EUROPEAN COMMISSION DG Competition

Only the English text is available and authentic.

REGULATION (EC) No 139/2004 MERGER PROCEDURE

Article 6(1)(b) in conjunction with Art 6(2) Date: 22/10/2021

In electronic form on the EUR-Lex website under document number 32021M10108

EUROPEAN COMMISSION

Brussels, 22.10.2021 C(2021) 7726 final

PUBLIC VERSION In the published version of this decision, some information has been omitted pursuant to Article 17(2) of Council Regulation (EC) No 139/2004 concerning non-disclosure of business secrets and other confidential information. The omissions are shown thus […]. Where possible the information omitted has been replaced by ranges of figures or a general description.

S&P Global Inc. 55 Water Street New York, NY 10041 United States of America

IHS Markit Ltd. th 4Floor, Ropemaker Place Ropemaker Street London, EC2Y 9LY England

Subject: Case M.10108 – S&P Global/IHS Markit Commission decision pursuant to Article 6(1)(b) in conjunction with 1 Article 6(2) of Council Regulation No 139/2004and Article 57 of the 2Agreement on the European Economic Area

1. THE PARTIES ............................................................................................................ 8

2. THE OPERATION...................................................................................................... 8

3. UNION DIMENSION ................................................................................................. 9

1OJ L 24, 29.1.2004, p. 1 (the ‘Merger Regulation’). With effect from 1 December 2009, the Treaty on the Functioning of the European Union (‘TFEU’) has introduced certain changes, such as the replacement of ‘Community’ by ‘Union’ and ‘common market’ by ‘internal market’. The terminology of the TFEU will be used throughout this decision.

2OJ L 1, 3.1.1994, p. 3 (the ‘EEA Agreement’).

Commission européenne, DG COMP MERGER REGISTRY, 1049 Bruxelles, BELGIQUE Europese Commissie, DG COMP MERGER REGISTRY, 1049 Brussel, BELGIË

Tel: +32 229-91111. Fax: +32 229-64301. E-mail: COMP-MERGER-REGISTRY@ec.europa.eu.

4. LEGAL FRAMEWORK ............................................................................................. 9

5. INTRODUCTION ..................................................................................................... 12

6. FINANCIAL DATA AND SOFTWARE PRODUCTS ........................................... 12

6.1. Introduction ..................................................................................................... 12

