Treatment of Efficiencies in the Competition Act - Appendix A

Appendix A: What Are Efficiencies In Merger Review?

What follows is a discussion of the economics of efficiencies in merger review. The first part outlines three categories of efficiencies. The second part reviews the implications of efficiencies for welfare standards and merger review methodology. The third part provides a sample of advances where efficiencies can be achieved outside the scope of merger review.

Categories of Efficiency

Mergers can either reduce or increase the overall efficiency of the economy.⁶² There are three main categories of efficiencies: allocative, productive and dynamic. A merger is said to result in "efficiency gains" when a merger enhances productive or dynamic efficiency.

Allocative Efficiency

Allocative efficiency is measured by the degree to which resources available to the economy are allocated to their most valuable use. Mergers that create, enhance or entrench market power create an incentive for firms to profitably increase their prices by reducing output.⁶³ Such an exercise of market power tends to distort prices and, therefore, provides artificial incentives for consumers to limit their consumption, and for producers to restrict their output.⁶⁴ This impedes consumption of goods and services that generate more value in consumption than the real resource cost of supply.⁶⁵

The impact of a merger on allocative efficiency is measured by its impact on the sum of producers' and consumers' surplus (often called the deadweight loss).⁶⁶ The size of the effect depends on how much the merging parties increase prices or otherwise restrict competition. This, in turn, depends on market power and, to some extent, on the effect of the merger on productive and dynamic efficiency.⁶⁷ The deadweight loss effects of a merger include the following:

• decrease in consumer surplus resulting from output reductions;
• decrease in producer surplus resulting from pre-existing market power; and
• any anticipated decrease in consumer and producer surpluses in interrelated markets.⁶⁸

The concept of allocative efficiency applies equally to the price and non-price aspects of competition.⁶⁹ For example, if a merger were likely to cause a product to be withdrawn from the market, the allocative efficiency cost would be the value that consumers place on product variety, minus the real resource savings from the ceasing of the variety.⁷⁰

Productive Efficiency

A merger may also provide an opportunity for a firm to change the way it employs productive inputs (such as labour and capital) to produce output. Productive efficiency is maximized when a given level of output is produced at the minimum real resource costs (sometimes called the minimum opportunity cost).⁷¹

In general, for a merger to improve productive efficiency, it must do the following:

• allow the firm to achieve minimum efficient scale in a manner that cannot be achieved through internal growth (e.g. in a declining market);
• allow a rationalization of output from high-cost to low-cost firms; or
• improve the incentive or ability of the firm's management to minimize costs.

Productive efficiencies include the following:

• product, plant and multi-plant savings in both variable and fixed costs; ⁷²
• savings associated with integrating new activities within the firm;⁷³ and
• savings arising from transferring superior production techniques and know-how from one merging party to the other.⁷⁴

Mergers that prevent or substantially lessen competition can also reduce productive efficiency as resources are dissipated through x-inefficiency and other distortions.⁷⁵ For instance, x-inefficiency may arise when firms are insulated from competitive market pressure to exert maximum efforts to be efficient. There may also be costs such as retooling that must be incurred to achieve the capture of efficiency gains. The overall impact of the merger on productive efficiency is the sum of the gains minus the losses (e.g. x-inefficiency) and other costs.

Gains in productive efficiency may affect either fixed or variable costs. When the effect on productive efficiency is positive, the resulting savings in variable costs are an incentive for the merging parties to reduce price (and/or make their product offering more attractive in other regards). Thus, in addition to saving resources and expanding the economic wealth of the overall economy, such variable cost savings also mitigate any adverse allocative efficiency effects of the merger.⁷⁶ Savings in fixed costs, by definition, are not an incentive to reduce price in the short run. They do, however, reduce average cost and, thus, could have positive effects on allocative efficiency in the long run.⁷⁷

Dynamic Efficiency

Dynamic efficiency is the effect of a merger on the introduction of new products, the development of more efficient productive processes, and the improvement of product quality and service. In particular, dynamic efficiency refers to the efficiency of the framework for decision making over time. For example, dynamic efficiency requires that proper incentives exist to make long-term decisions, such as those about investment and the introduction of new products. Dynamic efficiency also requires that the effects of decisions in one period be taken into account for future periods. The likely effects of a merger on dynamic efficiency are extremely hard to measure, in part because they are not guaranteed to happen⁷⁸ and, when they do, are often not realized for many years. Mergers that prevent or substantially lessen competition may reduce the rate of invention, technological change and dissemination of new technologies, with a resulting opportunity loss of economic surplus.

Gains in dynamic efficiency may affect other categories of efficiencies. If a merger enhances or impedes the efficiency with which new, innovative products are introduced to the market, such new products will mitigate or aggravate the allocative efficiency effects of the merger accordingly.⁷⁹ Similarly, if a merger enhances or impedes the efficiency with which new cost-saving production techniques are introduced to the market, such new processes will mitigate or aggravate the allocative efficiency effects of the merger accordingly.

Welfare Standards and Methodology in Merger Review

The need to consider welfare standards arises because a definition of efficiency gains inherently abstracts from income distribution effects. When positive and negative efficiencies are being "netted out," a potentially adverse income redistribution effect may be ignored.⁸⁰ Thus, a complete analysis must go beyond a pure quantification of efficiency effects and also consider the welfare consequences of income redistribution.

Historically, antitrust practitioners have distinguished between the effects of a merger on allocative efficiency and efficiency gains (e.g. productive and dynamic). This model was reasonably well suited to addressing the economic challenges that faced the Canadian economy in the 1970s and 1980s. At that time, there were concerns that the Canadian tariff had given rise to inefficient Canadian industries.⁸¹ Given the economic situation, the Economic Council of Canada, among others, argued that merger policy should be designed so that it did not impede the capture of economies of scale.⁸² This approach usually equated the effect of an increase in price due to the exercise of market power with the allocative efficiency effect. When an adverse effect could be demonstrated, these losses were weighed against any efficiency gains.⁸³

In recent years, concerns have arisen about the possibility that a merger with a negative effect on allocative efficiency but a larger positive impact through efficiency gains could also adversely affect income distribution.⁸⁴ Three approaches to welfare standards have been proposed for merger review: total surplus, balancing weights and consumer surplus.

Total Surplus

Under the total surplus approach, a merger would be permitted if it could be shown that the total of producers' and consumers' surplus would increase. The total surplus approach is the most permissive of the welfare standards.

The implication of this approach is that mergers that transfer wealth from consumers to shareholders are allowed. As Ross and Winter (2003) demonstrate, the total surplus approach cannot be solidly grounded in economics.⁸⁵ They note that an equal weighting on consumers' surplus and producers' surplus is a value judgment that may be at odds with the general progressiveness of Canadian fiscal policy.

Balancing Weights

Under the balancing weights standard, a merger would be allowed only if it would increase the weighted sum of the producers' surplus and consumers' surplus. This is a "balancing weight" because any standard that recognizes income transfer effects implicitly puts a higher weight on one person's income than another's.⁸⁶

The difficult question is exactly what the "weight" should be.⁸⁷ Ross and Winter (2003) argue that the weight should be aligned with other elements of Canadian fiscal policy. Absent evidence that merger policy is a more effective means of achieving the policy goal of preserving the income of less fortunate Canadians, merger policy should not be more progressive than other aspects of Canadian fiscal policy. Ultimately, the weights should depend on the specifics of the case. In practical terms, the balancing weights approach requires that the total benefits to the Canadian economy exceed the efficiency losses by a margin that reflects the importance of the wealth transfer caused by the price increase, in effect, raising the bar for this type of analysis.

Consumer Surplus

Under the consumer surplus approach, no mergers that would likely increase consumer prices (and therefore reduce consumers' surplus) would be permitted as part of an efficiency gains trade-off.⁸⁸ Efficiency arguments would be permitted only within the analysis of whether the merger would likely prevent or substantially lessen competition.

The advantage of this approach is that it prohibits all mergers that harm low-income Canadians who are consumers but not shareholders. It is also simpler to implement than the balancing weights approach. There is no need to debate what the weights should be: the weight on producers' surplus is zero, regardless of the facts of the case (e.g. whether the good in question is a luxury good or a necessity).

The weakness of the consumer surplus approach is that it ignores information — the mirror image of its advantage. It could be argued that a lower standard (somewhere in between the maximum implied by the consumer surplus standard and the minimum implied by the total surplus standard) is a better policy in view of the relative size of the Canadian economy. Also, it could be argued that there are certain luxury goods for which the wealth transfer effect from consumers to shareholders may be more appropriately treated as neutral (or even positive).

