INST Committee Report

Merger Review Process

       The Competition Act provides for the civil review of mergers (sections 91 through 96) by the Competition Tribunal. On application by the Commissioner of Competition, the Tribunal may issue a prohibition or divestiture order with respect to a merger that is deemed to prevent or lessen competition substantially. However, before such orders are granted, varied or denied by the Tribunal, a well-established review process must take place. As a starting point, the Committee will provide a simple sketch of this merger review process, which will provide the necessary background to comment on the operations and enforcement of the merger provisions in the Act.

       Section 91 of the Competition Act sets forth the definition of a “merger,” which is deemed to occur when direct or indirect control over, or significant interest in, the whole or a part of a business of another person is acquired or established. The principal issue in this section is the interpretation of the words “significant interest,” which is considered to occur when a person acquires or establishes the ability to materially influence the economic behaviour of the business of a second person (i.e., block Director resolutions or make executive decisions relating to pricing, purchasing, distribution, marketing or investment). In general, a direct or indirect holding of less than a 10% voting interest in another entity will not be considered a significant interest. However, a significant interest may be acquired or established pursuant to shareholder agreements, management contracts and other contractual arrangements involving incorporated or non-incorporated entities.

       In general, a merger will be found to be likely to prevent or lessen competition substantially when the parties to the merger would more likely be in a position to exercise a materially greater degree of market power in a substantial part of a market for two years or more. Market power can be exercised unilaterally or interdependently with other competitors and its ascertainment will be determined according to the following Bureau screening processes:

On the other issue, from an enforcement perspective, there’s a lot of discussion in the business about how few cases there are and how much guidance is available to the public at large and the business and consumer legal communities about how decisions are made. This issue has been debated probably longer than private access, but I think it’s time we institute some form of formal decision publication process. [George Addy, Osler, Hoskin & Harcourt, 59:11:15]  

The EU has a process where, even though a transaction isn’t challenged, a decision is released describing how the agency went through its review, what its findings were, and what it considered important or not important. I think that would serve as a very useful public information service for the Bureau to adopt. [George Addy, Osler, Hoskin & Harcourt, 59:11:15]

[W]hen you’re sitting in the room negotiating the resolution, you also talk about what should be published, and it can interfere with some of the remedy. If you’re having to divest of a core asset, if you put too much out there, it becomes a fire sale, which makes it more difficult to resolve. If you’re going to give me a penny for my asset or $100 million for my asset, you’re going to have a different negotiation coming up with a resolution. [Robert Russell, Borden, Ladner & Gervais, 59:12:10]

Merger Review Workload and Service Standards

       Virtually every witness appearing before the Committee admitted that the Bureau has faced an unprecedented number of merger reviews over the past several years, which has, and continues to put, extraordinary pressure on its Mergers Branch staff. Table 7.1 provides the data to back up the first part of this claim. Excluding asset securitizations (which, since 1999, have been exempted from filing), merger filings have hovered about 340 per annum in the past four years, which is up more than 70% from the average of about 200 filings per year recorded in the first half of the 1990s. So the trend is definitely up over the past decade, but it is also up over the past five years, with 373 mergers being filed in 2000-2001, the highest ever.

I have submitted for consideration a one-month initial review followed by a four-month timeframe. If, after the first month, the Bureau does not go into a full-scale investigative mode, the merger is cleared. If they do go into that mode, then there is a fixed period … of four months … to complete the Bureau’s investigation. [Calvin Goldman, Davies, Ward & Beck, 59:09:20]  

       Data submitted to the Committee provides evidence of the second part of the claim. The Mergers Branch at the Bureau averaged 38 full-time equivalent person-years in the early 1990s, but has gradually increased to 57 in 2000-2001. Therefore, the Bureau’s Mergers Branch has grown by just less than 50% over the employment levels of the early 1990s, which is significantly below the merger filings growth rate of more than 85% in the same period.²⁹ Moreover, Table 7.2 indicates that the complexity of mergers that the Bureau has had to review is also increasing. Complex mergers and very complex mergers, which are increasingly resource intensive, have augmented their respective shares in the past four years by 4% each. Although non-complex mergers make up the vast majority of cases under review (between 80-90%), their share of total reviews undertaken by the Bureau has declined substantially in the past four years. This trend, the Bureau claims, is due largely to globalization and the inherent complexities associated with multi-jurisdictional cases.