6.2. Market definition ............................................................................................. 13

6.2.1. Credit ratings ..................................................................................... 13

6.2.1.1. Overview and Parties’ activities ........................................ 13

6.2.1.2. Relevant product market.................................................... 13

6.2.1.3. Relevant geographical market ........................................... 17

6.2.1.4. Conclusion ......................................................................... 17

6.2.2. Company credit risk analytics data.................................................... 18

6.2.2.1. Overview and the Parties’ activities .................................. 18

6.2.2.2. Relevant product market.................................................... 18

6.2.2.3. Relevant geographic market .............................................. 19

6.2.2.4. Conclusion ......................................................................... 19

6.2.3. Indices................................................................................................ 20

6.2.3.1. Overview and Parties’ activities .................................. 20

6.2.3.2. Relevant product market.................................................... 21

6.2.3.3. Relevant geographical market ........................................... 24

6.2.3.4. Conclusion ......................................................................... 25

6.2.4. Identifiers and cross-reference services............................................. 25

6.2.4.1. Overview and Parties’ activities ........................................ 25

6.2.4.2. Relevant product market – Loan identifiers ...................... 26

6.2.4.3. Relevant geographic market – Loan Identifiers ................ 36

6.2.4.4. Relevant product market – CUSIP identifiers ................... 37

6.2.4.5. Relevant geographic market – CUSIP Identifiers ............. 39

6.2.4.6. Relevant product market – RED Code identifiers ............. 40

6.2.4.7. Relevant geographic market – RED Code Identifiers ....... 41

6.2.4.8. Relevant product market – Cross reference tools .............. 41

6.2.4.9. Relevant geographic market – Cross reference tools ........ 42

6.2.5. Pricing and reference data ................................................................. 43

6.2.5.1. Overview and Parties’ activities ........................................ 43

6.2.5.2. Relevant product market – Pricing and reference data...... 43

6.2.5.3. Relevant geographical market ........................................... 46

6.2.6. Desktop services ................................................................................ 47

6.2.6.1. Overview and Parties’ activities ........................................ 47

6.2.6.2. Relevant product market.................................................... 47

6.2.6.3. Relevant geographical market ........................................... 49

6.2.6.4. Conclusion ......................................................................... 49

6.2.7. Non-real time datafeeds (“NRTDs”) ................................................. 49

6.2.7.1. Overview and Parties’ activities ........................................ 49

6.2.7.2. Relevant product market.................................................... 50

6.2.7.3. Relevant geographical market ........................................... 51

6.2.7.4. Conclusion ......................................................................... 51

6.2.8. Fundamentals data ............................................................................. 52

6.2.8.1. Overview and Parties’ activities ........................................ 52

6.2.8.2. Relevant product market.................................................... 52

6.2.8.3. Relevant geographical market ........................................... 53

6.2.8.4. Conclusion ......................................................................... 53

6.2.9. Economic data ................................................................................... 53

6.2.9.1. Overview and Parties’ activities ........................................ 53

6.2.9.2. Relevant product market.................................................... 54

6.2.9.3. Relevant geographical market ........................................... 55

6.2.9.4. Conclusion ......................................................................... 55

6.2.10. Sector classification schemes ............................................................ 55

6.2.10.1. Overview and Parties’ activities ........................................ 55

6.2.10.2. Relevant product market.................................................... 56

6.2.10.3. Relevant geographical market ........................................... 56

6.2.10.4. Conclusion ......................................................................... 56

6.2.11. Leveraged loan market intelligence................................................... 57

6.2.11.1. Overview and Parties’ activities ........................................ 57

6.2.11.2. Relevant product market.................................................... 57

6.2.11.3. Relevant geographical market ........................................... 58

6.2.11.4. Conclusion ......................................................................... 59

6.2.12. Stock selection and strategy services ................................................ 59

6.2.12.1. Overview and Parties’ activities ........................................ 59

6.2.12.2. Relevant product market.................................................... 60

6.2.12.3. Relevant geographical market ........................................... 60

6.2.12.4. Conclusion ......................................................................... 61

6.2.13. Issuer solutions .................................................................................. 61

6.2.13.1. Overview and Parties’ activities ........................................ 61

6.2.13.2. Relevant product market.................................................... 61

6.2.13.3. Relevant geographical market ........................................... 61

6.2.13.4. Conclusion ......................................................................... 62

6.2.14. Investor event management solutions ............................................... 62

6.2.14.1. Overview and Parties’ activities ........................................ 62

6.2.14.2. Relevant product market.................................................... 62

6.2.14.3. Relevant geographical market ........................................... 62

6.2.14.4. Conclusion ......................................................................... 63

6.2.15. Issuance platforms ............................................................................. 63

6.2.15.1. Overview and Parties’ activities ........................................ 63

6.2.15.2. Relevant product market.................................................... 63

6.2.15.3. Relevant geographical market ........................................... 64

6.2.15.4. Conclusion ......................................................................... 65

6.2.16. Loan administration solutions ........................................................... 65

6.2.16.1. Overview and Parties’ activities ........................................ 65

6.2.16.2. Relevant product market.................................................... 65

6.2.16.3. Relevant geographical market ........................................... 67

6.2.16.4. Conclusion ......................................................................... 67

6.2.17. Digital design for financial services .................................................. 67

6.2.17.1. Overview and Parties’ activities ........................................ 67

6.2.17.2. Relevant product market.................................................... 68

6.2.17.3. Relevant geographical market ........................................... 68

6.2.17.4. Conclusion ......................................................................... 68

6.2.18. Equities and regulatory reporting (Dividend forecasting services) ... 69

6.2.18.1. Overview and Parties’ activities ........................................ 69

6.2.18.2. Relevant product market.................................................... 69

6.2.18.3. Relevant geographic market .............................................. 69

6.2.18.4. Conclusion ......................................................................... 70

6.2.19. Investor and administration services ................................................. 70

6.2.19.1. Overview and Parties’ activities ........................................ 70

6.2.19.2. Relevant product market.................................................... 70

6.2.19.3. Relevant geographical market ........................................... 71

6.2.19.4. Conclusion ......................................................................... 71

6.2.20. Institutional holdings/investor data ................................................... 71

6.2.20.1. Overview and Parties’ activities ........................................ 71

6.2.20.2. Relevant product market.................................................... 71

6.2.20.3. Relevant geographical market ........................................... 72

6.2.20.4. Conclusion ......................................................................... 72

6.2.21. Managed corporate actions data ........................................................ 73

6.2.21.1. Overview and Parties’ activities ........................................ 73

6.2.21.2. Relevant product market.................................................... 73

6.2.21.3. Relevant geographical market ........................................... 73

6.2.21.4. Conclusion ......................................................................... 74

6.2.22. Global securities financing data ........................................................ 74

6.2.22.1. Overview and Parties’ activities ........................................ 74

6.2.22.2. Relevant product market.................................................... 74

6.2.22.3. Relevant geographical market ........................................... 75

6.2.22.4. Conclusion ......................................................................... 75

6.2.23. Portfolio valuation tools .................................................................... 76

6.2.23.1. Relevant product market.................................................... 76

6.2.23.2. Relevant geographical market ........................................... 76

6.2.23.3. Conclusion ......................................................................... 77

6.3. Competitive assessment................................................................................... 77

6.3.1. Affected markets with respect to financial data and infrastructure products ............................................................................................. 77

