Anti-Competitive Agreements under the Competition Act
Competition law is a rapidly burgeoning subject and has grown enormously in recent years, especially since the early 1990s. An increasing number of countries have undertaken economic reforms and embraced the market economy as a result of which competition law has been introduced in order to promote and maintain competition.  Although there is some controversy as regards the objective of competition law, there is broad agreement that the principal objective is to make the market economy work better by stopping private power from obstructing markets.  Competition law regulates market power in order to promote competition, thereby enhancing economic efficiency and increasing social welfare. The starting point for an understanding of the rationale behind competition law is to understand the microeconomic concept of ‘market’, the perceived benefits of market efficiency, the role of competition and their causal inter-relationship.  ‘Economic efficiency’ refers to the optimal use and allocation of resources by markets, thereby maximizing ‘social welfare’. The researcher has assigned a very specific meaning to the term ‘social welfare’ i.e. the combined effect of allocative and productive efficiency maximizing society’s wealth overall. A gain of $1 to either producers or consumers is treated equally. Competition law has a significant positive effect on economic welfare. For example, it deters anti-competitive conduct that may otherwise result in welfare losses to society. In this manner, competition law is clearly beneficial.
The history of modern competition law is generally traced to the United States, where the Sherman Act was enacted in 1890, although some economists state that antitrust laws preceded it and the Act itself was a codification of the common law.  In 1957, the Treaty of Rome was signed by six European nations  bringing into existence the European Economic Community (EEC). Competition law was incorporated into this treaty in the form of Articles 85 (later renumbered Articles 81).  This article prohibits anti-competitive agreements. In competition law, there is a standard distinction between practices that are condemned per se, and those to which a ‘rule of reason’ can be applied, with positive and negative outcomes being balanced on a case-by-case basis. Firstly, courts may apply a per se approach that treats certain practices as so obviously anticompetitive that they are held per se unreasonable.  Under this approach, the global position of law is that the plaintiff need only prove that the action in question occurred, and the defendants are precluded from asserting any efficiency justifications.  It must be noted that per se offences can only be found if the enterprises in question are in the same product market. Secondly, actions that are not deemed per se illegal are analyzed under a rule of reason analysis. Under this approach, courts balance the anti-competitive harms resulting from the practice with the efficiency or consumer welfare benefits resulting from the action.  The following are the two research questions the researcher seeks to answer through this project work:
1) What is the rationale of competition policy in terms of law and economics and is there a difference between the approach of competition law of the US, the EU and India?
2) Can the introduction of the notion of a continuum of more or less differentiated rules instead of per se rules on the one hand and the rule of reason on the other hand be justified in case of anti-competitive agreements within the framework of S.3 of the Competition Act, 2002?
In sum, the researcher seeks to establish that the highest benefits can be reaped by finding simple and robust rules which are able to solve most of the competition problems without causing high regulation costs. The researcher draws analogies between the competition case-law of the EU and the US in order to realistically predict how justice would be efficiently met out in settling disputes that arise before the Competition Commission of India (CCI)  assuming that the inherent problems of applying the legal and economic models of competition law of developed countries in a developing country like India are negligible.
A Comparative Analysis of the Competition Laws of the EU, US and India
The competition law of the European Community (EC) and the Sherman Act of the United States are treated as the two benchmarks of competition law in the world. Unlike the US economy, which was largely integrated at the time of the passage of the Sherman Act in 1890, EC was created in 1957 in order to establish a new European common market, and with a view to promote throughout the Community a harmonious development of economic activities.  The word ‘competition’ has different meanings in the competition laws of different jurisdictions and states have different conceptions about what constitutes ‘harm to competition’ or ‘lessening of competition.’ In the U.S. law, lessening of, or harm to, competition is found when the conduct or transaction creates, will probably create, or increase economic power in a relevant market to the detriment of buyers.  Historically, the US antitrust law had defined harm to competition much more broadly than consumer harm through price rises. In the early 1980s, the ‘consumer welfare’ paradigm was introduced to limit the scope of the US antitrust laws.  The Sherman Act, 1890 is a simple, short statute and it was interpreted by case law as intended.  The Department of Justice is the enforcement authority for the Sherman Act for both civil and criminal matters.  Some of the deficiencies of the Act were sought to be rectified by the Clayton Act, 1914 which contains provisions for merger control and also tying, price discrimination and exclusive dealing.  American antitrust law fails to offer a single or stable vision in the debate over the importance and nature of competition policy because it remains tied to specific social and historical events that other countries may never experience, or at least never experience in synchronization with the events  in the United States that propel antitrust law.  Therefore, some believe that American antitrust law cannot be sold to other countries, which create and live by their own standards.
