Abstract
This article assesses how the use of economic analysis and quantitative tools has evolved in merger assessments in India and draws a comparison with practices in two of the advanced jurisdictions, the United States and the European Union. In addition, this article identifies the trends and the gaps that still persist in India, in terms of the adoption of analytical approaches in merger analysis.
Merger analysis is an ex ante assessment of the impact of a proposed combination on the competitive process and consumer welfare. While on the one hand, competition authorities must ensure that they do not stop a combination that could result inefficiencies that are likely to ultimately benefit the consumers, they have to do so keeping in mind that a combination cannot be unscrambled if a wrong decision is made. Therefore, the ideal approach for competition authorities is to focus on the anticipated effects of a proposed combination in a careful, balanced, and robust manner and aim to address the question: “What is the market going to look like when the combination is implemented?”
Competition law jurisdictions around the world have considered and adopted—to varying degrees—analytical approaches and tools to determine the positive and adverse effects of a combination on competition (commonly referred to as an effects-based approach) rather than a “tick-the-box” approach (referred to as a form-based approach). These tools have been used to define relevant markets, to assess pre- and postcombination concentration and market power, and the possible effect on prices.
While these tools are universal in terms of applicability, the application of and reliance on these tools/analytical approaches and the threshold levels used vary across competition authorities. 1 This article assesses how the use of economic analysis and quantitative tools has evolved in merger assessments in India and draws a comparison with practices in two of the advanced jurisdictions, the United States (U.S.) and the European Union (EU). In addition, this article identifies the trends and the gaps that persist in India, in terms of the adoption of analytical approaches in merger analysis.
I. Competition Law Frameworks for Application of Economic and Quantitative Tools in Merger Assessments: EU, United States, and India
Effective application of economic tools and analysis in undertaking robust effects-based competition assessments requires an understanding on how the framework of the competition law allows for different types of analysis that are to be conducted, their application, as well as their interpretation in assessing a combination. This section analyzes the prevalent competition law frameworks in the EU and the United States and compares it with India’s.
In the EU and the United States, there is a main competition legislation which is supplemented by guidelines (and supported by judicial precedent) explaining the broad merger philosophy, specifying the types of evidence acceptable, and proving clear economic frameworks for the application of the merger regulations and also outlining various key analytical techniques and practices including certain thresholds (see Table 1).
Competition Law Legal Frameworks in India, European Union, and United States.
Source: Author’s compilation from various documents.
Competition law in India is governed by the Competition Act 2002 and enforced by the Competition Commission of India (CCI). The merger control provisions of the Act (Regulations of Combinations—Chapter II, Sections 5 and 6) came into effect in 2011 through a government notification. However, there are no merger guidelines in India, and the Competition Act 2002 is the only constituent of the competition law framework in the country.
In the paragraphs below, the study outlines the key elements of the competition law framework in the EU and United States and compares it with India, for each of the three steps of merger assessment—defining the relevant market, assessing market power and concentration, and assessing the likelihood of anticompetitive effects as a result of the proposed combination. 2
A. Defining the Relevant Product Market
In the EU, the European Commission (EC) Merger Regulations define the relevant product market based on demand-side substitution, considering interchangeability or substitutability by the consumer, by reason of the products’ characteristics, their prices, and their intended use. However, the guidelines on relevant market definition also clearly mention the use of supply-side substitutability in markets where its effects are equivalent to demand substitution in terms of effectiveness and immediacy. The guidelines also emphasize evidence for substitution in the past.
Further, EU’s guidelines on relevant market definition mention the use of several economic and quantitative tools for relevant market definition including the hypothetical monopolist test with a threshold of 5%–10%, tests based on intertemporal similarity in price movements, analysis of causality between price series, and similarity and/or convergence of price levels. Besides these, the EC is open to accepting other quantitative evidence that is able to withstand rigorous scrutiny. 3
In the United States, relevant market definition is based on demand substitution, assessing the ability and willingness of customers to substitute one product with another in response to price and nonprice (e.g., quality, level of service) changes. While the guidelines acknowledge the “responsive actions of suppliers” for competitive analysis, they consider it while assessing market participants, measuring market shares, and analyzing competitive effects and market entry. Further, the U.S. merger guidelines describe tools and principles, their application by the regulator, and acceptable evidence (e.g., evidence of switching time and cost) for defining the relevant product market—hypothetical monopolist test (small but significant and nontransitory increase in price [SSNIP] test 4 —mostly use 5% price [or value-added] increase), critical loss analysis, 5 and SSNIP for targeted customers. 6
India’s Competition Act 2002 defines the relevant product market based on demand-side substitutability considering a list of factors including end use or physical characteristics of products or services, prices, and consumer preferences. 7 However, it does not specify any tools to define the market. The Act is also silent on supply-side substitutability, although the CCI has accepted it for relevant market definition in some combination assessments in the past which is discussed in the next section.