6.3.2. Affected markets – horizontal overlaps............................................. 80

6.3.2.1. Loan identifiers.................................................................. 80

6.3.2.2. Indices................................................................................ 94

6.3.2.3. Index calculation and administration services (horizontal overlap) ......................................................... 101

6.3.3. Affected markets – vertical relationships ........................................ 104

6.3.3.1. Loan identifiers (upstream) – Leveraged loan market intelligence (downstream) ............................................... 104

6.3.3.2. Loan identifiers (upstream) – Fundamentals data (downstream) ................................................................... 112

6.3.3.3. Loan pricing and reference data (upstream) – Leveraged loan market intelligence (downstream) ......... 114

6.3.3.4. Loan pricing and reference data (upstream) – Company credit risk analytics (downstream) .................. 117

6.3.3.5. Loan pricing and reference data (upstream) – Leveraged loan indices (downstream) ............................. 119

6.3.3.6. Loan pricing and reference data (upstream) – Credit ratings (downstream) ....................................................... 121

6.3.3.7. CDS pricing data (upstream) - Company credit risk analytics (downstream).................................................... 122

5

6.3.3.8. Municipal bond pricing and reference data (upstream) – Municipal bond indices (downstream) ......................... 125

6.3.3.9. RED Codes (upstream) – Cross-reference tools (downstream) ................................................................... 127

6.3.3.10. CUSIPs (upstream) – Indices (downstream) ................... 128

6.3.3.11. CUSIPs (upstream) – Other markets (downstream) ........ 132

6.3.3.12. Credit ratings (upstream) – Indices (downstream) .......... 135

6.3.3.13. Credit ratings (upstream) – Other markets (downstream) ................................................................... 141

6.3.3.14. Equity indices (upstream) – Portfolio valuation tools, Global securities financing data, Economic data (downstream) ................................................................... 144

6.3.3.15. Fixed income indices (upstream) – Multi-asset indices (downstream) ................................................................... 146

6.3.3.16. Indices (upstream) – Desktop services, Non-real time datafeeds (downstream) ................................................... 147

6.3.3.17. Index calculation and administration services (upstream) – Equity indices, Fixed income indices (downstream) ................................................................... 149

6.3.3.18. Managed corporate actions data (upstream) – Equity indices (downstream)....................................................... 151

6.3.3.19. Economic data (upstream) – Credit ratings (downstream) ................................................................... 152

6.3.3.20. Sector classification schemes (upstream) – Economic data, Trade analytics, Digital design for financial services, Stock selection and strategy services (downstream) ................................................................... 153

6.3.4. Affected markets – conglomerate effects ................................ 154

6.3.4.1. Conglomerate effects – indices markets .......................... 154

6.3.4.2. Conglomerate effects – issuance platforms and desktop services ............................................................... 156

6.3.4.3. Conglomerate effects – credit ratings and loan administration solutions................................................... 159

6.3.4.4. Conglomerate effects – credit ratings and issuance platforms .......................................................................... 161

6.4. Commitments ................................................................................................ 164

6.4.1. Framework for the assessment of the Commitments ...................... 164

6.4.2. Proposed Commitments................................................................... 164

6.4.3. The CUSIP Commitments ............................................................... 165

6.4.4. The LCD/LLI Commitments ........................................................... 172

7. COMMODITY PRICE ASSESSMENTS AND MARKET INTELLIGENCE...... 178

7.1. Introduction ................................................................................................... 178

