Round Table on Monopsony and Buyer Power

Directorate for Financial and Enterprise Affaires
Competition Committee

For Official Use

Organization for Economic Co-operation and Development

Note by Canada

JT03253147

Complete document available on OLIS in its original format

This note is submitted by Canada to the Competition Committee For Discussion at its forthcoming meeting to be held on 21-23 October 2008.

---

1. The Canadian Competition Bureau (the “Bureau”) is pleased to provide the following discussion of the roundtable topic “Monopsony and Buyer Power”.

1. Definition

1.1 What types of power exercised by buyers have raised antitrust concerns in your jurisdiction? Please define the different types of buyer power that are relevant.

2. The Bureau is generally concerned with monopsony power as it is defined by the OECD; that is, where the price of an input is depressed below the competitive level such that it results in a decrease in the overall quantity of the input produced or supplied in a relevant market.¹ This concern exists regardless of whether that market power is exercised by a single buyer, a coordinating group of buyers, or a buyer that is not otherwise the only buyer in a market (a dominant buyer). In this sense, the Bureau is also generally concerned with what is referred to by the OECD as “oligopsony power”. When considering likely output decreases as a result of a price decrease below competitive levels, the Bureau will take into account any type of quality-adjusted decrease in output² or a corresponding diminishment in any other dimension of competition. Also, when considering price decreases, the Bureau will take into account any change in terms of trade that amounts to a price decrease.

3. The Bureau’s concern with monopsony power has typically arisen in merger reviews. There has, to date, been no litigated case involving monopsony power in a merger³ but there have been negotiated settlements that, at least in part, have been designed to address such concerns.⁴

4. The Bureau’s concern about bargaining power (as defined by the OECD: power to reduce input prices but not so much so that prices fall below competitive levels) primarily (but not exclusively) arises in the context of buyer cartels. Such cartels fall under the conspiracy provisions of the Canadian Competition Act and as such are subject to the requirements of that provision, including the requirement that the buyer cartel prevent or lessen competition unduly. The Bureau has carried out a number of investigations into such cartels.⁵

5. As to bargaining power in unilateral conduct cases, in order for bargaining power to amount to abuse, it must be engaged in by a dominant firm, must constitute a practice of anti-competitive acts in that its purpose is an intended exclusionary, disciplinary or predatory effect on a competitor(s), and that the effect of this is or is likely to be a substantial lessening or prevention of competition. To date, there have been no abuse investigations involving bargaining power that have involved lower input prices. It is unlikely that lower input prices would, in isolation, meet the requirement of a practice of anti-competitive acts.

1.2 What is (are) the appropriate definition(s) of buyer power? What is the relationship between monopsony power, oligopsony power, bargaining power, and countervailing power?

6. The Bureau considers buyer power to be a general term that includes within it monopsony power, oligopsony power, and bargaining power. Countervailing power is used to mean the power to discipline an exercise of market power, monopsony power or bargaining power, and so the term countervailing power might be applied to either sellers or buyers.

7. Monopsony power is understood to mean those instances where a price decrease is such that it falls below competitive levels and there is a corresponding reduction in the input supplied or a corresponding diminishment in any other dimension of competition. As noted in footnote 1, here and throughout this submission, monopsony includes within its meaning situations where supply is perfectly inelastic such that a decrease in price below competitive levels does not result in a decrease in output. As is the case when examining downstream market power, the Bureau generally considers such price and output decreases in a relevant market. As noted above, monopsony power is generally used by the Bureau to include within its meaning oligopsony power.

8. Bargaining power means the power to decrease prices but those price decreases are such that they do not fall below competitive levels, and so the price decrease is associated with increased purchases of the input, rather than decreased purchases.

9. Countervailing power is normally the expression used when considering whether an act, such as a merger, is likely to result in the ability to sustain a material price increase and the Bureau assesses whether one or more buyers have a countervailing ability to constrain that exercise of market power. Conversely, in the case of monopsony, oligopsony or bargaining power, one or more sellers may have a countervailing ability to constrain the exercise of monopsony, oligopsony, or bargaining power.

