LEGAL AND SUBSTANTIVE STANDARDS IN COMPETITION LAW ENFORCEMENT: RELATIONSHIPS AND JURISDICTIONAL VARIATIONS

This article reviews recent literature on legal standards and substantive standards in competition law enforcement, with particular emphasis on the relationship between these two fundamental concepts, specifically on how the adoption of a specific substantive standard may affect the legal standards utilized by the courts and by competition authorities. We also examine the substantive and legal standards adopted in different jurisdictions, focusing on the EU (specifically, the EC) and the US, and discuss why legal standards are different in different countries.


INTRODUCTION
This article examines recent literature on legal standards (LS's) and substantive standards (SS's), with particular emphasis on the relationship between these two fundamental concepts and specifically on how the adoption of a specific SS may affect the LS's utilized by the courts and by competition authorities (CA's).
We start with the definitions of these two concepts: • the substantive (or liability) standard is the criterion for reaching a decision (by a CA or the court) on whether or not there is violation of competition law; • the legal standard (or decision rule) describes how decisions are reached: how do we show that the criterion (or SS) is actually satisfied -for specific conduct in a specific case?
Generally speaking, we can show whether or not the criterion is satisfied either through an inference of the effect of the conduct in the specific case, from the presumed effect of a more general population of cases (in which population we believe the specific case belongs), or we can rely on an investigation of the effect of the specific conduct in the specific case.
SS's and LS's are laid down in competition law and by the courts interpreting the law (i.e. by case law), and they evolve and are influenced by a multitude of factors, the most important of them being: 1 -socio-political/cultural factors/ideology; legal tradition; -historical/economic development; and -what economic theory suggests with regard to the effects of different types of business conduct and the optimal standards that should be adopted if the objective is to minimize decision errors or maximize social welfare.
In the next section we provide a description of the main types of LS's and SS's that are discussed in the relevant literature on competition law enforcement. Then in Section 3 we examine the SS's adopted in different jurisdictions -with emphasis on the EU, specifically the European Commission (EC) 2 and the US. In sections 4 and 5 we consider optimal (social welfare maximizing) LS's and the LS's adopted in assessing practices in different countries. Section 6 deals with the relation between SS's and LS's with focus on discussing and explaining why LS are different in different countries. Section 7 offers concluding remarks.

Types of SS's
We can distinguish between the following two broad types of SS's: (i) Welfarist SS's, which assume that the criterion for reaching decisions is consumer surplus or economic efficiency; this is usually assumed in academic discussions by economists, but is probably rarely used in practice.
(ii) Non-welfarist SS's, which can be categorized as: 1) SS's that are just one in a continuum of criteria that need to be examined in order to form a judgment about the ultimate criterion of welfare, e.g. the criterion of monopoly power or the criterion of an exclusionary effect/"disadvantaging rivals" (or "distorting the competitive process"); and 2) Other non-welfarist SS's related to "public interest concerns", e.g. concerns of equity, or competitiveness promoted through industrial policy, employment, etc. (see, for example, Baker, Salop 2016;Fox, Sullivan 1987;and Gal, Fox 2014).