7.2. Market definition ........................................................................................... 178

7.2.1. Commodity price assessments......................................................... 178

7.2.1.1. Overview of the Parties’ activities .................................. 178

7.2.1.2. Relevant product market.................................................. 179

7.2.1.3. Relevant geographic market ............................................ 182

7.2.2. Commodities market intelligence .................................................... 183

7.2.2.1. Overview of the Parties’ activities .................................. 183

7.2.2.2. Relevant product market.................................................. 184

7.2.2.3. Relevant geographic market ............................................ 187

7.3. Competitive assessment................................................................................. 188

7.3.1. Commodity price assessments......................................................... 188

7.3.1.1. Competitive dynamics ..................................................... 188

7.3.1.2. Overview of the affected markets.................................... 196

7.3.1.3. Price assessments - Biofuels............................................ 196

7.3.1.4. Price assessments - Coal.................................................. 199

7.3.1.5. Price assessments - LNG ................................................. 203

7.3.1.6. Price assessments - Metals .............................................. 205

7.3.1.7. Price assessments - Natural Gas ...................................... 207

7.3.1.8. Price assessments - Oil .................................................... 209

7.3.1.9. Price assessments - Petrochemicals ................................. 213

7.3.1.10. Price assessments - Power ............................................... 218

7.3.1.11. Price assessments - Shipping ........................................... 221

7.3.2. Commodities market intelligence .................................................... 223

7.3.2.1. Market Intelligence – Trade analytics ............................. 223

7.3.2.2. Market Intelligence – Downstream and midstream energy .............................................................................. 227

7.3.2.3. Market Intelligence – Petrochemicals ............................. 231

7.3.2.4. Market Intelligence – Sugar ............................................ 234

7.3.2.5. Market Intelligence – Vertical relationships between market intelligence products............................................ 235

7.3.3. Commodities price assessment and market intelligence – non-horizontal relationships .................................................. 237

7.3.3.1. Commodity price assessment (upstream) - Market intelligence (downstream) ............................................... 238

7.3.3.2. Conglomerate effects ....................................................... 241

7.4. The OPIS/CMM Commitments..................................................................... 243

7.4.1. Framework for the assessment of the Commitments ...................... 243

7.4.2. Proposed Commitments................................................................... 244

7.4.2.1. Initial OPIS/CMM Commitments ................................... 244

7.4.2.2. Market test ....................................................................... 245

7.4.2.3. Final OPIS/CMM Commitments..................................... 247

7.4.3. Commission’s assessment ............................................................... 247

8. CONDITIONS AND OBLIGATIONS ................................................................... 249

9. CONCLUSION ....................................................................................................... 250

Dear Sir or Madam,

(1) On 3 September 2021, the European Commission received notification of a proposed concentration (the “Transaction”) pursuant to Article 4 of the Merger Regulation by which S&P Global, Inc. (“S&P” or “the Notifying Party”) acquires sole control of IHS Markit Ltd. (“IHSM”).S&P and IHSM are designated hereinafter as the “Parties”.

1. THE PARTIES

(2) S&P supplies credit ratings, price assessments, analytics, and data to the capital and commodity markets worldwide. S&P is divided into four divisions: (i) S&P Global Ratings issuing credit ratings (ii) S&P Global Market Intelligence (“SPGMI”), which supplies company, industry & asset-level data and analytics and also credit ratings data; (iii) S&P Dow Jones Indices (“SPDJI”), which supplies financial indices focusing on equities indices; and (iv) S&P Global Platts (“Platts”), which supplies commodity price assessments as well as related market intelligence.

(3) IHSM delivers information, analytics and software/workflow solutions to customers in business, finance and government. IHSM has four core segments: (i) Financial Services – supplying financial information, solutions, and processing product offerings; (ii) Transportation – supplying automotive and maritime and trade product offerings; (iii) Resources – supplying upstream and downstream product offerings; and (iv) Consolidated Markets and Solutions – supplying product design, economics and country risk, and technology media and telecoms product offerings.

2. THE OPERATION

(4) On 30 November 2020, S&P and IHSM entered into a binding agreement to combine in an all-stock transaction. Under the terms of this agreement, IHSM will merge with a wholly-owned and solely controlled subsidiary of S&P. Upon completion of the Transaction, current S&P shareholders will own approximately

33 Publication in the Official Journal of the European Union No C 367, 13.09.2021, p. 8.

44 SPDJI is a joint venture with CME Group Inc. and CME Group Index Services LLC (together, “ CME”) in which S&P owns 73% of SPDJI and CME group owns 27%.

67.75% of the combined company on a fully diluted basis, while IHSM shareholders will own approximately 32.25%.

(5) As a result, the Transaction is an acquisition of sole control of IHSM by S&P pursuant to Article 3(1)(b) of the Merger Regulation.