I recommended earlier that in the area of merger review consideration be given to trying to define the time periods with statutory certainty so that business persons engaged in transactions, third parties interested in transactions and making submissions to the Bureau, … know there are fixed time periods, as opposed to the current service standard guidelines …This would promote certainty.  [Calvin Goldman, Davies, Ward & Beck, 59:09:15]

It will be interesting, now that this merger wave is sort of down, to see how resources are reallocated. As a result of that, it is certainly true that the other areas of the organization, such as the civil reviewable practices areas and conspiracy, are not nearly as well funded relative to other international comparisons. [Margaret Sanderson, Charles River Associates, 59:11:20]

       The revenue generated from fees related to merger review has been a significant but not a fully compensatory help to the Bureau’s budget constraint. The Bureau estimates that revenues from pre-merger notification, advance ruling certificates and advisory opinions will be in excess of $8.4 million in 2000-2001, $7.5 million of which will be available to the Bureau. Any fees the Bureau receives in excess of $7.5 million will be credited to the government’s Consolidated Revenue Fund. Given that the direct costs of merger review is estimated to be $9.5 million for 2000-2001, merger review revenues clearly fall short of cost recovery.

       In 1997, along with fees for certain services, the Bureau established and committed itself to meet a series of service standards when reviewing mergers. These standards are: non-complex mergers, 14 days; complex mergers, 10 weeks; and very complex, 5 months. Although the Bureau has, in a given year, met these targets 100% of the time, its performance level has varied without trend since 1997. In fiscal year 2000-2001, the Bureau met the three targets 95.7%, 92.5% and 100% of the time, respectively. The average and median turnaround times for merger review have at all times been shorter than the established standard. However, in every year since 1997, a relatively small number of merger reviews has fallen well outside the target date. These poor performances appear to be isolated cases that are not the result of systemic failures, but are more likely owing to human error — errors probably committed on the part of Bureau staff and merging parties. This performance and the targeted standards, the Committee finds, are reasonable. Although there were complaints about the merger review process made to the Committee, stakeholders had not complained about this aspect.

[T]he Bureau’s workload over the past few years has greatly increased. Unfortunately, our resources have not kept pace ... In a recent survey involving five comparable competition authorities, our Bureau had the second-lowest level of funding on a per-capita basis. Our demands continue to grow, largely due to globalization and our increased mandate. Ten years ago, the great majority of cases examined by the Bureau were domestic in nature. Today, not only are there more cases, but a very large number of them have an international dimension. This is demonstrated by the increasing number of multi-jurisdictional mergers and international cartels. [Gaston Jorré, Competition Bureau, 64:09:10]  

       The Committee believes that the routine merger review procedures of the Bureau are not the cause of selected protracted merger reviews of which people complain. These reviews bog down only when the Commissioner has unresolved issues with the merger (as proposed) and intense negotiation begins for restructuring the merger proposal or when seeking a consent order, or where a contested Tribunal proceeding is going to be launched. As a consequence, the Committee sees no benefit in enshrining strict deadlines for merger review in the Act, as some commentators have suggested. Indeed, the Committee sees more harm than good coming from such Act-imposed deadlines. Given an inviolable deadline, the Bureau would be forced to work more intensively on cases that are likely to run into difficulty and breach the deadline, sacrificing resources in other reviews and therefore delaying less problematic mergers. In effect, strict or Act-imposed deadlines will compress the time distribution of completed reviews, but only at the expense of higher average turnaround times.

Merger Enforcement Record

       The combination of an unexpected and uncontrollable merger review workload, growing at rates in excess of that of staffing, with that of quick turnaround times provided by the Bureau is a situation that lends itself to the perception that vigorous enforcement of the Act may have been sacrificed. The Committee will investigate.

       Table 7.3 provides the Bureau’s statistical record of merger enforcement under the Competition Act.³⁰ The Bureau’s entire enforcement record over the 1986-2000 timeframe is included, but the data is broken down into three four-year periods to look for trends in the statistics while overcoming a small numbers problem from which the data suffers. What is clear from the statistical record is that the past four years has involved almost as many merger examinations by the Bureau than that of the previous two four-year periods. Very little else can be discerned with such a high degree of confidence.