Types of LS's
Two (broadly defined) types of LS's can be distinguished: -Per Se (or object-based), where the decision in a specific investigation of a given conduct is reached on the basis of a general presumption about the impact of a general class of conducts, within which we must establish that the investigated conduct falls. 3 Beyond placing the conduct in the general class, by virtue of its characteristics, no further investigation is undertaken, e.g. to account for the context within which the behavior took place or to account for the situation that would emerge without the conduct. -Effects-based (EB, or rule-of-reason), where the decision is made after pursuing an investigation and assessment of the specific case and establishing the impact, in this specific case, on whatever liability criterion is used 4 .
The idea of a continuum from Per Se ("low") to Effects-based ("high") legal standards has gained popularity in recent years. As Jones and Kovacic (2017) note "the general progression in US doctrine has been toward recognition of an analytical continuum whose boundaries are set, respectively, by categorical rules of condemnation (per se illegality) or acquittal (per se legality) and an elaborate, fact-intensive assessment of reasonableness (Rule of Reason, RoR). These poles are connected by a range of intermediate tests that seek to combine some of the clarity and 3 Our definition is close to that in Hovenkamp (2017): "Correct application of the per se rule depends critically on a judgement that certain practices are unreasonable as a 'class' or family group. As a result, condemnation requires that they be correctly placed within that group" (p. 42). 4 The distinction, sometimes used by legal experts, between standards (like EB LS) and rules (like Per Se LS) is not used here. economy of bright-line rules with the greater analytical accuracy that a fuller examination of evidence can produce." 5 In order to locate different LS's along the continuum, it is useful to consider the additional analysis that is required as we move from a "low" (Per Se) to a "higher" (EB) legal standard. Specifically, the progression towards full effects-based legal standards requires that additional "blocks" or components of economic analysis be applied. These components are associated with: 6 1. the definition of the relevant market; 2. the assessment of extant market power; 3. the assessment of whether market power raising or exclusionary effects are present; 4. the articulation of a theory of harm to consumers; 5. the assessment of the efficiency effects on consumers and determination of the net effect of the conduct on consumers; and 6. the assessment of all potential efficiency effects and of what is, ultimately, the total welfare impact of the conduct.
Clearly, which of these blocks of analysis are taken into account depends on the LS adopted. For example, under Strict Per Se, only the first block needs to be taken into account while under full EB or rule of reason, all six (or at least, five) blocks (depending on the SS) need to be taken into account. If the substantive criterion in the assessment of a conduct is whether or not there is an adverse effect on consumer welfare, clearly, the economic analysis under block 6 is not relevant (i.e. is not needed), while block 6 is also needed when the SS is that of total welfare. Of course, if the SS was non-welfarist, e.g. if it was that of "nondisadvantaging rivals", even the 4 th block of analysis is not needed.  Gavil (2008) and Hovenkamp (2017). Alexander (2013) considered Justice Stevens as probably the first to point out that one should think of LS (for dealing with restraints under US Section 1) as forming a continuum with Per Se and Rule of Reason being at the opposite ends of this continuum. As Italianer notes, the US Supreme Court has explicitly recognized that "the categories of analysis cannot pigeonholed into terms like 'per se' or... 'rule of reason'. No categorical line can be drawn between them. Instead, what is required is a situational analysis moving along what the Court referred to as a 'sliding scale'". 6 The components can be identified by analyzing the documents on particular decisions made by a competition authority. Empirical analysis on the adoption of LS has been undertaken using this procedure (see Katsoulacos, Avdasheva, Golovaneva (2018) for a detailed description of the methodology and references to empirical work).

Components of economic analysis applied in assessment
Legal Standard Under the Strict Per Se (SPS) LS the CA makes decisions only on the basis of the purely formal characteristics of the conduct under investigation, relying on strong presumptions about the implications of the general class of conducts to which the specific conduct belongs regarding welfare (or, more generally, regarding whichever liability criterion is used). Alternatively, one can say that under a SPS LS the CA makes inferences about the effects from the formal characteristics of the conduct.
The Modified Per Se (MPS) LS can be considered as a Per Se rule subject to a Significant Market Power requirement or, more generally, as supplementing Per Se by undertaking an analysis of market characteristics, for example, when assessing conducts under abuse of dominance or in an information exchange agreement or in a concerted practice for which there is no strong, hard evidence of collusion. Alternatively, one can say that under MPS LS the CA makes inferences about the effects from the formal characteristics of the conduct, detailed analysis of market characteristics and, depending on the type of conduct, the implications of these on incentives for achieving sustainable collusion and/or on the size of extant market power.
Truncated Effects Based (TEB) LS is a higher LS, under which decisions about whether there is liability in the case of a specific conduct are reached by establishing that the characteristics of the specific conduct and of the market in which it is undertaken are such that it belongs to a class of conducts that distort the competitive process by disadvantaging rivals (i.e. through exclusionary effects, widely defined) or by enhancing market power (as in a concerted practice case) and, assuming a welfarist substantive standard, by establishing that the conditions present are such that a strong presumption can be made of adverse welfare effects. 7 Alternatively, one can say that under a TEB LS the CA decides that there is liability by inferring adverse welfare effects from the potential of the conduct to distort the competitive process by disadvantaging rivals (i.e. through exclusionary effects, widely defined) or by enhancing market power (as in a concerted practice case).
The inclusion of analysis 4 (recognizing factors that affect whether exclusion, reduces consumer welfare (before taking account of efficiencies) leads to an intermediate LS (between truncated and full effects-based).
Finally, Full Effects Based (FEB) represents the LS under which a finding of liability relies on all potential anticompetitive (exclusionary or market power enhancing) and all potential pro-competitive (efficiency) effects of the specific conduct being assessed for the specific case and compared and showing adverse effects on consumer welfare (if the SS is that of consumer welfare) or on total welfare (if the SS is that of total welfare).