The EC competition law is a sophisticated and effective system that serves the unique needs of the European Union (EU) by creating and enhancing a single European market that can promote prosperity and peace through the integration of several national economies.  In EC law, Article 81 deals with anti-competitive agreements. Article 81 may come before national courts of member countries in particular in the following scenarios  : a) in contract law, a party to an agreement may claim that a particular agreement is anti-competitive and thus unenforceable under Article 81(2)  ; b) individuals or undertakings who have suffered damages due to the application of anti-competitive agreements may seek compensation under tort law. Article 81(1) targets all agreements that prevent, restrict or distort competition. Article 81(2) declares these agreements are void. Article 81(3) provides that paragraph (1) may be declared inapplicable if the agreement in question improves production or distribution while allowing consumers a fair share of the benefits, if any restrictions are indispensable, and if the agreement does not confer on the undertakings the possibility of eliminating competition in a substantial part of the product market. Approval to an agreement under Article 81(3) is commonly called an exemption but it is not an exemption in the sense of the Sherman Act of the U.S. since it does not normally mean that an anti-competitive agreement is allowed.  Therefore, Article 81(3) throws a wide net that may catch pro-competitive agreements, and approval under Article 81(3) means that, viewed in its entirety, the agreement is pro-competitive and efficient.  EC law seeks to protect market actors as well as customers, and preserve the freedom to trade across state lines. Also, the European Commission has adopted Block Exemption Regulations (BERs) for agreements relating to specializations, research and development, technology transfer, motor vehicle distribution and some others.  Some scholars believe that whether EC competition law prohibits the same practices as the Sherman Act or whether EU decision-makers work in the same analytical tradition as U.S. competition officials and courts are questions that are irrelevant and misleading; EC law works because it ultimately serves the EU’s interests, regardless of whether it would serve U.S. interests or those of another society. 
India’s first experiment with a legislation to govern competition in the marketplace was the Monopolies and Restrictive Trade Practices Act (MRTP Act), 1969 which was adopted in pursuance of a Directive Principle of the Constitution  . The changing economic scenario required a law that would provide the framework for an economic analysis of the effects of a specific situation.  A High Level Committee was set up under the chairmanship of S.V.S. Raghavan in 1999 to advise the government on a suitable legislative framework for India relating to competition law.  Thus, the Competition Act was recommended to replace the MRTP Act with the view that regulatory focus should change from limitation of the size of undertakings to prohibiting trade practices which cause an appreciable adverse effect on competition.  The language used in the sections dealing with anti-competitive agreements in the Indian Competition Act, 2002 is similar to that of Article 81 of the EC treaty. Horizontal agreements  and certain types of vertical agreements  such as exclusive supply agreements and exclusive distribution agreements will be found to be anti-competitive if they cause an appreciable adverse effect on competition within India. The approach of the Competition Act, 2002 is also presumed to regard economic efficiency and social welfare as the prime object of the law.
A continuum of more or less differentiated rules instead of per se rules and rule of reason for anti-competitive agreements within S. 3 of Competition Act, 2002
There has been a long lasting battle between scholars preferring the so-called Efficiency Model  fostered by the “Chicago School" and the Interventionist Model  represented by scholars from the Harvard Business School.  Furthermore, both the European Commission and the European Courts have developed a more market-orientated approach over the last few years, taking into account the individual economic circumstances.  This is mirrored by the fact that the recent BERs all have included market share thresholds, e.g. they take market figures into consideration.  The per se rule is the judicial principle that a trade practice violates the Sherman Act simply if the practice is in restraint of trade, regardless of whether it actually harms anyone. “There are certain agreements or practices which because of their pernicious effect on competition and lack of any redeeming virtue are conclusively presumed to be unreasonable and therefore illegal without elaborate inquiry as to the precise harm they have caused or the business excuse for their use." 
