Advocate Ganeral Mazák has issued a call for less aggressive use of predatory pricing in his Opinion of 25 September 2008 relating to the Wanadoo decision where the Commission found that the appellant charged predatory prices below variable costs (AVC) until August 2001, and above AVC but below average total cost (ATC) from August 2001 onwards, as part of a plan to pre-empt the market in high-speed internet access during a key phase in its development. However, I think his views should not be followed by the ECJ.
Essentially the appellant won on three grounds:
1) CFI failed to state reasons when commenting on a dominant firm's right to align prices
2) CFI falied to state reasons why recoupment need not be shown
3) Likely recoupment must be demonstrated by the Commission in predatory pricing cases
In addition the AG also said that in his view a dominant firm should, on occasion, be allowed to align its prices with those of competitors.
Each of these four conclusions can be criticised.
1) Failure to state reasons regarding the dominant firm's right to align prices to competitors.
The passage of the CFI under appeal is this: "WIN cannot therefore rely on an absolute right to align its prices on those of its competitors in order to justify its conduct. Even if alignment of prices by a dominant undertaking on those of its competitors is not in itself abusive or objectionable, it might become so where it is aimed not only at protecting its interests but also at strengthening and abusing its dominant position." 
The AG's criticism was scathing because the CFI did not apply this paragraph to the facts at hand. In his words the CFI "should therefore have assessed whether or not that (new) formulation applied to the facts in the instant case, something which it manifestly did not do." 
With respect, this is wrong. Look at the passage in bold above and the facts of the case. On the facts of the case the CFI ruled that the firm's condct was predatory (below AVC for a period and between AVC and ATC with intention for another period). Therefore the pricing fell squarely withing AKZO [71 and 72]. So the behaviour is in itself abusive. Accordingly there was no need for the CFI to assess the formulation. It is merely an obiter dicta as to what the law is.
2) Failure to state reasons about recoupment
Here the AG notes that in Tetra Pak 2 the ECJ held that proof of likely recoupment was unnecessary "in the circumstances of that particular case" and criticises the CFI for not expaining whether there are also circumstances in this case that allow a finding of predatory pricing without recoupment. [58-60 of the Opinion]
This is a plausible line of attack, but another view can be taken of what the CFI said. It is worth quoting the CFI in France Telecom at length:
226 In relation to the recoupment of losses, the Court of Justice (in Tetra Pak 2) added, in paragraph 44 of that judgment: ‘[I]t would not be appropriate, in the circumstances of the present case, to require in addition proof that Tetra Pak had a realistic chance of recouping its losses. It must be possible to penalise predatory pricing whenever there is a risk that competitors will be eliminated. The Court of First Instance found, at paragraphs 151 and 191 of its judgment, that there was such a risk in this case. The aim pursued, which is to maintain undistorted competition, rules out waiting until such a strategy leads to the actual elimination of competitors.’
227 In line with Community case-law, the Commission was therefore able to regard as abusive prices below average variable costs. In that case, the eliminatory nature of such pricing is presumed. In relation to full costs, the Commission had also to provide evidence that WIN’s predatory pricing formed part of a plan to ‘pre-empt’ the market. In the two situations, it was not necessary to establish in addition proof that WIN had a realistic chance of recouping its losses.
So the reason why in Tetra Pak 2 proof of likely recoupment was unnecessary was because the Commission had proved that there was a risk that a competitor would be eliminated. This means that the risk that a competitor would be eliminated by predatory pricing one relevant circumstance when proof of likely recoupment is unnecessary. And at  the CFI gives reasons in conformity with this because the facts showed that the competitor was put at risk.
Of course you can argue that this is a bad reason. It is bad because it means that as soon as you satisfy the conditions set out in AKZO [71 or 72] then you have shown a risk of elimination and so you never need recoupment. But then perhaps this is what AKZO actually stands for (see below)
3) Likely recoupment in predatory pricing cases
This point is the most relevant for future cases. The AG takes the view that "case-law requires the possibility of recoupment of losses to be proven."  The reasoning is based on the fact that because the ECJ in Tetra Pak 2 said that proof of likely recoupment was unnecessary on the facts of that case, that it must mean that as a general rule proof of likely recoupment is necessary.
I think this is wrong. The bold part of para 44 of Tetra Pak 2 (see above) tells you that the rationale for penalising predtory pricing is to avoid the risk that a competitor is eliminated. Therefore whether the dominant company recoups or not is unnecessary, it is the risk to competitors that is the Court's concern.
The AG says that proof of likely recoupmet is supported by AKZO . Let us consider this passage in full:
"Prices below average variable costs (that is to say, those which vary depending on the quantities produced) by means of which a dominant undertaking seeks to eliminate a competitor must be regarded as abusive. A dominant undertaking has no interest in applying such prices except that of eliminating competitors so as to enable it subsequently to raise its prices by taking advantage of its monopolistic position, since each sale generates a loss, namely the total amount of the fixed costs (that is to say, those which remain constant regardless of the quantities produced) and, at least, part of the variable costs relating to the unit produced. "
The passage in bold would seem to me to indicate that there is a presumption that recoupment is likely if a firm prices below cost. A presumption that does not therefore require the Commission to show likley recoupment. But, perhaps, a presumption that may be rebutted by the dominant firm who could try and say that recoupment is unlikely because, for example, it is selling off old stock.
Nor does Hoffmann La Roche  support the necessity of recoupment, quite the opposite: it supports a finding of abuse whenever the degree of competition in the market is hindered. This occurs once the rival leaves.
One can however feel for the AG because he aligns himself with the views of AG Jacobs in Bronner that the purpose of Art 82 is to protect consumers, not individual competitors. However not all Advocates General believe this (see AG Kokott for instance).
4) Price alignment by a dominant firm
Should a dominant firm be entitle dto align its prices with those of competitors? The plea is inadmissible, but the AG suggests that "one should allow for circumstances where a dominant undertaking is exceptionally permitted to show that its pricing below average variable cost is objectively justified. " 
True, most commentators would say that one such circumstance is when the dominant firm is selling off old stock. The difficult question is whether the dominant firm should do this when it is doing so with the clear intention of damaging a new entrant in a market where it was the incumbent for many years.
The AG is calling for a major change in the law on below cost pricing, by placing a greater focus on the impact on consumer welfare and less on the economic freedom of market participants. The problem is that the case law which seemingly creates the basis for this shift does not support it. AKZO and Tetra Pak 2 are clearly cases where the dominant concern is the preservation of rivalry, of a competitive process. The radical change called for here should be made by advising the Court to change its approach rather than by asking the court to reinterpret its case law.
I hope the ECJ does not follow the AG on the first two points, the CFI gave reasons that can be understood. That they are bad reasons is another matter, so perhaps the ECJ could require proof of likely recoupment. Though it might want to note that the requirement to show this in the US has led to no successfu predatory pricing prosecution as far as I know. Is the ECJ so Chicagoan to want this result?