The SS in the EU
In the EU, under the influence of a strong Ordo-Liberal tradition, 8 the SS adopted by the Courts is not welfarist; it is that of the impact of the investigated conduct on rivals (or, the impact on the "competitive process", or impact on "consumer choice"). A number of articles and recent decisions confirm this (see Korah 2010;Blair, Sokol 2012;Gifford, Kurdle 2015;Jones, Kovacic 2017;Sokol 2017). The recent Intel (2014) decision provides a very explicit confirmation of this (see Katsoulacos 2019a). In this, the General Court (GC) explicitly rejected the idea that the criterion of "non-disadvantaging rivals" is not the main criterion governing adjudication in the EU.

The SS in the US
In the US, prior to the late 1970s, there were many "goals" including, importantly, that of protecting the competitive process. Since the end of 1970s however, the US Courts have accepted the view that antitrust law is a "consumer welfare prescription" (Jones, Kovacic 2017) and therefore a showing of liability requires that consumer welfare be reduced (e.g. because of an increase in price without any counteracting benefits to consumers such as improved service quality). It is worth noting that there have recently been quite a few voices that have argued that this should be changed and that the emphasis should return to the protection of the competitive process (see, for example, Werden, Froeb 2018; Wu 2018).

The SS in Other Countries
Welfarist SS are adopted in a number of other countries and in recent years some have also been moving towards a total welfare SS (e.g. Canada, Australia) where the liability criterion is that of total welfare, rather than consumer welfare. This requires taking into account a wider range of potential efficiencies under a FEB LS than in the case of the consumer welfare SS.
Non-welfarist SS's related to public interest concerns are popular in developing countries and BRICS, but they are recently also becoming more popular in advanced/mature jurisdictions (see, for example, Baker, Salop 2016;Fox 2018 Recent literature on optimal legal standards (see Katoulacos, Ulph 2009; Katoulacos, Ulph 2015; Katoulacos, Ulph 2016) assumes a welfarist SS, and adopts a welfare maximization approach (which generalizes the traditional decision-theoretic approach) taking into account both decision errors and deterrence effects (and legal uncertainty). The main result emerging from this literature is the following: shifting from Per Se to effects-based legal standards for presumptively illegal (resp. legal) conducts will be optimal (and improve welfare by reducing decisions errors and adverse deterrence effects) if: -The presumption of illegality (resp. legality) is not too high; 9 -The discriminatory quality of the economic models (in terms of their ability to distinguish between harmful and benign cases) is sufficiently high.
The presumption of illegality is extremely high for certain (hardcore) horizontal agreements (e.g. price fixing, market sharing, bid-rigging) and this is the reason for the essentially universal unanimity of using a Per Se Illegality LS for these conducts. It is important to understand that for such conduct this (Per Se) LS is indeed the LS that maximizes social welfare.
But what about other conduct types? Developments in economic theory (specifically in industrial organization) and evidence suggest that the two conditions mentioned above hold for a large number of other conducts in abuse of dominance and vertical constraints, and consequently, for these conducts, EB LS's seem appropriate (i.e. they are optimal, from the standpoint of social welfare maximization). Furthermore, legal uncertainty issues may not reverse this conclusion (Katsoulacos, Ulph 2015;. These developments in economic theory have been an important factor responsible for a move towards more EB LS's, for vertical restraints and abuse of dominance practices, in the US in the last 25 years. As noted in Gavil (2008), after the Sylvania decision "the Court systematically went about the task of dismantling many of the per se rules it had created in the prior fifty years, and increasingly turned to modern economic theory to inform its interpretation and application of the Sherman Act." But this move has not been universal and it has not been followed in the EU (including most of its member states) and many other countries. We need to provide an explanation, as this observation suggests that the LS's adopted in practice in many countries will not maximize social welfare. First, let we take a closer look at the LS's actually adopted in the EU and the US.

LEGAL STANDARDS IN US AND EU ENFORCEMENT
Below we discuss the LS's adopted in the US and the EU for the two main categories of conduct other than horizontal agreements (which are universally treated as Per Se): vertical restraints and abuse of dominance practices.