3. UNION DIMENSION

(6) The undertakings concerned have a combined aggregate worldwide turnover of more than EUR 5 000 million(S&P: EUR 6 524 million; IHSM: EUR 3 776 million in 2020). Each of them has a Union-wide turnover in excess of EUR 250 million (S&P: EUR […]; IHSM: EUR […] in 2020), but none of them achieves more than two-thirds of its aggregate Union-wide turnover within one and the same Member State. The notified operation therefore has a Union dimension.

4. LEGAL FRAMEWORK

(7) Under Articles 2(2) and 2(3) of the Merger Regulation, the Commission must assess whether a proposed concentration would significantly impede effective competition in the internal market or in a substantial part of it, in particular through the creation or strengthening of a dominant position.

(8) A merger giving rise to a significant impediment of effective competition may do so as a result of the creation or strengthening of a dominant position in the relevant markets. Moreover, mergers in oligopolistic markets involving the elimination of important constraints that the parties previously exerted on each other, together with a reduction of competitive pressure on the remaining competitors, may also result in a significant impediment to effective competition, even in the absence of dominance.

(9) In fact, the Horizontal Merger Guidelines describe horizontal non-coordinated effects as follows: “A merger may significantly impede effective competition in a market by removing important competitive constraints on one or more sellers who consequently have increased market power. The most direct effect of the merger will be the loss of competition between the merging firms. For example, if prior to the merger one of the merging firms had raised its price, it would have lost some sales to the other merging firm. The merger removes this particular constraint. Non-merging firms in the same market can also benefit from the reduction of competitive pressure that results from the merger, since the merging firms’ price increase may switch some demand to the rival firms, which, in turn, may find it profitable to increase their prices. The reduction in these competitive constraints could lead to significant price increases in the relevant market.”

(10) The Horizontal Merger Guidelines list a number of factors which may influence whether or not significant horizontal non-coordinated effects are likely to result from a merger, such as the large market shares of the merging firms, the fact that the merging firms are close competitors, the limited possibilities for customers to switch suppliers, or the fact that the merger would eliminate an important competitive force. That list of factors applies equally regardless of whether a merger would create or strengthen a dominant position, or would otherwise significantly impede effective competition due to non-coordinated effects. Furthermore, not all of these factors need to be present to make significant non-coordinated effects likely and it is not an exhaustive list.

(11) Finally, the Horizontal Merger Guidelines describe a number of factors, which could counteract the harmful effects of the merger on competition, including the likelihood of buyer power, the entry of new competitors on the market, and efficiencies. A merger between companies which operate at different levels of the supply chain may significantly impede effective competition if such merger gives rise to foreclosure. Foreclosure occurs where actual or potential competitors' access to supplies or markets is hampered or eliminated as a result of the merger, thereby reducing those companies' ability and/or incentive to compete.

(12) The Non-Horizontal Merger Guidelines distinguish between two forms of foreclosure. Input foreclosure occurs where the merger is likely to raise the costs of downstream competitors by restricting their access to an important input. Customer foreclosure occurs where the merger is likely to foreclose upstream competitors by restricting their access to a sufficient customer base.

(13) Pursuant to the Non-Horizontal Merger Guidelines, input foreclosure arises where, post-merger, the new entity would be likely to restrict access to the products or services that it would have otherwise supplied absent the merger, thereby raising its downstream rivals' costs by making it harder for them to obtain supplies of the input under similar prices and conditions as absent the merger.

(14) For input foreclosure to be a concern, the merged entity should have a significant degree of market power in the upstream market. Only when the merged entity has such a significant degree of market power, can it be expected that it will significantly influence the conditions of competition in the upstream market and thus, possibly, the prices and supply conditions in the downstream market.

(15) In assessing the likelihood of an anticompetitive input foreclosure scenario, the Commission examines, first, whether the merged entity would have, post-merger, the ability to substantially foreclose access to inputs, second, whether it would have the incentive to do so, and third, whether a foreclosure strategy would have a significant detrimental effect on competition downstream.

(16) Pursuant to the Non-Horizontal Merger Guidelines, customer foreclosure may occur when a supplier integrates with an important customer in the downstream market and because of this downstream presence, the merged entity may foreclose access to a sufficient customer base to its actual or potential rivals in the upstream market (the input market) and reduce their ability or incentive to compete, which in turn, may raise downstream rivals' costs by making it harder for them to obtain supplies of the input under similar prices and conditions as absent the merger. This may allow the merged entity profitably to establish higher prices on the downstream market.

(17) For customer foreclosure to be a concern, a vertical merger must involve a company which is an important customer with a significant degree of market power in the downstream market. If, on the contrary, there is a sufficiently large customer base, at present or in the future, that is likely to turn to independent suppliers, the Commission is unlikely to raise competition concerns on that ground.