From the Competition Bureau’s perspective, it has limited resources … the Bureau is in fact fairly strapped when it comes to resources, so it has to make responsible decisions as to how it deploys those resources. It currently has case-screening criteria that would bias its decisions in favour of bringing cases that have a broader economic impact. [Paul Crampton, Davies, Ward, Phillips & Vineberg, 65:10:10]

From the Competition Bureau’s perspective, it has limited resources … the Bureau is in fact fairly strapped when it comes to resources, so it has to make responsible decisions as to how it deploys those resources. It currently has case-screening criteria that would bias its decisions in favour of bringing cases that have a broader economic impact. [Paul Crampton, Davies, Ward, Phillips & Vineberg, 65:10:10]

       The vast majority of mergers pose no threat, or raises no issue, under the Competition Act. Donald G. McFetridge reports that about 1.6% of all publicly reported mergers (7.5% of those examined) between 1986 and 1994 raised an issue under the Act.³¹ According to the data in Table 7.3, the number of issues raised in merger cases has further declined in the latter half of the 1990s. When one subtracts mergers in which monitoring was the chosen enforcement response by the Commissioner — because they were never later challenged or brought back under investigation — the number of mergers that raised an issue under the Act has average only 2% of examinations undertaken by the Bureau.

Virtually all the cases that have been brought in the 15-year period since the Tribunal was created and the merger provisions were decriminalized have involved mergers that had already been consummated. At that point the merging parties had every incentive to hunker down and fight. By contrast, business people invariably have no appetite whatsoever to become involved in contested proceedings where their transaction has not yet been consummated. [Paul Crampton, Davies, Ward, Phillips & Vineberg 65:09:55]

[W]e can review any merger, no matter what the size. Where size comes in is whether you have to notify us. … And I guess … it’s a trade-off … if the world were cost-free, it would be nice to look at every merger and have notification. But given the costs imposed, there has to be some level before you create a notification process, and that’s why there is a threshold for notification.[Gaston Jorré, Competition Bureau, 64:09:30]

       The Committee finds it rather curious that, except for contested proceedings, all enforcement responses fell out of favour with the Commissioner (then the Director) in the mid-1990s. However, except for monitoring, all other enforcement responses, such as pre- and post-closing restructuring/undertakings and consent orders, have come back into favour. Moreover, what the Committee finds disturbing is that the number of mergers abandoned by their proponents as a result of the position taken by the Commissioner has declined substantially over the late 1990s. For example, 18 merger proposals were abandoned by their proponents of 1,614 merger examinations undertaken by the Bureau in the two four-year periods starting in 1988 and ending in 1996. Given 1,492 merger investigations and similar vigorous enforcement by the Commissioner, one would have expected about the same number of abandonments, 18, in the 1996-2000 period; however, there were only 4 such abandonments; less than one-quarter of what would reasonably be expected.

       To the Committee the data suggest one of three explanations: (1) mergers have become less problematic from a competition perspective; (2) the business community at large has in the past five years come to realize that the Commissioner is a vigorous enforcer of his Act and has increasingly acquiesced to other restrictive undertakings imposed by him/her as a means of realizing their mergers; or (3) the business community has in the past five years come to realize that the Commissioner’s budget is insufficient to vigorously enforce his Act and that he must acquiesce to the merging parties by seeking other non-vigorous merger enforcement methods than that of contesting them under a costly Tribunal proceedings.

       Without qualitative information on these mergers, the Committee cannot draw definitive conclusions. However, the Committee fears that the third explanation is more likely correct and, at least in part, explains the fewer merger proposal abandonments. Somewhat paradoxically, the lack of information published on mergers that the Commissioner did not oppose as a means of protecting private and strategic market information from being made public may be providing more protection, in terms of accountability, to the Commissioner — a state of affairs that the competition law community has long complained about.

       In any event, vigorous enforcement of the merger review provisions can be accomplished by providing the Bureau with adequate resources and allowing it to exercise greater selectivity in the review of mergers that are likely to pose a competition issue — recommendations that this Committee advocates.

