Merger Enforcement Guidelines (Banks) - Calculation of Market Shares and Concentration Levels

Merger Enforcement Guidelines (Banks) - Calculation of Market Shares and Concentration Levels

45. Although information which demonstrates that market share or concentration will be high cannot provide a sufficient basis, in and of itself, to justify a conclusion that a merger is likely to prevent or lessen competition substantially, it is a necessary condition that must exist before such a finding can be made. Absent high post-merger concentration or market share, the effectiveness of remaining competition in the relevant markets is generally such as to likely constrain the merged entity from acquiring, increasing or maintaining market power by reason of the merger.

46. Accordingly, the Director generally will not be concerned that the merging parties will be able to unilaterally exercise greater market power upon merger, where the post-merger market share of the merged entity would be less than 35 percent in the market. Similarly, the Director generally will not be concerned about a merger on the basis that the interdependent exercise of market power by two or more firms in the relevant markets will be greater than in the absence of the merger, where:

the post-merger share accounted for by the four largest firms in the market would be less than 65 percent; and,

ii) the post-merger market share of the merged entity would be less than 10 percent.⁽²¹⁾

47. If the sum of the merging firms' pre-merger market shares is below 35%, there are likely to be sufficient products and suppliers to which consumers can turn in response to any attempt by the merged entity to exercise market power. If the four-firm concentration level is below 65%, then coordination among firms in the market is likely to be too difficult to raise competition concerns. If there is other information to suggest that competition is likely to be lessened or prevented substantially even though these thresholds are not surpassed, the Bureau will consider this information in its assessment. These thresholds simply serve to identify mergers that are unlikely to have anti-competitive consequences from mergers that require more detailed analyses, before any conclusions regarding likely competitive impact can be reached. In all cases, an assessment of market shares and concentration is only the starting point of the Bureau's analysis.

48. Market shares are calculated both for firms that currently produce output in the relevant market, and also for firms that can potentially participate in the relevant market through a supply response. The market shares of existing market participants can generally be measured in terms of dollar sales, unit sales, or production capacity. In cases where products are undifferentiated and firms have excess capacity, capacity is normally a better reflection of a firm's relative market position and competitive influence than output.

49.In the case of bank mergers, it is inherently difficult to quantify capacity. Although the capacity of a bank or other financial institution to provide credit is partly determined by its access to deposits or other sources of funds, capacity can also be affected by the size of the delivery network, including the branch network, the availability of trained personnel who are familiar with the market or industry, and other factors. Since data on sales of banking products (i.e. loans and deposits) is more readily available than capacity data, the shares of market participants will be calculated on the basis of actual sales volumes. Information that suggests that this does not accurately reflect a particular firm’s competitive significance in the market will be taken into account in the assessment of the potential anti-competitive effects of the merger.

50. With respect to firms that can participate in the market through a supply response, only the output that is likely to be diverted to the relevant market within one year will be included in market share calculations. The Bureau will not in general assume that an institution that does not supply the relevant products (or supplies a minimal quantity of these products) is likely to respond to an increase in the price of the relevant products by diverting sales simply because it supplies similar products. For example, an institution that offers primarily large loans to large corporations will not be assumed to be able to easily switch to supplying smaller loans to small and medium-sized businesses. The profitable supply of different types of loans may require different types of activities (for example with respect to screening and monitoring), and an institution that is well adapted to supplying large loans may not be well adapted to supplying small loans, and may not be able to quickly supply such loans without expending considerable resources. The criteria used to assess whether a supply response is likely, and the likely magnitude of such a response, are discussed in the following section.

Firms That Can Participate in the Market Through a Supply Response

51.Firms that are likely to respond to a price increase in the relevant market within one year with minimal investments are considered at the market share stage of analysis. Firms that are likely to have an impact in the market after one year, but within two years of the merger, or whose entry requires considerable investment are considered when analysing Barriers to Entry (see paragraphs 76 to 87).