B. Defining the Relevant Geographic Market
The EU defines relevant geographic markets by assessing the distribution of market shares between the Parties and competitors and an analysis of pricing and price differences at the national and Community or European Economic Area level. Further, as the guidelines explain, demand (e.g., local/national preferences, purchase patterns) and supply (e.g., impediments faced by firms in developing sales on competitive terms throughout the EU region) substitutability across regions are considered important for defining the relevant geographic market. The guidelines also explain various economic and quantitative analyses for relevant geographic market definition including an analysis of current geographic purchase patterns, trade flows, switching costs, and chains of substitution. 8
In the United States, relevant geographic markets are primarily defined depending on the location of suppliers if no price discrimination based on customer location is possible. If such price discrimination is possible, the relevant geographic market is defined based on customer location. 9 The SSNIP test is an important tool mentioned in the guidelines for both the cases along with evidence of transportation cost and ease of transportation and the influence of downstream competition.
In India, the relevant geographic market definition takes into account, homogeneity of conditions of demand and competition for supply across geographies, considering a host of factors including regulatory trade barriers, local specification requirements, national procurement policies, distribution facilities, transport costs, language, consumer preferences, and need for secure or regular supplies or rapid after-sales services.
C. Assessing Market Shares and Concentration
The EU and United States have well-defined tools and thresholds in their guidelines for assessing market power and concentration. The EU Horizontal Merger Guidelines mention the use of historic, current, and future market shares for competition assessment (as required). They also clearly state that market shares are interpreted not in isolation but subject to likely market conditions. Further, based on “well-established case law,” 10 the guidelines specify postmerger thresholds of (i) >50%—may be evidence of dominance; however, smaller firms can act as a sufficient constraint; (ii) <50% (40%–50% or <40%)—may also raise competition concerns in view of factors such as strength and number of competitors, capacity constraints, or the extent of substitution of products of the merging Parties; and (iii) <25%—presumed to not impede competition. 11 The guidelines also establish the use of the Herfindahl–Hirschman Index (HHI) as a measure of market concentration along with certain thresholds (see Table 2).
Herfindahl–Hirschman Index (HHI) Thresholds in European Union and U.S. Guidelines.
Source: Compiled from guidelines on the assessment of horizontal mergers under the Council Regulation on the control of concentrations between undertakings and U.S. Department of Justice and the Federal Trade Commission, Horizontal Merger Guidelines.
The U.S. Horizontal Merger Guidelines explain the regulators’ approach toward assessing market shares and concentration in conjunction with other market factors to determine whether a merger may substantially reduce competition. Further, the guidelines clarify that market shares are typically based on historical data and may understate or overstate a firm’s future competitive strength depending on ongoing and/or future changes (e.g., technological advancements). Hence, it is important to consider such changes while interpreting market share and concentration numbers. The guidelines provide possible interpretations of market shares and concentration figures under various merger scenarios and also recommend appropriate indicators for their measurement in terms of future competitive significance in the relevant market. Additionally, they explain the need to include market participants who are currently earning revenue in the market as well those who are likely to enter quickly (rapid entrants) in the event of a small price increase by the incumbents (SSNIP), without incurring significant sunk costs. Similar to the practice in the EU, the U.S. guidelines also mention HHI as a tool for measuring market concentration and specify its thresholds in the context of merger assessments (see Table 2). The guidelines clarify that these thresholds are only a means to identify mergers that are likely to raise anticompetitive concerns and for whom it is important to undertake a more detailed assessment of other competitive factors.