Review Thresholds

       The claim that the Bureau receives insufficient funding for optimal enforcement of the Act, in particular mergers, is not new. In fact, the competition law community has made the Committee aware of this fact since it undertook its study of the Competition Act and its publishing of the Interim Report. The desire for a more complete evaluation that would consider other consequential impacts on enforcement has held the Committee from venturing beyond the call for more resources to be allocated to the Bureau. Given the concern raised in the preceding section, the Committee is now prepared to evaluate specific proposals to raise the merger review thresholds as a way of focusing scarce resources on the larger merger reviews and the enforcement of other aspects of the Act.

 [I]f parties to smaller transactions — mergers, for example — want to proceed with their transaction without notifying the Competition Bureau and try to fly below the radar screen, they have to take the risk that the Competition Bureau isn’t going to find out about the transaction for three years, because if the Bureau does, it can bring an application to the Tribunal for up to three years and force divestiture. That’s a huge risk, and business people typically do not want to assume that risk without comfort. So I find myself frequently, at any given time, having several matters on the go that involve transactions that are not above the notification thresholds, but the parties nevertheless want comfort from the Competition Bureau in the form of a no-action letter or an advance ruling certificate before they put their money on the table and proceed with the transaction. [Paul Crampton, Davies, Ward, Phillips & Vineberg, 65:10:10]

       Since the adoption of the Competition Act in 1986, the parties to any significant merger — that is, a merger of a certain size as set out in the Act — are required to notify the Commissioner before closing the transaction. Although all proposed mergers may be reviewed by the Commissioner, only those mergers (i.e., asset or share acquisitions) involving more than $35 million in gross revenue from sales per annum in or from Canada, or involving more than $400 million in combined assets or sales (including affiliates) in Canada, must notify the Commissioner of the proposed transaction. The transactions threshold for amalgamations is $70 million. Both the gross sales and combined asset thresholds have remained unchanged since 1986.

       Between 1986 and 2001, inflation of more than 40% (as measured by the consumer price index or CPI) has occurred. Consequently, the $35 million and $400 million thresholds have captured many more mergers than Parliament had intended when the Act was adopted. Indeed, the possible over-inclusiveness of mergers that must automatically undergo review may have been a constraint on optimal enforcement of the Act — the Bureau suggests that the gross-revenue-from-sales threshold of $35 million has been particularly binding. In other words, some resources currently devoted to merger review may be more effectively allocated to other activities, either to the review of larger mergers or to the enforcement of other provisions of the Act.

       The Bureau performed a special request for the Committee that indicates that approximately one in ten mergers examined by its Mergers Branch in the past year fell within the $35 to $50 million transactions range. This statistic, one in ten, suggests that raising the transactions threshold to $50 million would reduce the total number of merger filings by about 40 per year. Unfortunately, we were unable to find out how many of these one-in-ten mergers posed an issue under the Act. Nevertheless, given the deficiency in filing revenues to cover the direct costs of merger review and the Committee’s belief that there are more pressing needs for enforcement of other activities, we believe that it is best to raise the $35 million transactions threshold to $50 million. The Committee, therefore, recommends:

One thing that would help … is the elevation of the thresholds to align them with the economic value of the threshold as it was when it first came in, in 1988. In 1988 a $35 million threshold on the transaction size was put in place. … In the meantime, the value of the dollar has eroded by more than a third, and if we were to make that adjustment today, I think it would release from the system, from the review, maybe 40% of the cases they now deal with, and would enable more people to be freed up to do other things. [Tim Kennish, Osler, Hoskin & Harcourt, 59:09:25]

From an enforcement perspective, I would like to see increasing attention paid to other provisions of the Act, perhaps becoming a little less risk-averse from an enforcement perspective in dealing with mergers. We also heard this morning about the possibility of increasing thresholds. That might help too. [George Addy, Osler, Hoskin & Harcourt, 59:11:15]

Furthermore, the Committee believes there is merit in formalizing such considerations and, therefore, recommends:

Mergers and Efficiencies

       Section 96 of the Competition Act sets Canada’s competition legislation apart from those of other countries. This section states that: “The Tribunal shall not make an order if the merger brings about gains in efficiencies that are greater than, and will offset, the effects of any prevention or lessening of competition”; this has been interpreted by some as being consistent with what is known as the “total surplus standard.”