(18) In assessing the likelihood of an anticompetitive customer foreclosure scenario, the Commission examines, first, whether the merged entity would have the ability to foreclose access to downstream markets by reducing its purchases from its upstream rivals, second, whether it would have the incentive to reduce its purchases upstream, and third, whether a foreclosure strategy would have a significant detrimental effect on consumers in the downstream market.

(19) Lastly, a concentration may also give rise to conglomerate effects. According to the Non-Horizontal Merger Guidelines, in most circumstances, conglomerate concentrations do not lead to any competition concerns.

(20) However, foreclosure effects may arise when the combination may confer on the merged entity the ability and incentive to leverage a strong market position from one market to another closely related market in particular by means of tying or bundling. The Non-Horizontal Merger Guidelines distinguish between bundling, which usually refers to the way products are offered and priced by the merged entity and tying, usually referring to situations where customers that purchase one good (the tying good) are required to also purchase another good (the tied good) from the same supplier. While tying and bundling have often no anticompetitive consequences, in certain circumstances such practices may lead to a reduction in actual or potential competitors' ability or incentive to compete. This may reduce the competitive pressure on the merged entity allowing it to increase prices.

(21) In assessing the likelihood of such conglomerate foreclosure effects, the Commission examines, whether the merged firm would have the ability and incentive to foreclose its rivals, and, whether such strategy would have a negative impact on prices and choice, and thus ultimately on competition. In practice, these factors are often examined together as they are closely intertwined.

5. INTRODUCTION

(22) The Transaction relates to different types of markets which can be broadly categorized into (i) financial data and software products and (ii) commodities data and analysis, leading to horizontal overlaps and non-horizontal relationships.

6. FINANCIAL DATA AND SOFTWARE PRODUCTS

6.1. Introduction

(23) Within financial data and software products, the Parties are active across a variety of products and value chains.

(24) S&P is primarily active as one of the top three global credit rating agencies (alongside Moody’s and Fitch), providing ratings regarding the creditworthiness of corporate and financial assets. Credit ratings revenue accounted for […] of S&P’s revenue in 2020. S&P also manages CUSIP identifiers, which is an alphanumeric code that identifies financial securities for the purposes of facilitating the clearing and settlement of trades. The company is also active in the provision of financial indices, primarily of equity indices, via SPDJI, a joint venture with the CME Group.

(25) IHSM is a provider of financial indices, primarily of fixed income and CDS indices. In addition, IHSM is also an important player throughout the fixed income value chain, offering pricing and reference data (for CDS, loan and bonds), issuance platforms, and other issuer solutions.

(26) The Parties’ activities relevant for the assessment of the Transaction include: credit ratings (S&P), company credit risk analytics (S&P), indices (both), identifiers (both) and cross-reference services (both), pricing and reference data (IHSM), desktop services (S&P), non-real time data feeds (S&P), fundamentals data (S&P), economic data (IHSM), sector classification schemes (S&P), leveraged loan market intelligence (S&P), stock selection and strategy tools (IHSM), issuer solutions (IHSM), investor event management solutions (IHSM), issuance platforms (IHSM), loan administration services (IHSM), digital design for financial services (IHSM), equities and regulatory reporting (IHSM), institutional holdings/investor data (IHSM), managed corporate actions data (IHSM), global securities financing data (IHSM), portfolio valuation tools (IHSM), commodity price assessments (both) and commodity market intelligence (both).

22 Non-Horizontal Merger Guidelines, paragraphs 95 to 118.

23 Financial data products are products that deliver financial information to the end-customer. This data is sometimes the by-product of the trading or other activities of financial players. Often financial data has also undergone aggregation, processing or enrichment. Financial data can also be packaged with functionalities and workflow tools to create comprehensive solutions for the end-customer.

24 CUSIP stands for Committee on Uniform Securities Identification Procedures.

25 CDS stands for credit default swap.

6.2. Market definition

6.2.1. Credit ratings

6.2.1.1. Overview and Parties’ activities

(27) Credit ratings are an opinion regarding “the creditworthiness of an entity, a debt or financial obligation, debt security, preferred share or other financial instrument, or of an issuer of such a debt or financial obligation, debt security, preferred share or other financial instrument, issued using an established and defined ranking system of rating categories”. Credit ratings of a company are issued by credit rating agencies like S&P based on a contract between the company requesting the rating (also known as “the issuer”) and the credit rating agency. A company rating is generally updated once per year and the company and the credit rating agency normally have a long-term relationship. These credit ratings for companies are referred to as “non-transaction ratings”. By contrast, credit ratings of a financial instrument are more ad hoc/one-off engagements, where a credit rating agency rates an individual issuance (e.g. a bond), and are referred to as “transaction ratings”. Ratings can be made public or remain private. They can also be unsolicited (i.e. produced by the credit rating agency without a client request) as opposed to requested by an issuer.