2. Identifying Buyer Power

2.1 What determines the extent to which a buyer can exercise monopsony power? Bargaining power? Other types of buyer power?

10. A necessary condition for an exercise of monopsony power is that the input be supplied in a market characterised by an upward-sloping supply curve in the relevant range of production.

11. The fewer selling options there are, the more likely it is that a large buyer or a coordinating group of buyers will be able to exercise either monopsony or bargaining power. What will largely determine whether the resulting price decrease remains above competitive levels will be the extent of selling options. The fewer the selling options, the more likely it is the bargaining power will extend to monopsony power.

2.2 What metrics can, and have been used, to identify monopsony power? Bargaining power? Other types of buyer power? What are their strengths and weaknesses?

12. The Bureau first determines whether it is likely a firm has buyer power.⁶ It then tries to determine whether that buyer power is likely to entail the special case of monopsony power. This involves the inherently difficult exercise of trying to determine whether prices are competitive or not.

13. The first step in assessing whether an entity is likely to have buyer power is typically a determination of the relevant market in which the entity makes its purchases. The conceptual basis used for defining markets is, mirroring the selling side, the hypothetical monopsonist test. Conceptually, a relevant market is defined as the smallest group of products and the smallest geographic area in which a sole profit-maximising buyer (“hypothetical monopsonist”) would impose and sustain a significant and non-transitory price decrease below levels that would likely exist in the absence of the act in question (for example, a merger). The relevant product market definition question is thus whether suppliers, in response to a decrease in the price of an input, would be able to profitably switch to alternative buyers or modify the input they sell in sufficient quantity to render the hypothetical monopsonist’s input price decrease unprofitable.

14. Buyers currently buying the input in question will generally be considered participants in the relevant market. Buyers not currently buying the input may be considered participants in the relevant market provided, in the event of a small but significant input price decrease, the buyer would buy the input and the seller would sell it. It is of note that buyers need not participate in the same downstream market in which the buyer at issue (for example, the merging parties) participates. For example, a grocery story likely participates in a local market for the sale of groceries, but it may purchase a food input, such as corn, from a producer that may have regional, national and even international buyers for the sale of its product.

15. Once the relevant market is known and its buyers identified, the size of the entity’s purchases of the input relative to its suppliers’ total sales of the input in the relevant market is determined. If the entity only accounts for a small percentage of its suppliers’ sales in the relevant market, these suppliers are generally considered well placed to forgo sales to the entity in favour of other buyers when faced with an attempt to lower input prices.

16. If the entity accounts for a significant portion of input purchases, barriers to entry into buying are considered.

17. If the entity accounts for a significant portion of input purchases and barriers to buying the input are high, likely factors for consideration when trying to distinguish between bargaining and monopsony power include as follows:

• The shape of the supply curve in the relevant range of output: As noted above, upward-sloping supply curves are a necessary condition for monopsony power. If there is evidence that supply is highly elastic over a relevant period of time, the Bureau is less likely to pursue a monopsony power case. Relevant considerations when evaluating the shape of the supply curve include any factors, such as pre-existing contracts, that may influence the time it takes for supply to adjust to new demand conditions.
  
  
• Whether upstream supply of the input is characterised by a large number of sellers and low barriers to entry such that the normal selling price of a supplier is likely competitive.
  
  
• Whether it seems likely that certain suppliers will exit the market in response to the anticipated price decrease or will scale back production.
  
  
• Where possible, empirical techniques for analysing the effect of historical changes in supply on price and quantity to help determine whether monopsony power exists.

18. If it appears that monopsony power is possible (including the case where the supply curve is perfectly inelastic), the likelihood of such power actually being exercised may be considered. Factors relevant to this include the possibility of an exercise in monopsony power jeopardizing a long-run source of supply, and the possible costs to the potential monopsonist of decreased output in the downstream market that may follow decreased input purchases.

