Antitrust And The Schumpeterian Dynamic In Erp Software

Lynne Kiesling

Today the Wall Street Journal had an editorial (subscription required) commenting on the antitrust policy implications of last week’s District Court decision to allow Oracle to proceed with its hostile takeover offer for PeopleSoft. Their observations focus on a crucial issue in antitrust policy, one that antitrust economists have been struggling with for a long time – the inherently static nature of the tools used to diagnose and measure potential anticompetitive implications. In particular, the WSJ editorial notes the shortcomings of the Herfindahl index, a measure based on the sum of the squared market shares of the firms in the industry. A heavily concentrated industry at any point in time will have a higher Herfindahl index to the extent that there are dominant firms in the industry that themselves have high market shares (the squaring of the market shares serves to emphasize high market shares and market share differentials).

Instead of now appealing Judge Walker’s ruling, Mr. Pate [Assistant Attorney General] would be better off asking his lawyers to rethink their Herfindahl index and other static measures of competition that are outdated in a world of rapid technological change. Let’s hope the Walker ruling also instructs the antitrust barons of the European Union, which is also scrutinizing any Oracle-PeopleSoft deal.

Indeed, the judge’s opinion in the decision reveals that he thought about the dynamic implications of interfering in the Oracle action. He discussed the heterogeneity of software providers, and the potential disciplining value that niche providers can bring even in a market in which there are three “dominant firms”. Even if you take at face value the market definition of ERP software, with Oracle and PeopleSoft as two of the big three firms, the calculated Herfindahl index is pretty grim. But Judge Walker correctly points out that market definition is the lynchpin in the antitrust argument for opposing the takeover attempt, and moreover, that market definition is a highly dynamic and moving target, particularly in enterprise software.

This is a very interesting issue. The judge’s decision leaves open the possibility for the PeopleSoft shareholders to assess for themselves whether such a merger would provide value creation for customers, thereby increasing the value to shareholders themselves. Some people worry that if Larry Ellison succeeds in folding PeopleSoft into Oracle, customers will have fewer choices in ERP software, enabling Oracle, SAP, and even Siebel and Microsoft to raise prices on their ERP offerings. That is one way to create shareholder value.

But it’s a short-run way, and focusing on it has a high opportunity cost by cutting off potentially beneficial consolidation in a bloated part of an industry that has already shown its inclination to opportunistically exploit any short-run strategies of competitors.

Put another way, if the PeopleSoft shareholders agree to the takeover on the basis of their belief that it will raise prices and therefore revenues and profits, the more fool they. They will get steamrolled. The shareholders should agree to the takeover if they see complementarities and cost savings. Other software companies are already in the space, or could easily move into the space, that Oracle and PeopleSoft serve. Moreover, the space that they serve, enterprise resource planning, has itself morphed a few times over the past decade, and the companies that have succeeded are the ones that are nimble. Existing competitors, potential competitors, the heterogeneity of customer needs, and potential changes in the market itself and the needs of enterprise customers all compose a messy soup of disciplines on the ability of a large Oracle to succeed in any attempt to raise prices via consolidation. As the WSJ editorial put it,

The virtue of Judge Walker’s opinion is that he took into consideration the dynamic, highly competitive nature of today’s technology businesses. For one thing, those markets are global, especially the competition to serve large multinational customers. For another, the Justice complaint ignored the many other software suppliers that could enter the high-end business applications market if Oracle began to raise prices after any merger. One of those new entrants might well be that tiny little outfit known as Microsoft, which bought European software company Navison in 2002.

Judge Walker has shown himself to be Schumpterian, staking a claim for the prospect of “creative” in the “creative destruction” that undergirds all healthy, thriving, dynamic markets. In so doing he has done something extremely valuable, which the WSJ editorial also notes – he has preserved the opportunity for PeopleSoft shareholders to choose, based on their diffuse knowledge, experience, and objectives, whether or not this is a merger that makes sense, and if so on what terms. Why need the Department of Justice intermediate when such a variety of disciplining dynamics already exist?

Analysts following the case think that this was the right decision, as seen in this Info World article:

One antitrust attorney following the case, Paul Friedman of Dechert LLP, said the DOJ would face long odds on an appeal, thanks to Judge Walker’s apparent meticulousness in mapping out his findings of fact. Walker’s decision ran to 164 pages.

“Judge Walker took as much time as he did in delivering the opinion because it is very grounded in his review of the evidence,” Friedman said. “The appeals court generally gives great deference to the lower court in findings of fact. Unless he made a really serious mistake, and it would surprise me if he did, the government may conclude that it can’t appeal.”

Those who watched the case unfold say the decision was unsurprising. Though the DOJ initially appeared to have a strong case for its limited definition of the high-end market for financial and human resources applications, observers say Oracle’s legal team effectively shredded that argument and demonstrated that the enterprise applications market is fragmented and highly competitive.

“Everything I saw about the case made this the only reasonable judgement,” said independent analyst Josh Greenbaum, who runs Enterprise Applications Consulting Inc. “I think justice has definitely prevailed.”