(28) Credit ratings data, i.e. information about the credit rating values of companies rated by a credit rating agency are licensed and used by investors of any kind (individual and institutional, such as insurance companies, pension funds, or governments), investment banks and financial index providers to inform investment decisions or inclusion of a company/financial instrument in an index.

(29) S&P Global Ratings (“SPGR”) is a credit rating agency, whose credit ratings data and related information products are licensed and distributed by S&P Global Market Intelligence (“SPGMI”). These products provide credit ratings data themselves (i.e. opinions on credit risk created by SPGR) and ratings-related research. S&P credit ratings data are also distributed by third party data vendors such as Bloomberg or Refinitiv.

(30) IHSM is active neither in the issuing of credit ratings nor in the distribution of credit ratings data.

6.2.1.2. Relevant product market

(A) The Commission precedents

(31) The Commission has not previously assessed the relevant product market for the supply of credit ratings.

26 Article 3(1)(a) of Regulation (EC) No 1060/2009 of the European Parliament and of the Council of 16 September 2009 on credit rating agencies, OJ L302, 17.11.2009, p.1 (the “CRA Regulation”).

27 IHSM provides credit assessment services, i.e. a valuation service for asset portfolios, bonds and private debt. These do not compete with credit ratings offered by rating agencies. Customers of IHSM are primarily financial institutions seeking the valuation of e.g. debt instruments not rated by a credit ratings agency or unrated counterparties (i.e. companies that do not have a credit rating established by a credit ratings agency). These financial institutions use credit assessment services captively, alongside their own internal credit risk assessments of the same debt instruments or counterparties.

13

(B) The Notifying Party’s view

(32) The Notifying Party considers credit rating issuance as a distinct product market from the distribution of credit ratings data, that would be downstream from credit rating issuance. This is mainly because suppliers and customers are not the same players for both activities. Credit ratings data is distributed by companies including e.g. Bloomberg, Refinitiv or FactSet who are themselves not credit rating agencies, to customers who are themselves not necessarily issuers.

(33) The Notifying Party considers that the markets for credit rating issuance should not be further sub-segmented by asset-class or geographical coverage, even though the lack of material demand-side substitutability may suggest separate markets for particular rating types. S&P also does not consider as plausible separate markets within overall credit ratings issuance public versus private ratings and transaction-related versus non-transaction ratings.

(34) Regarding the market of credit rating distribution, the Notifying Party does not consider it appropriate to separate markets based on the downstream use case or based on the type of credit rating being distributed.

(C) The Commission’s assessment

(35) The Commission does not consider it appropriate to separate credit rating issuance and the distribution of credit ratings for the purposes of this case.

(36) The Commission acknowledges that credit rating issuance is an activity which occurs between the rating agency and the entity soliciting a credit rating on a case-by-case basis, while credit ratings distribution occurs later on and usually entails the provision of credit ratings data in bulk to different users across the financial and public sectors. Moreover, in response to the CRA Regulation, the main credit rating agencies have created group structures separating the legal entities issuing credit ratings from those distributing credit ratings data.

(37) However, the distinction between the two activities appears largely artificial from a competition perspective, in particular for the assessment of the vertical relationships that arise as a result of the Transaction.

(38) First, the issuance of credit ratings and the distribution of credit ratings are more akin to two-sided market than separate markets. Competitive dynamics of the two activities currently do not appear to be markedly different. Indeed, the market position of a credit rating agency in terms of issuance has a direct translation into demand for its credit ratings data; similarly, the more widely used a credit rating agency’s ratings data, the more attractive it is for entities who need to obtain a rating. The market investigation confirms this. One competing credit rating agency mentions for instance that “in terms of licensing rating information, depending on the region and the asset class, investor customers look at the three CRAs and choose

28 Form CO, Chapter on Vertical Relationships, paragraphs 3.2-3.3.

29 Form CO, Chapter on Vertical Relationships, paragraph 3.4.

30 Form CO, Chapter on Vertical Relationships, paragraph 3.54.

31 Form CO, Chapter on Vertical Relationships, paragraph 3.55.

33 the licenses to cover their needs”. Customers of credit ratings data active in downstream markets also cite credit rating agencies as the relevant suppliers for credit ratings, not intermediaries.

