We need retrospective analysis of Competition Authority decisions

What is the point of merger control? In 11 years, and over 600 deals, Irish regulators have intervened to block four. Some are now asking if this is too few. This is the wrong question.

To be sure, the Competition Authority will block suspect deals if appropriate. Under a 2002 law, it has self-executing power to enjoin transactions that substantially lessened competition.

But few mergers harm competition. At EU level, in 23 years and over 5,000 deals, 24 have been blocked. Ryanair's bid for Aer Lingus was the last one. US outcomes are similar. In the lingua franca of economists, a deal is harmful solely if it "creates or enhances the merged firm's ability or incentives to exercise market power either unilaterally or through coordination with rivals".

The question is what does this mean? Is it gobbledygook – what one critic described as “bureaucratic discretion masquerading as law enforcement”? Or is it a clinical and apolitical assessment, as those (regulators and practitioners) who make a living from it say?


Certainly nobody suggests it is a brightline test. Just issued Competition Authority guidelines say “there are no standard measures of competitive effects”. Every case must be assessed “on its merits and in its own particular circumstances”.

Some also charge that merger control, no matter the substantive test, involves non-economic considerations. Opposing the 2002 merger control law, Pat Rabbitte TD rightly argued “there is, whether the technocrats like it or not, a political and public interest dimension to the issue”.

Take the political consideration first.

A merger or takeover can raise fears of local plant closures and job losses. Or industry consolidation may be championed by government to allow Irish firms the scale to compete on worldwide markets, even if local markets are monopolised.

By law, the Competition Authority must be blind to such considerations. And in practice it is.

There are many examples. In 2007, the Competition Authority required divestment of an overlapping business, in the face of accusation and blame for threatened closure of Erin Foods’ Thurles plant and loss of 95 jobs. On-going consolidation in dairy processing continues to receive close agency scrutiny, though government policy envisages creation of one or two super-dairies.

What about its substantive competition law assessment? By its own apolitical standards, has the Competition Authority generally called it right?

To use industry labels, bad decisions are either “type one” or “type two” errors: blocking a pro-competitive or neutral deal is type one, while allowing an anti-competitive deal is type two. Type two errors are difficult to scope for. Upwards of 595 deals have been cleared since 2003 (a few with mandated divestment or other conditions). Should any of these have been blocked as anti-competitive?

Some economists have criticised the Competition Authority’s analysis in a number of clearance decisions. But economists, like lawyers, are often involved in advising parties proposing or opposing mergers, which could skew their views. In reality, no detailed retrospective study has been conducted that would show anti-competitive deals have been approved.

Nor has there been serious retrospective analysis of Competition Authority prohibition decisions for "type one" errors.

Positive outcome
In 2004, the Competition Authority blocked IBM's acquisition of Schlumberger. In the agency's view, the deal would have allowed a "merger to monopoly" in business continuity services. Following the Competition Authority's intervention, Schlumberger – which was intent on exiting the Irish market – sold its Irish business to Hewlett Packard, which competes with IBM to this day. This suggests a positive outcome for consumers.

In 2008, the Competition Authority blocked Kerry's 2008 acquisition of various brands from Breeo. The agency predicted the deal would give rise to "unilateral price increases" for, among other things, rashers. That decision was overturned on appeal to the High Court in March 2009, providing a rare opportunity to assess the agency's predictions against actual market outcomes.

If the Competition Authority was right, the merged entity's rasher prices should have increased soon after the High Court verdict. But National Consumer Agency price surveys suggest prices fell post merger. A February 2009 survey by the NCA found a 180g 6pk of Denny Rashers (a Kerry brand) cost €3.59 in Dunnes, Tesco and Superquinn. An August 2010 survey found the same pack cost €2.99 in Tesco, Dunnes Stores and Superquinn.

This is hardly robust analysis. Nor does it take account of the other two deals blocked following Competition Authority intervention: a 2006 prohibition of Kingspan’s acquisition of Xtratherm and a 2012 intervention to prevent Eason and Argosy merging. It may suggest, however, that some retrospective analysis of Competition Authority decisions would be worthwhile.

Another question is why the Competition Authority rarely interjects in deals notified to the European Commission?

Most mega-deals are notifiable to Brussels, even if the primary effect may be felt on Irish markets. That's why deals like Ryanair's bid for Aer Lingus and Hutchinson 3's bid for Telefonica O2 Ireland are subject to European Commission approval. Although it has the right to do so, the Competition Authority rarely asks for these cases to be referred back for its review.

Philip Andrews is a partner and co-head of McCann FitzGerald’s competition, regulated markets and EU law group