The Indian Competition Act 2002, however, specifies no tools (except market share), thresholds, or underlying approach toward assessing market power and concentration in merger assessments. It only lists “market shares” and “level of combination in the market” as part of the fourteen factors mentioned under section 20(4) of the Act to assess the possibility of appreciable adverse effects on competition (AAEC) as a result of the proposed combination.
D. Analyzing the Effect of the Combination
In the EU, the EC guidelines explain a number of factors to assess the likelihood of coordinated (reaching terms of coordination, monitoring deviations, and presence of deterrent mechanisms) and/or noncoordinated effects (e.g., merging firms have large market shares, limited switching and options for customers) as a result of the proposed merger. The guidelines also clarify that the list is not exhaustive and taken separately may not be decisive. Further, the guidelines explain several other factors for analyzing anticompetitive effects such as countervailing buying power; market entry—likelihood, barriers to entry, timeliness, and sufficiency; possible efficiencies—merger-specific, verifiable, and creating benefits for consumers; and the failing firm defense.
The U.S. merger guidelines provide detailed explanations about assessing unilateral and coordinated effects for industries with different product characteristics (e.g., differentiated goods vs. homogenous products) through various types of quantitative and qualitative economic analysis. For instance, for assessing unilateral effects, the guidelines mention the use of diversion ratio, econometric models, and merger simulations. Assessment of other factors such as the presence of powerful buyers; market entry—ease, likelihood, timeliness, and sufficiency; merger of competing buyers; and the failing firm defense is also explained elaborately in the guidelines. The guidelines also allow for merger-specific verifiable efficiencies, but these are mostly accepted only if the merger is not likely to cause significant adverse effects. It is difficult for competition agencies to verify the efficiency claims as detailed information to assess the credibility is in possession of the merging firms. Therefore, the responsibility of substantiating efficiencies rests on the merging firms in terms of likelihood of realizing the efficiency, magnitude of the efficiency, plan of action to realize the efficiency, the impact of the claimed efficiency to improve the merged entity’s incentive, ability to compete in the market, and so on. 12 Efficiencies justified using past experiences and as an integral part of the business planning process are considered to be creditable. Also, cognizable efficiencies that are not a result of an anticompetitive effect of the merger such as a fall in output, and estimated after removing costs incurred in realizing the efficiency, are taken well by the competition agencies. 13
The Indian Competition Act 2002 lists a set of fourteen factors under section 20(4) to assess the possibility of AAEC as a result of the combination. The factors include an assessment of competition through imports, the extent of barriers to entry, degree of countervailing power, ability to increase the price significantly and sustainably, the likelihood of removal of a vigorous competitor, the extent of close substitutes, nature, extent of innovation and vertical integration in the market, and the failing firm defense. The factors also provide scope for adopting an effects-based approach, allowing firms to demonstrate efficiencies as a result of the merger 14 and weighing the benefits and adverse effects of the combination against each other. 15 However, the Act is silent on the underlying approaches and possible economic analysis that it would consider appropriate to assess the impact of the combination under each of these factors.
The discussions above indicate that while India’s Competition Act 2002 covers the significant ground and provides for analysis of almost all important aspects of merger assessments, it does not provide any guidance on how it evaluates combinations (including economic and quantitative tools and their thresholds) or the underlying approach behind the assessments. This is usually done with the help of merger guidelines in advanced jurisdictions, the lack of which has left significant room for discretionary application and interpretation of the law and economic and quantitative tools for merger analysis in India. As a result, from a thorough review of the CCI’s orders under Sections 5 and 6, it is seen that there are several ambiguities and inconsistencies in the application of the law.
II. Analyzing the Indian Jurisprudence
An analysis of CCI’s orders shows that it has accepted certain economic and quantitative tools that have been proposed by the Parties and has adopted them in its own assessments. Over time, some of them have become standard tools that have gained legitimacy among practitioners. However, there is a fair amount of ambiguity in terms of their application and interpretation, and this is compounded by the fact that there is no codified merger guideline as in the case of the EU or the United States.