       The Act also goes to considerable lengths to explain both what should and should not be included as a gain in efficiency. For example, the Act states that “the gains in efficiency” to be considered are those that “would not likely be attained if an order were made in respect of the merger”; that is, they must be merger specific. This implies that if the efficiencies could be realized in a manner that generates less anticompetitive harm than that created by the merger, then the efficiencies would not be ascribed to the merger. For example, efficiencies that could occur through internal growth or unilateral rationalization would not be ascribed to the merger. Alternatively, there may exist other cooperative means of achieving the efficiencies, such as joint ventures or a restructured merger, which would create lesser anticompetitive effects. Additionally, the efficiencies must be real and not just pecuniary; that is, the merger must bring about a real savings in resources and must not stem from greater bargaining or purchasing power that is essentially redistributive among members of society.

But in looking at it historically, in countries that have had strong competition laws, like the U.S., and countries that had very weak competition laws, like Japan, they found that they didn’t end up with very productive and efficient economies when they didn’t foster competition and make sure those efficiencies, that productivity and efficiency, were there. So when the cases are looked at, it’s not just on the basis of the consumer or the small business alone, but the Canadian economy and what benefits consumers as a whole.[Robert Russell, Borden, Ladner & Gervais, 65:10:15]

The analysis of efficiencies in competition law in this country is in a state of disarray, to say the least. We’ve had 15 years or more of toing and froing on it, and still don’t know if we have anything we can work with. So if you’re going to go for the section 45 reform … [focus on] what constitutes the civil test. [Donald McFetridge, Carleton University, 59:10:05]

       Canada is the only country known to have a competition legislation that requires the efficiencies likely to be produced by a merger to be weighed against the likely anticompetitive effects of the merger. This approach occupies the middle ground between the European Union approach, whereby the merging parties are invited to make claim to efficiencies that the Merger Task Force will consider (which introduces lobbying into the mix), and the U.S. approach, which requires efficiency gains to be so great that prices will not rise as a result of the proposed merger (the so-called “price standard”). In retrospect, this is not an unreasonable approach and, in fact, may be a strategically sound one given Canada’s relatively smaller and open market economy.

       Although this legislative defence is unique among the industrialized countries of the world, its 15-year history has not been very hospitable to merger proponents. The Commissioner has not even once found the efficiency gains to a merger proposal sufficient to offset any lessening of substantial competition. This behaviour contrasts sharply with the Commissioner’s findings of efficiency gains on many occasions pertaining to exclusive dealing and tied selling cases. Furthermore, in this same 15-year period, the Tribunal has only once decided (Superior Propane) and twice commented on efficiency gains (Imperial Oil and Hillsdown). The elucidations, however, have been confusing to say the least. Just when the Tribunal has come to agree with the Bureau’s guidelines on the treatment of efficiencies according to the “total surplus standard” (Superior Propane), the Bureau abandoned its guidelines. To further confuse the issue, the Federal Court weighed in and partially overturned the Tribunal’s decision in favour of expanding the strictly quantitative analysis of the “total surplus standard” to include redistributional and other qualitative effects of the merger, while neither advocating the “consumer surplus standard” or the American “price standard” approach. This Court direction had the consequence of opening the door to the Commissioner, as well as to the lone dissenting Trial judge sitting on the Superior Propane case, to advocate the “consumer surplus standard.”³² Sensing that the latter standard would render section 96 virtually ineffective, the majority opinion of the Tribunal panel chose to supplement the “total surplus standard” with a calculation of what is described as the “adverse social effects” of the merger, i.e., the wealth redistributed from “poor” Canadian consumers to the shareholders of the merging parties.

 [W]hether the efficiencies outweigh and offset the anti-competitive effect and really, in principle, that includes everything. It includes all the anti-competitive effects and some of those are measured quantitatively but … [t]hen you have other factors which are more qualitative and you can’t really measure. To give you a very simple example, how do you weigh the impact of loss of choice. If you go from having two people you can buy something from to just having one, you’ve clearly lost something, apart from price and it’s not something you can really value but it’s certainly something that has to be weighed in. [Gaston Jorré, Competition Bureau, 64, 10:00]

      The Tribunal’s decision in Superior Propane may or may not be satisfactory; it is not clear if such precise calculations of the wealth redistributed from “poor” consumers to the shareholders of producers will be possible in future cases. Moreover, so many different interpretations of Parliament’s intentions when it stated that the “effects of a merger that would prevent or lessen competition” must be weighed against the “gains in efficiency” suggest that more expert study is required.³³ Accordingly, the Committee recommends: