Predatory pricing requires a dominant supplier to incur a profit sacrifice to eliminate a competitor. Prices that imply a sacrifice if applied in the long run (i.e. below long run average incremental costs) may not imply a sacrifice in the short run (i.e. above average variable costs).

This raises the question whether prices in this range actually imply a profit sacrifice at all. This in turn challenges the notion that such prices imply predation in the presence of predatory intent. This is because without profit sacrifice the economic evidence does not corroborate that the dominant firm in fact acted on such intent.

In this article, Lau Nilausen, Senior Director at FTI Consulting, discusses the need for a more robust approach to assessing predatory pricing than the traditional reliance on standard cost measures and evidence of predatory intent.

Reconnecting to the logic behind predation theory

europe has developed extensive case precedent for the assessment of predatory pricing over the past two decades. prices either have to be below average variable costs or below average total costs supported by intent to foreclose to imply predation. The logic is that such prices demonstrate or corroborate that a dominant firm's actions are rational only if they result in the elimination of a competitor. This article challenges this logic with a focus on the rationality of prices in the range between average variable and average total costs.

The discussion below is structured as follows. First, a very brief summary of the principles established in european case precedent for finding predation is provided. second, the proper interpretation of the long run average incremental cost concept vis--vis average total cost is explored. Third, the question of whether prices below average total costs imply profit sacrifice is assessed. Finally, the implications for finding predation based on prices below average total costs and predatory intent are set out.

1. AKZO, Deutsche Post, and the rest

To find predation it has to be the case that the dominant undertaking incurs a short term profit sacrifice to foreclose a competitor.1 Case precedent has established that prices below average variable costs (AVC), average total costs (ATC), and long run average incremental costs (LRAIC) may constitute a profit sacrifice. These cost measures are introduced briefly below as a basis for the subsequent discussion.

The European Court of Justice (ECJ) established in AKZO v Commission that prices below AVC "must be regarded as abusive" as "A dominant undertaking has no interest in applying such prices except that of eliminating competitors".2 A cost is variable only if it changes in response to variations in the quantity produced.3

Neither case precedent nor the Commission's guidelines have discussed the economic merits of including common costs in ATC.

In the same judgment, the ECJ stated that prices above AVC but below ATC "must be regarded as abusive if they are determined as part of a plan for eliminating a competitor".4 Such plans are referred to as predatory intent.

In Deutsche Post, the European Commission (the Commission) found that "only the additional costs of providing a particular service vary with volume produced".5 Whereas the judgment in AKZO v Commission required a cost to be sensitive to changes in volumes to be variable, a cost is treated as incremental in the Deutsche Post decision if it varies with the provision of the entire service in question. This is equivalent to LRAIC.

Following Deutsche Post, the LRAIC concept has been applied across a number of European cases.6

2. LRAIC is equivalent to ATC

By definition, LRAIC is "the difference between the total costs incurred by the firm when producing all products, including the individual product under analysis, and the total costs of the firm when the output of the individual product is set equal to zero, holding the output of all other products fixed".7 This includes both variable and fixed costs. For a company producing a single product, LRAIC is therefore self-evidently equal to ATC.

Multiproduct firms may incur costs that can only be avoided if the company ceases supply of all products. These so-called common costs are not incremental to any specific service and therefore not included in LRAIC. This raises the question whether common costs should be included in the ATC measure established by the ECJ.

The Commission states in its guidelines that "If multi-product undertakings have economies of scope, LRAIC would be below ATC for each individual product, as true common costs are not taken into account in LRAIC".8 However, neither case precedent nor the Commission's guidelines have discussed the economic merits of including common costs in ATC. Doing so would in fact represent a break from the logic supporting all precedent since AKZO v Commission. It would also represent significant conceptual challenges. Specifically:

1) prices above LRAIC imply returns on investment in excess of the cost of capital for that product. It is therefore not the case that a firm "has no interest in applying such prices except that of eliminating competitors",9 "would have been unable to offer its own retail services without incurring a loss",10 or incurs "losses that could have been avoided",11 as is the logic underpinning case precedent and the Commission's guidelines. Prices above LRAIC hence do not imply a profit sacrifice as required to find predation;

2) the amount of common costs allocated to any individual product depends on how much can be allocated to other products and thereby on the performance of all such other products offered by the firm. Poor commercial performance by one product may thereby increase the allocation of common costs to all other products. This could lead to a counterintuitive finding of predation in one market as the direct result of a potential failure by the firm to sell sufficient volumes in unrelated markets; and

3) by definition, "the allocation of common costs may not be related to the relative use of common cost by the separate products or services, which could make the allocation rather arbitrary".12 This introduces subjectivity into what should be an objective economic assessment. Whether a company is predating or not cannot depend on which of several equally reasonable allocation methods is applied to allocate costs not incremental to that product.

For these reasons, there is no economic merit to including in ATC any costs in addition to LRAIC. Rather, to do so would contradict the Commission's stated requirement that predation "always entails a sacrifice".13 ATC therefore must equal LRAIC to be economically meaningful and consistent with the logic reflected in case precedent.

3. Do prices below LRAIC imply profit sacrifice?

There are therefore two relevant cost boundaries for assessing profit sacrifice:

1) AVC on the basis that the dominant firm would be better off not selling a specific volume, even in the short run, than doing so at prices below this level; and

2) LRAIC on the basis that the dominant firm would be better off completely withdrawing from the market than continuing to supply at prices below this level in the long run.

The different cost measures assess whether prices imply short run or long run profit sacrifice. Prices below LRAIC but above AVC imply a long run profit sacrifice even though there is no short run profit sacrifice. This raises the question whether such prices actually can be said to imply a profit sacrifice.

To answer this question it is helpful to return to the statement in the Commission's guidance that profit sacrifice implies "losses that could have been avoided".14 For example, LRAIC may include sunk costs. Failure to recover sunk (i.e. unavoidable) costs does not per se imply "losses that could have been avoided". This is also reflected in the Commission's recognition that fixed (and therefore a fortiori sunk) costs have a "less direct" impact on pricing incentives.15 LRAIC may therefore be too high a benchmark for the assessment of profit sacrifice in the presence of sunk costs.

However, the cost of an asset is not sunk at the time of reinvestment. It would therefore not seem rational for the dominant firm to invest whilst setting prices that would not provide a sufficient return on that investment. This suggests that it may be appropriate to include ongoing investments in the assessment of profit sacrifice. This would be consistent with the use of LRAIC in European broadband predation cases for which investment coincided with the alleged predatory conduct.

Prices below LRAIC but above AVC imply a long run profit sacrifice even though there is no short run profit sacrifice.

Prices below LRAIC may hence not imply profit sacrifice as the dominant firm may be better off making the associated sales than not doing so. On the other hand, prices above AVC may imply a profit sacrifice if they do not justify concurrent investments or other ongoing incremental fixed cash cost. Whether prices between AVC and LRAIC imply a profit sacrifice therefore depends on the specific circumstances of the firm in question.

4. Misinterpreting intent

As noted above, the ECJ found that prices above AVC but below ATC "must be regarded as abusive if they are determined as part of a plan for eliminating a competitor".16 Under this framework, the presence of predatory intent hence makes the difference between finding predation or not (given that the economic evidence does not allow to make this distinction). Prices between AVC and LRAIC thereby only serve as confirmation that the dominant firm acted on such intent.

However, as just explained above, prices in this range may not imply any profit sacrifice. But in the absence of profit sacrifice the firm's conduct would not in fact confirm that the firm acted on any predatory intent. Mechanistically assessing whether prices are below ATC with intent to foreclose hence implies a risk of finding predation without support in economic evidence. This suggests a need to return to the economic substance of predation testing: whether the dominant firm's pricing actually implies a profit sacrifice.