2.3 What is the relevance of market definition for identifying monopsony power? Bargaining power?

19. Market definition is relevant for identifying both monopsony and bargaining power, and is used by the Bureau in the manner described above.

20. Identifying relevant markets is of particular importance in monopsony cases since the decrease in input production of concern need only occur within the relevant market. It need not occur more generally. For example, a grain purchaser may be able to exercise monopsony power in its grain purchases in a region of Canada. That region of Canada is a relevant geographic market because grain producers located therein would not be able to discipline a decrease in grain prices by selling to other buyers or geographic locations. Grain production in other markets, say, for example, those located outside of Canada, may, however, increase output in response to the decrease in output in the relevant market of concern. This, however, would not be a consideration when assessing the likelihood of monopsony power in the relevant market of concern (except to the extent that it may influence the potential monopsonist’s incentives to exercise monopsony power in the first place). Where the Bureau is concerned with monopsony power (where, again, this includes the case where the supply curve is perfectly inelastic so that there is no decrease in output as a result of a decrease in price below competitive levels), it is only concerned with monopsony power in the identified relevant market (this also means that there need not be a decrease in output in the corresponding downstream market). This position is consistent with inefficiencies arising from such behaviour more generally. Economic theory suggests that an output decrease in response to monopsony power in one relevant upstream market that results in output increases in other relevant upstream markets is typically the result of inefficient substitution towards less efficient producers.

2.4 What issues arise in adapting the hypothetical monopolist test to define markets to identify monopsony and/or bargaining power?

21. The main issue in adapting the hypothetical monopolist test to define markets to identify monopsony and/or bargaining power arises in regard to product market definition. Since the hypothetical monopsonist test is applied from the perspective of sellers and the buyers to whom they can sell, the relevant product market definition question is whether suppliers, in response to a decrease in the price of an input, would be able to profitably switch to alternative buyers or modify the input they sell in sufficient quantity to render the input price decrease unprofitable to the buyer. This opens the door to the possibility that the product market would consist of products that are otherwise unrelated or, from the buyer’s perspective, not substitutes. For example, if grain suppliers could discipline a grain price decrease by switching to fruit production, this would imply that, for the purposes of assessing the likelihood of monopsony power, grain and fruit are in the same product market.⁷

22. While this is technically correct, the time and money required by sellers to switch into the production of other products so as to discipline a price decrease are often either too costly and/or take too long to be an effective discipline. Consequently, while it is possible that relatively modest changes to an input so as to expand suppliers’ selling options can be achieved at little cost and time, and so would be a consideration, the Bureau’s more likely focus in regard to market definition in upstream cases is geographic, whereby alternative locations and the buyers in those locations are identified.

3. Welfare Effects

3.1 What are the welfare effects of the exercise of the different kinds of buyer power? How does the exercise of the different types of power by a buyer affect input suppliers and consumers of the firm exercising the buyer power? Do the welfare effects of the exercise of buyer power depend on the market structure downstream?

23. Monopsony power will generally result in allocative inefficiency in the upstream market in which it takes place.⁸ The decrease in input purchases can also result in decreased downstream output with corresponding higher prices, which will adversely affect downstream consumers. Such an effect in the downstream may not arise, however, since that market may be competitive such that a decrease in output by one participant does not have any impact on market price and output. Under Canadian law, there need not be harm by way of a price or output effect in the downstream market in order for an exercise of monopsony power to be considered harmful. Consequently, it is typically sufficient that there be a reduction in output (or a corresponding decrease in another dimension of competition) in the upstream market in order for an antitrust concern to arise.