Under the rule of reason, however, the essential question is not whether the claimed restrictions are “indispensable" to improve production or distribution, or to promote technical progress, but rather whether the anti-competitive effects of the restrictive agreement outweigh its pro-competitive effects.  The classic definition of the rule of reason dates back to 1918: “The court must ordinarily consider the facts peculiar to the business to which the restraint is applied; its condition before and after the restraint was imposed; the nature of the restraint and its effect, actual or probable. The history of the restraint, the evil believed to exist, the reason for adopting the particular remedy, the purpose or end sought to be attained, are all relevant facts."  Accordingly, analysis under the rule of reason is effectively open-ended; a full-scale market intervention is necessary and as a necessary corollary, no justifications brought forth by the defendants can be excluded a priori.[WHAT does this mean?] This corollary turned out to be a major obstacle to effective antitrust enforcement and led to the evolution of various per se rules. The U.S. Supreme Court explicitly stated that, “This principle of per se unreasonableness avoids the necessity for an incredibly complicated and prolonged economic investigation - an inquiry so often wholly fruitless when undertaken."  However, current enforcement practice pays increasing attention to the peculiarities of specific cases and weighs the advantages and dangers of the conduct in question, thus effectively performing a rule of reason analysis.  There are similar developments in European Competition Policy, albeit with a certain time gap compared to the US.  In this project, the researcher seeks to propose a general model of optimally differentiated rules which is still not in operation, neither in the US nor in the EU. A competition rule is optimally differentiated if the marginal reduction of the sum of error costs (as the marginal benefit of differentiation) equals the marginal costs of differentiation.
Distribution of Positive and Negative Welfare Effects of Business Behaviour
The expected benefit of a greater differentiation of competition rules is that it leads to a more effective separation of business behavior  with positive welfare effects and negative welfare effects, and therefore enables a better control of behavior. Positive welfare effects may be in the form of preventing anti-competitive conduct that may otherwise reduce or prevent competition, creating an environment conducive to adopting deregulatory competition policies and removal of excessive regulation, etc. Negative welfare effects are associated with excessive market power and reduced incentives to innovate. Both in antitrust law and the general law and economics literature, the problem whether more general or specific (precise) rules should be applied has been discussed from an economic point of view by applying an error-cost approach.  The fundamental principle underlying this idea is that the optimal rule is illustrated by the minimum of the sum of the welfare losses through wrong decisions (error costs) and regulation costs.  When certain types of harmful conduct escapes prosecution, the problem of “under-inclusiveness" i.e. Type I error arises (false positive) whereas when certain rules prohibit beneficial behavior the problem of “over-inclusiveness" arises i.e. Type II error (false negative). Therefore, such differentiated rules will amount to a decrease in both Type I and Type II error costs. The size of the benefit due to the optimally differentiated rules depends crucially on the frequency distribution of the welfare effects of the controlled business behavior.  Apart from the frequency distribution of welfare effects of business behavior, the other two determinants of the optimal differentiation of rules are the different kinds of regulation costs and the impact of rent-seeking activities of both industries and the law enforcement agency.
In the above figure, the welfare effects of five different business behaviours B1 to B5 have been depicted.  The vertical axis represents the relative frequencies  and the horizontal axis shows the welfare effects of Bi which on the right side are positive and on the left side are negative. For simplicity sake it has also been assumed that the welfare effects are evenly distributed between a minimum and a maximum value.  Analysing B1 and B2, it is clear that per se rules that allow or prohibit the respective behaviours are optimal rules. Type I and Type II error costs are zero and therefore, any additional differentiation of rules cannot lead to any benefit. Many types of business behavior like B3 mostly entail positive welfare effects but in a limited number of cases decrease welfare. Allowing this behavior per se would lead to Type I error costs (“false positives"  ). Additional differentiation that can identify and prohibit the cases of B3 towards the left side of the vertical axis might be useful because it will reduce Type I error costs. B4 is the symmetrical case where a per se prohibition leads to Type II error costs (“false negatives")  . B5 represents the type of behavior with highly varying welfare effects and a per se permission or prohibition, both would entail large error costs. Thus, we see that there is a complex continuum of (more or less) differentiated rules between both extremes of a per se and a full-scale market analysis.