Vertical Restraints -US
The US makes a vertical restraint lawful if consumers benefit (therefore the consumer welfare SS is used, as previously mentioned) after investigating all the implications (on consumer welfare) of the specific restraint (therefore, the rule of reason LS or FEB is used). For example, resale price maintenance can raise prices to consumers but not be prohibited if the defendants can show there are non-price benefits to consumers (such as retailers providing better service) which means consumers are not harmed on net.
Even if exclusive dealing is shown to significantly foreclose rivals, plaintiffs must go further and establish that it is detrimental to consumers, such as higher prices or reduced product variety. With regards to customer and territorial restraints, anticompetitive concerns are raised only if those restraints impair inter-brand competition (as opposed to competition among retailers of the same brand) and it is then shown that consumers are worse off.
To conclude, in the US it is required that a vertical restraint is shown to harm consumers regardless of how much it might foreclose the market to competitors or raise prices to consumers.

Vertical Restraints -EU
Under EU law, Article 101(1), the European Commission (EC) must establish that the vertical agreement has as its "object or effect, the prevention, restriction or distortion of competition within the common market." If it establishes that claim, the defendants can turn to justifying that it is procompetitive by drawing on Article 101(3), and showing that there are efficiencies to offset the anticompetitive effects and that consumers receive their "fair share" so as not to be harmed.
Despite its apparent similarity to the US approach, there is a significant difference. As mentioned "This framework would approximate the US rule of reason if the Commission's burden... were to show likely adverse effects on consumer welfare. This, however, does not appear to be the case. The Commission's burden does not require an analysis of competitive effects of the sort undertaken in the US. Rather, EU case law suggests that it is enough for the EC to show that the agreement in question restricted the 'economic freedom' of either a party to the agreement or a third party, without regard to a likely effect on prices, output, or consumer welfare generally." (Cooper et al. 2005) That is, the SS concerns "impact on the competitive process". In turn, depending on the type of the restraint, the LS is either a MPS Illegality or a TEB Illegality LS. As we indicate below, the non-welfarist SS adopted in the EU can be considered an important factor responsible for this.

Abuse of Dominance (AoD) Practices -US
With regard to legal standards in assessing AoD cases, since the end of the 1970s there have been two competing schools in the US: the Chicago School, arguing for MPS (or TEB) legality, and the Post-Chicago School, arguing for FEB or Rule of Reason (see, for example, Evans, Padilla 2005). Especially, perhaps, in the hi-tech/innovation intensive markets, the MPS legality prescription has been winning, despite the growing voices that things should change.

AoD Practices -EU
As noted in Geradin, Petit (2010), under a presumption of illegality, the assessment of AoD cases in the EU has relied on "old, formalistic legal appraisal standards, and (has shown) a reluctance to endorse a modern economic approach." In another review, Neven (2006) reached the same conclusions. In Ibanez (2016), the extensive review of the European Courts' choice of legal standard in AoD cases shows that, for a large number of practices associated with such cases, the standard is one of Modified Per Se Illegality while for the rest the legal standard is the so-called truncated effect-based LS (which certainly falls short of the full effects-based). However, a more recent detailed empirical analysis of all the EC antitrust decisions, looking at the LSs implicit in the economic analysis utilized by the EC CA (DGCOMP), shows that while these are findings are confirmed on average for a period of over 20 years (from 1992 -2016), there is a systematic move towards full effects-based in AoD cases in recent years.
A summary:

Explaining Different Presumptions in the EU and the US
AoD practices, conducted by firms with a dominant position, is considered presumptively illegal in EU, while they are considered presumptively legal in the US. Both positions can be correct, if we take into account the differences in SS in the two jurisdictions. Conduct that is considered on average to have exclusionary effects (and as such is considered presumptively illegal under the EU SS) can simultaneously be considered, on average, not to reduce consumer welfare (and as such is considered presumptively legal under the US SS).

Explaining Different Legal Standards
While normative economic theory examines the choice of legal standards to maximize social welfare, empirical observations are not always consistent with this theory's predictions, as discussed above. According to the recently developed positive theory of the choice of LS's (see Katsoulacos, 2019b), these are adopted by the CA in anticipation of the standards that are adopted by the appeal courts, taking into account reputation-related and cost concerns of the CA. Thus, in this approach, one needs to examine the factors that influence the courts' choice of LS's, an important one being the courts' SS. Recent analysis of the relationship between legal and substantive standards shows the important influence of the latter on the choice of the former. Specifically, it is explained why the adoption of non-welfarist substantive standards makes more likely the adoption of Per Se LS's for presumptively illegal conducts (Katsoulacos 2019a).