24. As noted above, however, the downstream market structure can impact the profitability of an exercise of monopsony power and, in turn, the potential monopsonist’s incentive to reduce input purchases. In competitive downstream markets, the only downstream effect of decreased input purchases and the corresponding decrease in downstream output will be lost downstream market share. This will reduce, although not necessarily eliminate, the incentive to exercise monopsony power. Reduced input purchases may nonetheless be profitable for the firm if the benefit of the cost savings upstream outweighs the costs associated with decreased market share downstream.

25. Cases of non-conspiracy related abuse of bargaining power are normally (but not necessarily exclusively) considered under section 79 of the Canadian Competition Act. Under this section the Bureau considers (1) whether a firm is dominant; (2) whether the firm has engaged in a practice acts intended to exclude, discipline or predate a competitor; and (3) whether the practice has resulted in a substantial prevention or lessening of competition in a market. That market can be either the upstream market in which input purchases are made or the downstream market in which product based on those inputs are sold. The practice of anti-competitive act(s) at issue would be assessed to see whether it likely has the effect of substantially lessening or preventing competition in the relevant market by way of creating, enhancing or preserving the dominant firm’s market power.

26. Acts that result only in transfers of wealth among various market participants as a result of lower prices that either do not constitute monopsony power (where monopsony power includes those situations where the supply curve is perfectly inelastic such that a decrease in price below competitive levels only entails a transfer) that has the effect of substantially lessening or preventing competition, or do not ultimately result in substantially lessened or prevented competition as a result of an adverse effect on a competitor, generally do not raise antitrust concerns in Canada. The exception to this arises under the cartel provisions, where conspiracies to lower prices that do not necessarily amount to prices below competitive levels may be found to warrant investigation.

3.2 With respect to conduct that increases buyer power, what welfare effects suggest that such conduct should be a concern of competition policy? If the conduct increases buyer power but does not result in an increase in prices for downstream consumers, should it be enjoined? Why?

27. As noted in the response above, under the abuse provisions, anti-competitive conduct that increases buyer power is only a concern if the intended effect is an exclusionary, disciplinary or predatory effect on a competitor such that there is a substantial lessening of competition in the market in which that competitor participates or otherwise would have participated. If the conduct increases buyer power but does not, through the mechanism described above, result in created, enhanced or preserved market power, the required elements of the Canadian abuse of dominance provisions would not be met and the conduct could not be enjoined. If the market of concern is the downstream market, the required elements of the abuse provisions will normally entail an increase in downstream consumer prices (or a negative effect on some non-price element of competition, such as quality). If the market of concern is the upstream market for the purchase of inputs, prices to consumers need not necessarily increase for the conduct to be considered anti-competitive, but upstream input prices would typically have to fall below competitive levels.

3.3 What are the welfare effects of the secondary line price discrimination implied by the cycle hypotheses, perhaps augmented by a waterbed effect? Is the change in downstream market structure necessarily harmful for consumers? If not, on what basis can consumer welfare improving cycles be distinguished from those that are deleterious?

28. The Canadian abuse provisions do not have an efficiency defence, but business justifications that are pro-competitive will be considered when assessing particular conduct. Such a justification can overcome the reasonably foreseeable effects of the conduct if the firm(s) can show that those anti-competitive effects were not the overall purpose of that conduct. The Bureau considers credible efficiency or pro-competitive rationales to generally fall into one of two categories: activities that minimise costs of production or operation, independent of the elimination or discipline of a rival; and activities that improve a firm’s product, service, or some other aspect of the firm’s business. Consequently, in order for a change in downstream market structure, which may result from secondary line price discrimination implied by the cycle hypotheses, to be considered harmful, it must be the case that the intended purpose of the discrimination was the exclusion or disciplining of competitors (as opposed to some pro-competitive rationale), and that such exclusion/disciplining results in enhanced downstream market power. A simple observation of price discrimination in the input market would not normally be sufficient to conclude any anti-competitive effect.