Efficiencies achieved in other dimensions

In the past decade, there has been a revolution in our understanding of how, in imperfectly competitive markets, firms' business strategies (including mergers, joint ventures and various forms of a business' contracting practices and internal organization) can be used strategically. These concepts were firmly incorporated into Industrial Organization and "Game Theory" by numerous economists, most notably John Nash, who, jointly with John Harsanyi and Reinhard Selton won the Nobel Price for this work in 1994. More recent work has built on these technical advances to show how market incentives affect economic behaviour, and thus achieve efficiency, in numerous dimensions. A non-exhaustive list includes:

Network competition and product standardization: Many recent papers have illustrated that the integration or expansion of a network increases the value of the network to individuals already on the network. Product standardization gives rise to similar gains.

Efficient firm organization in light of information and transactions costs: Private information and imperfect contracts can limit the ability of individuals to use arms-length contracts to achieve mutually beneficial arrangements. Thus, it has been shown that direct equity interests, through joint ventures, vertical integration, long-term contracts or mergers may correct for a market failure that would otherwise exist due to transactions costs between arms-length entities.

Efficient use of firm-specific intangible capital: Intangible capital such as a brand name, a reputation for quality, learning-by-doing or know-how generally require large up-front investments, but involve very low marginal costs (or sometimes even zero) when being used. Various forms of equity arrangements, restrictive contracts or licensing arrangements can promote the optimal use of such productive assets.

Innovation and investment in intangible capital: Innovation and investments in intangible capital such as brand names, reputations or know-how, often require very long time-horizons. Technological competition often provides strong incentives to invest and innovate, but at the same time, imperfect appropriability and "all or nothing" innovation races can lead to inefficiencies that can be addressed through mergers or similar transactions.

Investment in physical capital: Investments in physical capital often indivisible (or lumpy). Optimal investment incentives may require a scale that is large relative to the market, and where transactions costs limit the ability of firms to use arms-length contracts, mergers or similar arrangements may promote efficiency.

Specialization of production and multi-plant production: In the presence of imperfect competition, firms' competitive decisions on how much output to produce or mow many varieties to produce may result in insufficient specialization, and mergers or other similar arrangements may promote efficiency.

Diversification and efficient decisions in the presence of risk: Entry decisions generally involve uncertainty and risk, and the ex post outcomes in terms of market structure may be sub-optimal. Thus, mergers or other similar arrangements (including business exit or failure) may promote efficiency. In addition, contractual arrangements (such as through futures markets) may allow for efficient "trade" in risky outcomes.

---

62. Mergers may also lead to income transfers, such as from consumers to shareholders. The analysis in this section first looks at the effect on efficiency and then comments on the effect of potential income transfers.

63. In general, when evaluating the effects of a merger, the primary concerns are often price and output. In some cases, however, effects on other aspects of competition, such as quality, product choice, service and innovation, may also be important. To simplify the discussion, the term price refers in this appendix to all aspects of firms’ actions and includes, for example, an increase in the nominal price as well as a reduction in quality, product choice, service or innovation.

64. In other words, market prices (when, as noted above, price encompasses all relevant dimensions) are an efficient means of aggregating information and coordinating economic decisions. (For a useful discussion of this role of markets, see Fredrick von Hayek, “The Use of Knowledge in Society,” American Economic Review 35 (1945), p. 519.) For example, when the cost of a resource or input increases due to outside factors (e.g. weather, shortage), a resulting change in the price signals producers and consumers to conserve their consumption of this, now more expensive, resource or input. Price increases due to the exercise of market power tend to impede this market signal function and, thereby, distort the allocation of economic resources.

65. This paragraph describes the analytical framework for assessing market power from the perspective of a seller of a product or service. This framework applies equally when assessing the exercise of market power by buyers of a product or service. Buyers’ market power is the ability of a single firm or group of firms to profitably depress prices paid to sellers (for example, by reducing the quantity of inputs purchased) to a level that is below the competitive price (often called monopsony power). The economic effect of an exercise of such power on allocative efficiency is symmetric.

66. The effect on allocative efficiency is a net concept. It represents a quantification of the benefits to consumption in a market (in terms of utility) minus the opportunity cost of the resources required to produce it.

67. Such efficiencies may not only save productive resources, but may also increase the incentive of the merged entity to compete. This would occur, for example, if the productive or dynamic efficiency were to reduce variable costs.

68. When the products purchased by the merging firm include intermediate goods that are used as inputs in other products, price increases in intermediate goods can contribute to allocative inefficiency in other markets.

69. These non-price effects can be difficult to quantify. For example, consider the case in which a merger reduces product choice by eliminating a product that would have been introduced but for the merger. Introducing a new product is economically equivalent to reducing the price from the prohibitive level (at which demand is zero) to the equilibrium level. This is inherently unobservable, from both the quantitative and qualitative perspectives; there is simply no data or qualitative experience with such a price.