The Relation Between SS's and LS's
Two recent articles (Katsoulacos 2019a, andKatsoulacos 2019b) examine the relation between SS's and LS's-specifically how the former influences the choice of the latter. The analysis is used to shed light on the question of the factors that determine the differences in the choice of LS's in different jurisdictions. It is shown that a very important factor behind these differences can be the fact that, in different countries, the courts do not adopt the same SS's. Indeed, it is argued that this may be one of the main factors behind the most significant remaining divergence in antitrust enforcement in the EU and North America, which is that the former continues to use Per Se or "object-based" LS's to a much greater extent.
As noted above, in the EU, under the influence of a strong Ordo-Liberal tradition, the SS is not welfarist; it is related to the impact of the investigated conduct on rivals (or, impact on the "competitive process"). Generally, when a court would consider it best to use a nonwelfarist SS such as that of not-distorting the competitive process or not-disadvantaging rivals), it would be more likely to consider conducts as presumptively illegal rather than presumptively legal. Also, for the former, the use of this non-welfarist SS will lower the LS adopted by the court closer to Per Se, by increasing the strength of the presumption of illegality, since the probability that a conduct is exclusionary (i.e. that it violates the liability criterion of exclusion) is higher than the probability that it is consumer-welfare reducing (i.e. that it violates the liability criterion of consumer harm). Anticipating this, a utility-maximizing CA (motivated by reputation and cost concerns) will adopt (at best) the same (low) LS, or an even lower LS than the court (Katsoulacos, 2019b). As a consequence, this will significantly reduce the extent to which economic analysis will be relied upon in determining liability in any specific case.
More specifically, Katsoulacos (2019a) illustrates that welfarist SS's should not be confused with EB (rule-of-reason) LS's, as is commonly done in literature. We can have EB LS's without welfarist SS's and we can have Per Se LS's with welfarist SS's. A characteristic example of the misunderstandings is the exchange in Wils (2014) and Rey, Venit (2015), concerning the EC's Intel case. The latter criticizes the EC and GC "for not using (full) effects-based" (meaning FEB under an welfarist SS), while the former commends the EC and GC "for using the (appropriate) effects-based" (associated with a non-disadvantaging rivals SS): Wils (2014) was right, in the sense that given the non-welfarist SS adopted by the GC, it did not even make sense to use a full effects-based LS (see also Peepercorn 2015).
In the case of Intel, if under an welfarist SS the socially-optimal LS is FEB, with a disadvantaging-rivals SS (as the one applied by EU Courts), the LS adopted is likely to be even lower than TEB, i.e. it is likely to be the MPS LS -which is in fact the LS adopted by the GC. The GC's LS choice is completely rational given its objective (which is essentially to protect rival firms). In summary, given the SS of EU courts, it is natural to find that conduct by dominant firms is considered presumptively illegal and is often assessed using the MPS Illegality LS. If the courts were to switch to welfarist objectives, in AoD cases the LS's are likely to move closer to FEB.