52. The following factors are relevant to determining if a firm will divert sales within one year in response to a post-merger price increase:

i) the cost of substituting production in the relevant market for current production ("switching costs");

ii) whether, and to what extent the firm is committed to producing other products or services; and,

iii) the profitability of switching from current production.

53. In general, the Bureau will determine whether a firm not currently supplying the relevant product can profitably respond to a small but significant increase in the price of this product within one year. Only the volume of output that is likely to be supplied in the relevant market at this price will be included in market share calculations.

The Initial Screening Test

54. In analyzing the competitive effects of a bank merger, it is difficult in practice and likely unnecessary for the Bureau to define markets associated with each product supplied by merging banks and with each location from which these products are supplied, and identify potential supply responses and evaluate the likelihood of entry into each of these markets. The fact that banks offer a vast number of products and services at a large number of locations to different types of customers implies that such an exercise would be extremely resource intensive and time-consuming. In practice, the Bureau will apply an iterative approach which, although entirely consistent with the framework described in the MEGs, allows the Bureau to more quickly identify the products and geographic locations which are more likely to create concern with respect to the loss of competition.

55. The Bureau will begin its analysis by conducting an initial screening test. The objective of this test is to "screen out" product offerings and geographic areas where a bank merger is unlikely to pose competition problems. The Bureau will apply the market share and concentration threshold tests, as outlined in paragraphs 46 and 47 to a pre-defined set of product offerings and geographic areas. Because the focus is to screen out markets from further analysis, the set of pre-defined product offerings and geographic areas will be narrow, and will depend on the availability of data.⁽²²⁾ As a result, use of this initial screen will tend to overreport the number of geographic areas where potential competition concerns might arise. This is not problematic, however, since this is only an initial screen and is not determinative for the transactions as a whole. The Bureau will rectify this deficiency in its subsequent competitive effects analysis, as described more fully in paragraphs 59 to 100.

56. If the post-merger market share and concentration thresholds are not exceeded for a given pre-defined product offering and a pre-defined geographic area, the Bureau is unlikely to be concerned that competition in the supply of that product in that area will be lessened substantially as a result of the merger. In the absence of information suggesting otherwise, the Bureau will have no cause to conduct a further review of this product offering and geographic area.⁽²³⁾

57. Finally, the product and geographic areas which are not excluded by this screening process will be subject to a full competitive effects analysis, as described in paragraphs 59 to 100.

58. In order to make the initial screening test analytically tractable, the Bureau will use a geographic mapping software program developed by Statistics Canada. ⁽²⁴⁾ This program is capable of quickly matching the market shares of each reporting financial institution for each pre-defined product offering within each pre-defined geographic area. The software program will also apply the market share and concentration thresholds to each area and list the results in tabulated form.

---

⁽²¹⁾ Given that the Bureau's definition of the market may differ from that of the parties, full information should be provided to the Bureau regarding the merger and its likely effect on competition, where either the anticipated four-firm concentration level (CR4), or the market share accounted for by the merged entity, is close to the above-described thresholds.

⁽²²⁾ Having explored available data sources at the Bank of Canada, OSFI and the Canadian Bankers Association (CBA), the Bureau intends to use the CBA database in its initial screening test as this is the most comprehensive, readily available database. The data consists of branch level sales information on a number of product offerings for many of the CBA members and non-members (including the four merging parties) based on the first three digits in the postal code of each represented branch (referred to as FSAs or forward sorting areas). While the database does not contain information on financial activity in all FSAs in Canada, it does cover all of the branches of the four banks currently proposing mergers. The Bureau will also be gathering additional information from other sources, including the parties directly and their existing and potential competitors.

⁽²³⁾ As noted in paragraph 55, the pre-defined geographic areas based on the CBA database are likely to be narrow and do not necessarily represent defined geographic markets. This will reduce the chances that true relevant geographic markets are incorrectly ruled out of any further competitive effects analysis by the screen.

⁽²⁴⁾ Statistics Canada has assisted the Competition Bureau in developing a spatial analysis tool to examine multi-product mergers in a local market context, which can be used for banking or other industrial sectors.