A. Supply Substitutability: Competition Act 2002 versus Case Law
The Competition Act 2002 only mentions demand substitutability as the criteria for defining the relevant product market. However, the CCI has applied supply-side substitutability for relevant product market definition while undertaking competition assessments for some proposed combinations. For instance, in the Linde–Praxair (2018) combination, while segmenting the retail market for helium, the CCI found that liquid and gaseous helium are not demand substitutable due to different characteristics and end uses. However, there is significant supply-side substitutability as suppliers can use their trans-fill centers for vaporizing, compressing, and filling helium into cylinders for gaseous supplies, as well as for repackaging liquid helium into dewars for liquid supplies. Therefore, based on supply substitutability, the CCI defined a single product market for helium without any further delineations. 16 The CCI also used supply-side substitutability in the Holcim–Lafarge (2015) combination to consider all grades of ready mix concrete (RMC) as one market as they are manufactured using the same raw materials and in the same manufacturing plant. The various grades of RMC differ in terms of strength and applications and are not demand substitutable. 17
In our assessment, CCI was correct in applying supply substitutability to defining the relevant product market in the above instances. Supply-side substitutability is also considered in the EU for relevant product market definition. However, unlike the EU, the CCI does not have any guidelines explaining when supply-side substitutability will be acceptable, as it is not assessed by the CCI in all cases. Further, since supply-side substitutability is not mentioned in the Act, its acceptability is left dependent on the current members of the CCI, who will change in the future and may choose to follow the Act in letter.
B. Elzinga–Hogarty (E–H) Test and Catchment Area Analysis: Ambiguities in Application
The CCI has used the E-H test 18 and catchment area analysis for relevant geographic market delineation in several merger assessments. However, the manner in which they were applied in certain instances raises several ambiguities. In the Indusland–Royal Bank of Scotland (2015) combination, the CCI used data on the average distance traveled by customers—fifteen to twenty kilometers—to avail banking services to define the primary service area of the Parties around their bank branches. Thereafter, the CCI acknowledged that the primary service area of other competitors may also have a similar fifteen- to twenty-kilometer radius and significant overlaps with the Parties, resulting in a chain of substitution sequences. Customers in all areas forming part of the chain reaction will be sensitive to small changes in the service charge and credit rates and hence can be part of one relevant geographic market. Thus, in light of this chain reaction and acknowledging the well-established connectivity in Mumbai, the CCI defined the relevant geographic market to be at least Mumbai. 19
However, the CCI has been arbitrary in its application of the catchment area analysis in subsequent combinations. In the PVR-DT (2016) combination, the CCI rejected the application of the catchment area analysis and chains of substitution, stating that the characteristics of the market, in this case, are different and hence application of chains of substitution to widen the relevant geographic market to Delhi National Capital Region (NCR) or even Delhi is not appropriate. 20 Instead, considering the responses submitted by the competitors and distributors, the CCI concluded that the geographic market is more local than the city level based on the following: (a) Prices charged by multiplexes are different across the regions and (b) there are differences in the clientele across regions (e.g., Hollywood films mainly run in South Delhi theaters). Given these arguments, the CCI divided the market into South Delhi; North, West, and Central Delhi; East Delhi; Gurgaon; Ghaziabad; Faridabad; Noida, and Greater Noida. 21 However, the CCI did not provide any quantitative or economic analysis it relied upon to delineate the market. There was also no clarity in the order regarding the rationale for clubbing North, West, and Central Delhi into one market. Further, inconsistent with the Indusland–Royal Bank of Scotland order, the CCI ignored the argument on good transport connectivity in Delhi NCR proposed by the Acquirer.
In the Ultratech–Jaypee (2016) combination, the CCI rejected the relevant geographic market definition (as groups of states) submitted by the Acquirer based on the E-H Test at 90% threshold based on interstate trade flows of cement as it found it to be too wide and not reflective of competitive constraints. The CCI undertook an independent application of the E-H test and used different little in from outside/little out from inside (LIFO/LOFI) thresholds for each state addition and relevant market. For instance, for overlaps in Andhra Pradesh, the CCI expanded the market to include Karnataka (threshold 11%) and Maharashtra (threshold 18%). The CCI did not add further states (as proposed by the Acquirer) based on an analysis of actual interstate trade flows—finding the proportion of exports to total production across the states to be insignificant. 22 However, the CCI did not define the thresholds for deciding “insignificance.” Similar ambiguities can also be seen in the application of the catchment area analysis and E-H Test in the Linde–Praxair (2018) and Holcim–Lafarge (2015) combinations.