3.4 Under what circumstances is a waterbed effect possible?

29. The Bureau has no experience with waterbed effect cases. The OECD notes that a waterbed effect arises if the effect of the exercise of the bargaining power is to increase the input price paid by other buyers. Such a case would only be subject to the Canadian Competition Act if a dominant firm engaged in an act that relatively increased its competitors’ input costs (a theory of raising rivals’ costs) through a practice of anti-competitive act, such that the intended effect of the act was to exclude or discipline their market participation. Merely being able to negotiate relatively more favourable prices does not typically meet these requirements.

3.5 What should be, and is, the legal status of the exercise of the different types of buyer power? Why?

30. As noted above, acts that result only in transfers of wealth among various market participants as a result of lower prices that either do not constitute monopsony power (where monopsony power includes those situations where the supply curve is perfectly inelastic such that a decrease in price below competitive levels only entails a transfer) that has the effect of substantially lessening or preventing competition, or do not ultimately result in substantially lessened or prevented competition as a result of an adverse effect on a competitor generally do not raise antitrust concerns in Canada.

4. Buyer Power and Conduct

4.1 Horizontal Mergers. When and why does a horizontal merger create buyer power that gives rise to competition policy concerns? Can the creation of buyer power, in particular countervailing power, be a benefit of a horizontal merger?

31. As described above, as to buyer power, the Bureau is generally concerned with horizontal mergers that create, enhance or preserve monopsony power. The Bureau is generally not concerned with bargaining power unless it is an anti-competitive means by which a competitor’s costs are raised such that it is excluded or disciplined with the net result that there is a substantial lessening or prevention of competition in a well-defined market. While such cases are typically considered under the abuse provisions, the Bureau does not rule out the possibility of such a consideration under the merger provisions.

32. The Bureau, when considering countervailing power in the merger context, normally considers it from the perspective of existing entities being able to discipline a possible exercise of market power by merging parties. It does not typically consider the creation of countervailing power through a merger as one of the benefits of that merger.

4.2 Vertical Restraints. What are the implications of buyer power for the competitive analysis of vertical restraints? Does buyer power give rise to competitive concerns regarding buyer-led vertical restraints? What kinds of vertical restraints can, and have, raised concerns because of buyer power? Can buyer power and vertical restraints result in welfare-reducing foreclosure?

33. The vertical cases involving the buying of inputs with which the Bureau typically concerns itself involve the driving up of input prices to competitors by engaging in a practice of anti-competitive acts. Examples of such vertical acts might include the pre-emption of scarce inputs by a variety of means, including exclusive deals with the main suppliers of the relevant input.⁹ In such a way, buyer-led vertical restraints can give rise to competitive concerns. While the Bureau generally acknowledges that differences in the relative costs of inputs to a dominant firm versus that to its rivals might be to the rivals’ disadvantage, the Bureau does not generally consider the achievement of such relative cost differences by way of a dominant firm simply being able to negotiate a lower input price, to be an anti-competitive act.

34. For an act to be considered anti-competitive, it must be found to be for the purpose of an intended negative effect on a competitor that is exclusionary, disciplinary, or predatory. Such acts could include exclusive dealing, bundling, tying and so forth. As noted above, exclusive dealing in the purchase of scarce inputs was the basis of the Nielsen case. In examining the purpose of the act, the Bureau takes into account the fact that a non-monopsony price reduction in input prices can be pro-competitive and so efficiency enhancing.

35. The Bureau is generally of the opinion that buyer power and vertical restraints can result in welfare-reducing foreclosure. While this is the case, it should be noted that the abuse provisions do not provide for an efficiencies defence.

4.3 Predatory Bidding. Predatory bidding might create buyer market power. It involves overbidding for inputs in the short-run to reduce the profitability and induce the exit of competing firms that use the input in order to create monopsony power in the long-run. Under what circumstances is predatory bidding profitable and harmful? What should be the legal standard for a finding of anticompetitive predatory bidding?