70. Assume, for example, that the gross value to consumers and producers generated by the existence of product variety is $100, and the real resource cost of producing it is $30. Pre-merger, the sum of producers’ and consumers’ surplus created by this variety is $70. If the merging parties choose to eliminate this variety, the allocative efficiency effect of this action would be minus $70.

71. Timing differences in the realization of efficiency savings are accounted for by discounting to the present value.

72. Both variable and fixed-cost savings are relevant to the analysis because both generate producer surplus, even though only variable (i.e. marginal) cost savings generate an incentive for the firm to cut prices.

73. These include reduced transaction costs associated with contracting for inputs, distribution and services that were previously performed by third parties.

74. Such improvements in the efficiency with which superior production techniques and know-how are diffused throughout the industry can be looked at as either productive efficiency or dynamic efficiency. Productive efficiency tends to focus on the impact of the transfer of the superior production techniques and know-how on the firm’s cost, while dynamic efficiency tends to focus on the incentive to allocate the efficient amount of resources to create the superior production techniques and know-how, and/or adapt for application, or otherwise make it accessible.

75. X-inefficiency typically refers to the difference between the maximum (or theoretical) efficiency achievable by a firm and the actual efficiency attained.

76. The size of the effect on allocative efficiency depends on how much the merging parties increase prices (or otherwise restrict competition). The effect of any productive efficiency that reduces variable costs must be incorporated into this analysis.

77. Not all reductions in fixed costs qualify as productive efficiencies. For example, if a merger in an industry in which firms require specialized capital equipment that has no value in other uses allows the firms to employ one rather than two such machines, the elimination of this duplication gives rise to an efficiency gain of zero. More generally, under the economic definition of efficiencies, the value of any resources saved through merger must be evaluated at their opportunity cost.

78. That is, a merger that affects dynamic efficiency often means that the probability that innovation will occur will be different with the merger than without it. Evaluating the value of such changes to consumers depends on, for example, the degree of risk aversion.

79. The exercise of market power provides an incentive for the firm to produce less output with a given innovation and therefore reduces the net benefit the innovation produces for the overall economy. At the same time, however, a merger may promote dynamic efficiency when it allows the merged entity to improve the ability of firms to “appropriate” a return that reflects the social benefits generated by dynamically efficient activities, avoid duplication generated by the all-or-nothing nature of some innovation “races,” or achieve the necessary scale to achieve minimum research and development costs.

80. For example, a merger that adversely affects allocative efficiency by $100 but allows efficiency gains of $120 has the same net welfare effect as a merger that adversely affects allocative efficiency by $20 and allows no efficiency gains. These two mergers may, however, have very different effects on income.

81. In particular, H. C. Eastman and S. Stykolt, The Tariff and Competition in Canada (Toronto: MacMillan of Canada, 1967), argued that weak competition law together with tariff protection allowed Canadian firms to raise prices above competitive levels to slightly less than the tariff-inclusive price and thereby extract monopoly profits. Such prices encouraged inefficient firms to enter the market and caused them and others to operate at an inefficiently small scale.

82. It is important to note that, at the same time, liberalization of international trade enhanced the role of foreign competition. Thus, the adverse effects on allocative efficiency due to increases in concentration in the Canadian market would likely be temporary, since trade liberalization would likely expand geographic markets in the future.

83. It was usually assumed that the effects on productive and dynamic efficiency were positive.

84. In general, any case in which efficiency losses of one type are weighed against efficiency gains of another type can have implicit income redistribution effects. For example, a merger that reduces allocative efficiency (by increasing price) or productive efficiency (by increasing cost) but also increases dynamic efficiency by leading to increased innovation may redistribute income from current consumers to younger consumers who may not be consuming today but will demand new, innovative products in the future.

85. Tom Ross and Ralph Winter, “Canadian Merger Policy Following Superior Propane”, Canadian Competition Record (Summer 2003), p. 7.

86. Rejecting the total surplus approach (which puts equal weight on the income of all people in the economy) unavoidably means accepting some other set of weights on the income of individuals in society. Any welfare standard is inherently arbitrary (Arrow Impossibility Theorem). However, the decision to allow or block a merger has these effects; thus, merger policy implicitly embodies some weighting.

87. This is a difficult question, since it requires considerable discussion and thought to determine the correct way to establish the particular weight (e.g. it would require detailed guidelines).

88. Maximizing consumer surplus is equivalent to assessing mergers based on their likely impact on prices. Thus, these standards are discussed together.