Optimizing the Degree of Differentiation: How is it done?
Before delving into the method of optimizing the degree of differentiation of rules, the researcher seeks to outline the underlying assumptions for the present analysis:
(i) the aim of competition policy is wealth maximization;
(ii) competition policy enforcement is always imperfect.
Let us investigate what the possible reasons  for decision errors are:
(i) the problem of over-inclusiveness or under-inclusiveness of the rules;
(ii) political pressure and various interest groups influencing the government and governmental agencies;
(iii) rent-seeking activities of the regulating authority i.e. the CCI in this case in case of large discretionary powers being vested in it besides a lack of clear and transparent rules  ;
(iv) rent-seeking activities of specific industries through the provision of one-sided and distorted information leading to information asymmetry between the CCI and the regulated industries.
The following is the graph showing how the optimal degree of differentiation can be determined.
Error costs (ce), regulation costs (cr) and their sum referred to as total costs (c) are represented on the vertical axis while the degree of differentiation (d) is represented on the horizontal axis. A very small degree of differentiation on the left side indicates simple per se rules while a full-scale market analysis would imply a high degree of differentiation on the extreme right of the horizontal axis. A higher degree of differentiation will generally mean a reduction in error costs (ce) because further assessment criteria through more differentiated rules will lead to a better identification of pro and anti-competitive behavior. However, with rising differentiation the marginal cost reduction will diminish leading to a less decreasing curve ce. Marginal cost of additional differentiation presumably rises and therefore, cost curve cr increases and even slopes upwards. The total cost curve c is derived by the vertical aggregation of the increasing function cr and the decreasing function ce. The optimal degree of differentiation d* can be found at the minimum of the total cost curve c. Two notions of minimization are involved in this graph:
(i) Reaching the optimum degree of differentiation;
(ii) Reaching the optimum at the least cost of regulation.
“The crucial implication of this error-cost approach is that the optimally differentiated rule usually does not lead to correct decisions in every single case, but rather minimizes the sum of error and regulation costs on the average of all cases." 
Competition law is the principal legislative instrument for furthering competition policy and combating anti-competitive agreements. Market deregulation is complementary to competition law i.e. it is not the absence of regulation but the elimination of excessive government regulation in particular sectors of the economy. Both competition law and market deregulation are together called as competition policy. The policy as is evident in the newly notified Section 3 of the Indian Competition Act, 2002 is concerned is such that the researcher has drawn suitable analogies with U.S. and EU law to successfully establish that it is much more beneficial for the CCI to apply differentiated rules instead of deciding each dispute that comes before it on a case-by-case basis. Generally, even application of more sophisticated economic models to the decision-making process is not indicative of a case-specific analysis but rather evolving an optimal set of (more or less) optimally differentiated rules. A full-scale analysis of all positive and negative welfare effects is warranted and is desirable only in the rarest of rare cases whereas in respect of the millions of daily transactions, simpler rules are preferable, which can be monitored, enforced and complied with much more easily. Simple, robust, optimally differentiated rules can be accomplished if the marginal reduction of error costs equals the marginal increase in error costs through the application of additional differentiation criteria. These criteria shall vary from case to case. The factors that determine the optimal degree of differentiation include frequency distribution of welfare effects, regulation costs, rent-seeking behavior, etc. These determinants shall also vary from case to case and the competition rules can range from simple per se rules to (more or less) optimally differentiated rules to a full-scale market analysis, presumably in exceptional circumstances. In conclusion, the CCI would fare well to apply (more or less) optimally differentiated rules to cases that would come before it for adjudication instead of a blanket application of either per se rules or the rule of reason across the board, ensuring that cost of regulation is also kept within limits.