36. To date, the Bureau has not investigated any matter involving predatory bidding. While the Bureau has considered matters where the price of a scarce input to competitors was increased, for example, through pre-emption or exclusive dealing, the Bureau has generally been concerned with the consequent impact the exclusion or disciplining of rivals would have on the downstream, rather than upstream, market (which, as noted above, is not to say that downstream effect is strictly necessary for the Bureau to have an issue). As noted above, however, the Bureau is generally concerned with monopsony power and its creation, and so would investigate matters involving allegations of predatory bidding.

37. The Bureau generally considers predatory bidding, as described above (and so ignoring any possible effect in the downstream market), to be profitable if its net effect, upon the exclusion of competitors, is a lower price paid for inputs than that which would otherwise have prevailed absent the bidding by an amount sufficiently large and/or for a period sufficiently long so as to allow the predating bidder to recoup the costs of the higher input prices it paid. If such bidding does not lead to a substantial lessening or prevention of competition by way of created, preserved or enhanced market power downstream, or enhanced market power upstream by way of creation of monopsony power, the bidding would not generally be considered abusive by the Bureau.

4.4 Raising Rivals’ Costs (RRC) Overbuying. RRC Overbuying involves a downstream firm increasing its purchases of an input to raise demand and therefore the price of the input, with the intent of raising the cost of its rivals in the downstream market. Under what circumstances is RRC Overbuying profitable and harmful? What should be the legal standard for a finding of anticompetitive RRC Overbuying?

38. The Bureau considers RRC overbuying to be profitable to the firm engaging in it, if the net effect of the exclusion or the disciplining of rival firms whose costs have been raised is a higher downstream price than that which would have prevailed absent the overbuying by an amount sufficiently large and/or for a period sufficiently long so as to recoup the costs of the higher input price. If having also found that the intended purpose of the overbuying was exclusion or disciplining, rather than some pro-competitive rationale, the Bureau will generally consider created or enhanced downstream market power to be the result of abuse.

4.5 Buyer Groups. Under what circumstances does the formation of agreements, arrangements, and institutions that facilitate collective purchasing by buyers raise antitrust concerns? What should be the legal standard for the determination that a buyer group should be enjoined due to competition policy concerns?

39. Monopsony power and also bargaining power that result from the formation of a cartel will generally raise concerns under the conspiracy provisions of the Canadian Competition Act.

---

Footnotes

¹Cases where the supply curve is perfectly inelastic such that a price decrease below competitive levels does not result in a decrease in output but only a transfer may also give rise to concerns. This scenario should be understood to be generally included in the category of monopsony.

²In such a situation, output would deteriorate not in terms of output, but rather in terms of quality.

³In a hearing regarding the merger of meat rendering companies, the Competition Tribunal, while examining the merger from the selling side, indicated that it could analyse the competitive effects of the merger from the perspective of either a monopsonist or a monopolist, and that no significant difference resulted from the two characterisations (The Director of Investigation and Research v. Hillsdown (Holdings) Canada Ltd., et al., Reasons and Order, Competition Tribunal, March 9, 1992).

⁴See, for example, “Competition Bureau reaches agreement to preserve competition in two B.C. forestry markets”, Media Release, Competition Bureau, Ottawa, December 7, 2004.

⁵See, for example, “Competition Bureau Concludes Inquiry into Snow Crab Processing in Newfoundland and Labrador”, Media Release, Competition Bureau, Ottawa, December 17, 2004.

⁶Which is as defined above and so subject to the note contained in footnote 1.

⁷It may also be the case that buyers who participate in wholly different downstream markets will be relevant purchasers in the upstream market. For example, a buyer who purchases corn for flour production may participate in the same market as a buyer who purchases corn for ethanol production.

⁸Where the supply curve is perfectly inelastic, there will be no allocative inefficiency resulting from a decrease in price below competitive levels. As noted in footnote 1 and elsewhere in this submission, such a situation may still give rise to concerns.

⁹Director of Investigation & Research v. D&B Co. of Canada, Reasons for Order, Competition Tribunal, 1995 (“Nielsen”).