the market for frozen deserts,” and then erase everything that isn’t “the market for ice cream.” In that tiny little sub-sub-sub-sub-market, Dreyers’ brands Dreamery, Godiva and Starbucks battle it out with Nestlé’s Häagen Dazs, and Unilever’s Ben & Jerry’s. The big issue, according to the FTC — the reason for government to intervene and cost Dreyer’s shareholders a billion dollars — is that “this deal will reduce the number of significant competitors from three to two” and “would likely raise prices and reduce choice for consumers.” But, even granting that the government ought to be concerned with such matters at all, none of it means a thing in this case unless you accept the idea of “the market for superpremium ice cream” as a meaningful reality. Who’s to say that’s a market of any importance? Who’s to say that market needs the government to interfere with private business decisions to be sure that there’s a certain amount of competition in it — or any competition at all? Let’s say that there were so little competition in “the market for superpremium ice cream” that choice was narrowed to a single brand and prices became astronomical. What if “the market for superpremium ice cream” vanished from the face of the earth altogether? So what? Consumers could simply buy any of dozens of premium, regular, or cheap-o ice cream brands instead — or some other desert. Let them eat superpremium cake! Even if we grant that “the market for superpremium ice cream” needs to be policed for competition, who’s to say that the correct number of competitors is two rather than three? Is it always better to have more competitors in a market? In this country we have only two political parties of any influence — would we be better off with more, like Italy or Israel? Maybe we would, in which case we are free to have them – or not. But in the case of ice cream — or rather, superpremium ice cream – we are not free to have less than three. And why should we necessarily be concerned that prices might rise with a drop in the number of competitors? First, there’s no axiomatic reason to be absolutely positive that prices would rise in the first place. And what’s the correct price for superpremium ice cream, anyway? Maybe prices should go up – maybe today’s prices are too low. To insist arbitrarily that lower prices are always better – and to bring government power to bear to make sure that they are – amounts to a form of price controls in the name of antitrust. It takes what should be a free bargaining process between producer and consumer and stacks it in favor of the consumer. But why are people who make ice cream less entitled to equal protection under the laws than people who eat ice cream?

Where to begin with such insanity? First of all, I love superpremium ice cream and pound cake is not a perfect substitute. So much for understanding what the relevant market is. Secondly, Luskin shows his hand as far as his lack of knowledge about basic economics and well is his desire to make money off the stock market even at the expense of economic efficiency (never mind that those who take his investment advice typically lose money as in “they are poor because he is stupid”).

It seems neither Anderson nor Luskin seem to even understand the process, which allows for the corporations to hire economists to make their case that the merger is efficiency increasing rather than competition reducing. Of course, neither Anderson nor Luskin have an integrity nor expertise on anything, which is why they are asked to write for rags such as the National Review and why your company would be ill-advised to hire them as your economic experts.