The key point to note from this discussion is that there is no clarity on what kind of market characteristics allow for an application of the E-H test and catchment area analysis, and in what circumstances will these tests be given more or less weightage compared to other market characteristics and qualitative arguments. Of course, the Competition Act 2002 is silent on this, and the lack of guidelines leads to a fair amount of ambiguity and discretion in application.
C. Market Shares: What Is Considered to Be Significant?
Market shares are mentioned as one of the factors listed under Section 20 (4) of the Competition Act 2002 for assessing the likelihood of AAEC as a result of the proposed combination. However, the Act does not specify any thresholds or guidelines on the interpretation of the market shares. This provides significant room for discretion to the CCI to interpret the market shares as high or low or based on a form or effects-based approach.
In the PVR-DT (2016) combination, the CCI estimated market shares in terms of the number of screens in each relevant geographic market. For the Noida market, the CCI order said that the incremental market share as a result of the combination is 10.1% which is significant. 23 However, the incremental market share for the North, West, and Central Delhi markets was 5.1% and found to be not significant by the CCI. 24 Thus, it is not clear where the market shares transition from being insignificant to significant. In the Tata Steel–Bhushan Steel (2018) combination, the incremental market share for cold-rolled coils and sheets based on installed capacity was between 5% and 10% and was found to be significant. Further, the order states that the combined market share of the Parties in this market is “only” 20%–25%. 25 Even for galvanized products, the incremental market shares based on installed capacity have been found to be significant at 5%–10%; however, the order goes on to state that the combined market share of the parties is “only” 25%–30%. 26 In the market for precision tubes, the CCI order states that the combined market share of the parties is in the range of 35%–40% (without mentioning the term “only”). 27
The question then is, whether incremental market shares above 5% will be considered significant by the CCI? Also, does the CCI indicate by qualifying the combined market shares up to 30% by “only” that such combined market shares are not significant? Furthermore, in which circumstances will incremental market shares be examined in detail and will become part of the calculus in the decision of the CCI in terms of the post combination market concentration?
While the CCI has not relied solely on market shares for undertaking assessments in the cases discussed in this section (for instance, low capacity utilization has been given significant weightage in the Tata Steel–Bhushan Steel combination assessment), it is important for Parties and practitioners to have some clear indication about when market shares are considered high or low by the CCI and the typical economic analysis that it relies upon in making such assessments.
D. Measures of Concentration: What Are the Thresholds?
The Competition Act 2002 does not outline any measures of concentration. It only lists the “level of combination” as one of the factors for determining the likelihood of AAEC as a result of the proposed combination under Section 20 (4) of the Act. However, the CCI has assessed market concentration using HHI and the four-firm concentration ratio (CR4) in the past. In the Linde–Praxair (2018) combination, the CCI called analysis based on HHI as “standard concentration analysis” and seems to have used EU thresholds of a concentrated. 28 Similarly, the CCI used HHI analysis in the Vodafone—Idea (2017) combination as well, seemingly based on the EU thresholds—order states that “…with pre-combination HHIs exceeding 2000 in all telecom circles (except Haryana, Mumbai, and Punjab)…,” although they do not clearly specify or acknowledge the thresholds. 29 In the PVR–DT combination (2016), the CCI order clearly mentions the use of EU HHI thresholds for its concentration analysis. 30 The CCI has also used HHI in the concentration analysis for the Ultratech–Jaypee (2016) combination; however, it does not specify the thresholds used, but identifies changes in HHI as significant or insignificant across the various relevant geographic markets. 31 However, while the CCI has used HHI extensively in several merger assessments, there are orders where the CCI does not mention HHI at all. For instance, in the JK Tyre–CIL (2016) combination order, there is no mention of the HHI analysis. 32
In some cases, CCI has also used the CR4 analysis (along with HHI) for the concentration assessment of proposed combinations. For instance, in the Airtel–Telenor (2017) combination, the CCI found the retail mobile telephony services market to be concentrated with a CR4 of more than 65% and the premerger HHI more than 2,000 in all the overlapping circles. 33 The CCI also used CR4 in the Ultratech–Jaypee (2016) combination to assess concentration in the market. 34
The discussion above outlines two key ambiguities regarding CCI’s concentration analysis in merger assessments: (a) When will HHI and CR4 be considered by the CCI to assess change in market concentration as a result of the proposed combination? and (b) What are the thresholds for interpreting such analysis?
