The ACCC’s draft new Merger Guidelines (“the Guidelines”) are long overdue and, even if for that reason alone, should be welcomed. The Commission’s previous Guidelines, issued in 1999, were the culmination of a lengthy process of elaboration, and represent a substantial and enduring intellectual achievement. But the theory and practice of merger analysis has moved on considerably since then. Updated Guidelines were badly needed if clear guidance was to be provided about the Commission’s approach to analysing mergers.

The proposed new Guidelines do a useful job of setting out an analytical framework for assessing mergers for possible breaches of the merger control provisions set out in s.50 of the Trade Practices Act.  However, there are important respects in which they are puzzling, and others in which they raise serious concerns.

The most striking puzzle arises with respect to one of the key novelties in the Guidelines. This is the price test they set out for assessing whether a merger is likely to substantially lessen competition. More specifically, the Guidelines say that the Commission’s “starting point for analysis” will be to see whether the merged entity could profitably “increase prices by around 5 to 10 per cent for a non-transitory period of time”. If it could, then (recognising that what constitutes a substantial lessening of competition will vary from merger to merger) the presumption would be that the merger would be likely to breach s.50 (para 2.4).

Given this price test, it would seem to follow that if a merger was not likely to “increase prices by around 5 to 10 per cent for a non-transitory period of time”, then the merger would not be likely to substantially lessen competition. If so, then what the Commission has done is to adopt what amounts to a “consumer welfare” standard, in which the key issue is whether a merger would make consumers worse off: if it would, then the merger would be blocked (unless it was authorised under the authorisation provisions); if not, it would not be.

There is merit in such a test – but the Commission does not seem to have considered its implications. In particular, it means that merger-induced efficiencies that affect price must be taken into account in assessing the transaction. Those efficiencies will affect the profit-maximising price level for the merged entity, usually reducing it; whether prices will be higher or lower than they would be in a counterfactual world cannot be known without assessing those efficiencies and their impacts.

There is merit in that too – and it follows naturally from the logic of having a “consumer welfare” standard. But the Commission (and this is the puzzle) denies it. More specifically, the Guidelines, when they come to discuss efficiencies, seek to limit the extent to which merger-specific efficiencies will be considered in the assessment, and say that “a merger that removes or weakens competitive constraints.. will (unless authorised) contravene s. 50 – even if the merger results in a more efficient firm with a lower cost structure” (para 6.64).

Taken as it stands, this sets up a seeming contradiction: a substantial reduction in competition is one that allows a firm to profitably increase price by 5 to 10 percent, i.e. the profit-maximising price post-merger is 5 to 10 percent greater than it would be in some counterfactual; however, even a merger that does not have that effect (because the merger-related efficiencies mean that profit-maximising prices post-merger are not 5 to 10 percent greater than in some counterfactual) may substantially lessen competition.

Now, it is, of course, possible that the price test is merely a sufficient, but not a necessary, condition for a substantial lessening of competition; but, if so, the test would not provide a great deal of guidance; moreover, if that is all the Commission intends the test to be, then it should say so, and explain why. In effect, it would be a shame were that so, as one would have thought that mergers that made consumers no worse off, and presumably increased efficiency in the process, ought not to be prevented. In contrast, if the price test does mean what it seems to say – that a merger lessens competition if it creates scope for profitable, substantial, price increases – then the treatment of efficiencies in the Guidelines needs to be changed.

This lack of clarity may simply reflect the draft nature of the Guidelines. What is more concerning is the seeming expansiveness in the range of circumstances the Guidelines discuss as being potentially problematic. This expansiveness takes two forms.

First, the Guidelines do away with the previous safe-harbours, i.e. the market share thresholds below which the Commission would presume that mergers were unlikely to substantially lessen competition. The lack of such safe-harbours, at least as rebuttable presumptions, makes the ACCC Guidelines different from those in other major jurisdictions. This might reasonably be viewed as signalling a widening of the net.

Second, the Guidelines devote a great deal of space to discussing vertical and conglomerate mergers and the ways in which they could, at least in theory, harm competition: indeed, the discussion of the potential unilateral effects of these mergers dwarfs that of horizontal mergers. This is striking, as it is usually sensible to presume that mergers that are purely vertical or purely conglomerate will not materially harm competition. Indeed, a recent survey of the empirical literature on vertical mergers (by Francine Lafontaine & Margaret Slade, "Vertical Integration and Firm Boundaries: The Evidence" in the Journal of Economic Literature, 2007) finds that the “weight of evidence is overwhelming” that vertical mergers tend to benefit consumers, even in industries where concentration levels are high.

As a result, what the Guidelines seem to be doing is signalling a strong interest by the Commission in mergers which, if the results of empirical studies are to be believed, very rarely harm competition. This could open the door for a significant compliance burden to be placed on business, as those mergers and their impacts are subjected to the rigours of the Commission’s merger evaluation process. The extent of that burden may be all the greater as the Guidelines make it clear that the Commission will not be shy in seeking access to internal documents that relate to the proposed transaction.

It may be that it is not the Commission’s intention to so widen the net – but, if so, that is hardly clear from the Guidelines. Moreover, the Guidelines’ discussion of vertical and horizontal mergers seems seriously unbalanced, placing weight on theoretical possibilities that do not appear to be of great practical relevance, especially if offsetting efficiencies are properly taken into account.

There are many other aspects in which these Guidelines would benefit from further thought. For example, the discussion of coordinated effects needs work, as does that of entry; presumably, the Commission will take some time to get these right. The 1999 Guidelines were extensively tested, and that contributed to their quality; the new version would benefit from a similar process. It is to be hoped that the issues raised above will be dealt with as that process unfolds.