Explaining Avoidance of FEB in AoD Cases in Both Jurisdictions
It is important to observe that in a number of recent hi-tech AoD cases (such as those involving Google and Intel) both the EU 10 and the US jurisdictions avoided the adoption of FEB LS's. To explain this we start by noting that the discriminatory quality of economic models and economic evidence in AoD cases is lower than that of mergers and vertical restraints. This fact makes such modeling and related evidence much more likely to be rejected (and decisions reversed or annulled). This constitutes a very serious disincentive for the use of FEB by the CA in AoD cases. Agencies (and in the case of the EC, more specifically, the EC'c Legal Service) are making decisions by anticipating the courts' choices and minimizing the risk of annulment (thus avoiding adverse reputation effects). The intensive use of economic modeling and evidence entailed by using FEB can increase the disputability of decisions and consequently it is expected, ceteris paribus, to increase the rate of annulment by appeal courts.
There is some evidence of this last effect. Neven (2006) looks at all the appeals against EC decisions during the 1994-2006 period, and computes the proportion of cases in which the EC prevailed (so decisions were not annulled). He determines a 98% success rate of Art. 82 (now Art. 102) AoD decisions that "have remained focused on form", which "is striking". The proportion of mergers and Art. 81 (now Art. 101) cases that focus on the effects is much lower -75%. Neven considers this "probably the most important insight of (his) findings". The evidence presented by Neven is not however confirmed when a much larger/updated dataset (that considers EC decisions until 2016) is examined in order to identify whether moving closer to FEB in antitrust decisions by the EC's CA (DGCOMP) increases the rate of annulment (Katsoulacos, Makri, 2020) .
Thus we have: -In the EU, for infringement, the courts need a showing of only exclusionary potential, with conduct considered presumptively illegal. Given this, it is natural for the EC to adopt MPS (or TEB) Illegality LS's. -In the US, for infringement the courts require a showing of reduction in consumer welfare, with conduct considered presumptively legal. Given this, agencies attempting to show infringement through a FEB LS would be too risky (and more costly), so they adopt MPS (or TEB) Legality.
The US Google case perhaps best illustrates the situation in the US. Concluding its 2013 investigation the FTC noted that "the law protects competition, not competitors" and that there was not enough factual evidence of alleged "search bias" to support a complaint. Interpreting the outcome, Kovacic (2018) states that "Supreme Court rulings have given dominant firms more freedom [than in the EU] to control pricing, product development and marketing" noting that "I believe that the crucial factor that led the FTC to back down was its perception that Google ultimately would prevail in the Courts. If the US doctrine resembled the EU antitrust doctrine regarding dominant firms, there is a strong chance that the FTC would have brought its own case." 6.5. Will the ECJ Intel Decision Change the Situation in the EU?
The European Court of Justice (ECJ) decision 11 on Intel represents a move in the right direction for the EU. The ECJ decision does ask the GC to move away from MPS Illegality, towards more effects-based illegality, whereby it must be established that rivals are indeed disadvantaged, by undertaking an As Efficient Competitor investigation for the specific case (and not only relying on inference). But this is certainly not a request for FEB, which takes seriously into account efficiencies and undertakes a balancing test, given that the ECJ, as the GC, also adopts the "nondisadvantaging rivals" SS under which there is no scope for a FEB LS.
6.6. The Importance of Innovation in the Choice of SS and LS It has recently been stressed by many authors that enforcement in innovation-intensive markets must be re-calibrated to preserve the incentive to innovate (by placing more weight on dynamic competition over static competition). This is right. It has also been suggested by some authors that the situation could be improved by abandoning the consumer welfare SS and switching to a "protection of the competitive process" SS. For example, Wu (2018) states that "this small change... would make antitrust far more attentive to dynamic harms." The suggested switch in the SS, could shift the presumption from legality to illegality or, if the presumption is one of illegality, it will increase the strength of the presumption of illegality. In either case, the impact will be to make it even more likely to adopt MPS or TEB Illegality LS's, rather than FEB LS's. From a welfare-maximization (or decision-error minimization) perspective, this suggestion would be sensible if one accepts that exclusionary conduct against new highly-innovative firms is highly likely to be harmful to welfare in the long run. In this case, a showing of exclusionary effects can be presumed to be welfare-reducing with a high probability, i.e. the strength of the presumption of illegality of exclusionary conduct is very high (and hence a TEB Illegality LS is indeed justified).

CONCLUDING REMARKS
While economics literature, under the assumption of a welfarist SS, suggests that EB LS's should be very widely adopted in competition law enforcement, for conducts other than hard-core horizontal agreements, in practice this is not observed in the assessment of many business conducts. Among the more mature jurisdictions, this is particularly pronounced in the EU (though FEB is also avoided in the US in recent hi-tech AoD cases). The non-welfarist SS's and illegality presumptions that the EU Courts have been adopting are likely to be mainly responsible for the differences in the LS's adopted (especially in relation to abuse of dominance cases) in the EU (MPS or TEB Illegality) relative to North America (MPS or TEB Legality).
In summary, the European courts adopt non-welfarist SS's and this influences their presumptions and choice of LS's: the former are presumptions of illegality and the latter are Modified Per Se (or TEB) Illegality LS's. The EC, in turn, chooses Modified Per Se (or TEB) Illegality standards, anticipating the choice of the courts. Clearly, it does not make sense to criticize the EC for acting in anticipation of the courts' choices. It also does not make sense to criticize the courts for a LS choice that is rational, given their SS. But it may indeed make sense to criticize the EU courts for using the wrong (non-welfarist) SS's.