So far, the study discussed the ambiguity in the application of various economic analyses and tools by the CCI. It is also worthwhile to look at the broader picture to assess whether the CCI is undertaking merger assessments based on a form-based or an effects-based approach. Recent case law suggests that overall the CCI is moving toward an effects-based approach, in line with the practice in advanced jurisdictions such as the EU and the United States. In the Vodafone–Idea (2017) combination, in the relevant market for retail mobile telephony services, the CCI found that the incremental HHI was significant (ranging from around 400 Andhra Pradesh to 1,500 in Kerala) in all the fourteen circles where the combined market shares of the Parties were over 30%. However, the CCI considered various other factors in its assessment (e.g., fair distribution of spectrum, the significant quantity of unsold spectrum to mitigate access issues, significant buyer power due to easy number portability) to conclude no likely AAEC in the market as a result of the proposed combination. 35 Further, in the Tata Steel–Bhushan Steel (2018) combination, although the combined market share of the Parties was 35%−40% in the precision tubes relevant product market, the CCI found no likely AAEC based on the analysis of other market factors including the presence of other significant competitors and availability of unutilized capacity which will make it possible for competitors to increase production as required and deter any anticompetitive action of the Parties post combination. 36 However, there continue to be one-off cases where the CCI tilted toward a form-based analysis in its assessment of proposed combinations. For instance, in the PVR-DT (2016) combination, the CCI went too far to impose significant structural remedies (divestitures) in the South Delhi market over and above the behavioral remedies already offered by the Acquirer (limiting expansion, price caps on food and ticket prices, quality commitments, and commitment to not demand exclusivity from distributors), which were adequate to overcome competition concerns of the CCI. This was also the view expressed in the dissent note issued by some members of the CCI. In its decision, the CCI undermined the importance of other factors (besides market shares and HHI) such as the presence of significant substitutes, low barriers to entry, strong industry rivalry (low occupancy ratios), and high bargaining power of distributors which would have made it very difficult for the Parties to undertake anticompetitive practices postcombination, particularly amid the behavioral commitments. 37
III. Conclusions
This article shows that India’s Competition Act 2002 has drawn extensively from practices in advanced jurisdictions and provides an opportunity for analyzing the most critical factors for assessing proposed combinations in the country. However, as is mostly the case in India, there are limitations in the implementation of the Act, primarily due to the lack of detailed merger guidelines (including thresholds). While the young CCI has made noteworthy efforts in adopting international practices for undertaking merger assessments—be it in terms of following an effects-based approach and/or applying economic and quantitative tools, the lack of merger guidelines has created significant scope for discretion by the CCI in application and interpretation of merger analysis. This has led to several ambiguities and uncertainties for the business community (particularly the private sector and foreign investors) who definitely need more predictability on the CCI’s assessment while considering future potential combinations. The use of guidelines (for instance, as in the case of EU) would have definitely made past assessments more predictable, consistent, and clear for the Parties. For investors, it would have also added credibility to the assessment from a perception point of view, instilling greater confidence in the robustness and thoroughness of the CCI’s investigations. Perhaps in some cases, better application and interpretation of economic analysis and tools may have led to more appropriate decisions by the CCI. Further, the current situation, in the absence of merger control guidelines, carries the risk of a complete alteration in the approach toward merger assessments in India as members of the CCI change in the future.
One can argue that these ambiguities arising in Indian case law are a result of the CCI’s commitment to adopting an effects-based approach due to which it does not consider thresholds or predetermined predictable ways of undertaking analysis to be important. However, the CCI needs to clarify its position and suggest if practitioners and parties should infer its philosophy to merger analysis from past orders, particularly in the absence of any guidelines. If an open-ended approach is a way of allowing for flexibility in the Indian system, the CCI must make its orders very detailed and clear in terms of explaining its approach and methodology as well as the calculations.
Footnotes
Authors’ Note
This article is the modified version of the paper presented in the Fourth National Conference on the Economics of Competition Law organized by the Competition Commission of India.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.