(39) In addition, customers of credit ratings data, including those who purchase the data via third parties, typically require a license from the credit rating agency which issued the rating itself, as confirmed by the Parties themselves and by respondents to the market investigation, for instance for the use of credit ratings in financial indices.

(40) As a result, credit rating issuance and the use of credit ratings data are more akin to a two-sided market than they are to separate markets, as acknowledged by the relevant regulatory authority. As mentioned by the European Securities and Markets Authority (“ESMA”), “the credit rating industry is a two-sided market [… ]. Issuers in this market prefer to use the CRAs that are recognised by the largest number of investors, and investors prefer to use those CRAs who can offer the greatest coverage of the issuers and instruments they want to invest in”.

(41) Second, direct distribution of credit ratings data by credit rating agencies and indirect distribution by intermediaries do not appear to form part of the same market. While the Notifying Party argues that third party data vendors also compete in a credit ratings distribution market by distributing credit rating agencies’ data, it is clear that this is not a direct and independent competition. Third party data vendors, which typically aggregate credit ratings and other datasets into their desktop solutions, can only continue to distribute this data if allowed to do so by credit rating agencies, and they would not be able to generate or obtain the data independently from the relevant credit rating agency(ies).

(42) As for further segmentations of the credit ratings market by entity type, geography or rating type (based on the ESMA classification, between public or private rating, between solicited and unsolicited ratings or between entity and transactions ratings), there appears to be supply-side substitutability across all three dimensions. The three main credit rating agencies in particular already provide ratings in all segments, being registered at ESMA and having the know-how to rate different entities and asset classes. In any case, additional costs to cover a new segment in which they are not active would be limited, given that there would be no new registration costs.

33 See Minutes of a call with a competitor on 9 June 2021, 16:30 CET, paragraph 8.

34 See for instance replies to Questionnaire 4 for supplier of financial indices. Respondents nonetheless consider vendors such as Bloomberg and Refinitiv as important distribution channels for indices. See replies to question 25 of Questionnaire 4.

35 See S&P’s official website “FAQs: Licensing S&P Global Ratings’ Data” which reads that “SPGMI charges license fees to Indirect End Users based upon that End User’s particular Licensable Use Cases”.

36 Replies to question 35 of Questionnaire 4.

37 ESMA Thematic report on fees charged by Credit Rating Agencies and Trade Repositories, 11 January 2018, p. 9.

38 This is also the case when companies active in the issuance of credit ratings distribute the credit ratings of competitors. For instance, Fitch is licensed to distribute credit ratings from Moody’s on its platform. See Minutes of a call with a competitor on 9 June 2021, 16:30 CET, paragraph 9.

39 ESMA assesses different categories of ratings separately, distinguishing between “Corporate: Non-Financial”, “Corporate: Financial”, “Corporate: Insurance”, “Sovereign and Public Finance” and “Structured Finance”.

15

(43) Most smaller credit rating agencies active in Europe also provide ratings for different asset classes and types, although they often have a more limited coverage.

(44) By contrast, demand-side substitutability is limited. Ratings are issued on a case-by-case basis to a particular entity or financial instrument, so they are not interchangeable, and as such, e.g. a credit rating for a sovereign is not substitutable with a credit rating for a bank. This is also valid for users of credit ratings data, since they often need to know the credit ratings of a certain entity/instrument or group of entities/instruments (e.g. all insurance companies). The types of credit ratings to which a customer needs access to may depend on their activity. For instance, providers of indices will typically require access to all types of credit ratings based on the ESMA classification. However, a large majority of responding companies explained that they generally use the same credit rating agency(ies) for their entity-level and transaction-level ratings for efficiency reasons. As such, the demand pattern by rating type does not appear to be markedly different between these two categories. Different rules can also apply to different type of ratings these under the CRA Regulation. For instance, the regulation imposes the rating of structured finance instruments by at least two credit ratings agencies. The market investigation is not conclusive regarding the relevant categories. One competing credit rating agency broadly agrees with the ESMA categories and “considers a segmentation of the rating market between corporate ratings, sovereign ratings, structured finance ratings and financial institution ratings to be appropriate”.

(45) It results from the above that, for the purposes of this decision, the segmentation between credit rating issuance and the distribution of credit ratings data does not appear relevant. The precise scope of the market for credit ratings, i.e. whether they are segmented by (i) type of credit ratings (based on the ESMA classification, between public or private rating, between solicited and unsolicited ratings or between entity and transactions ratings), and/or (ii) based on which credit rating agency issues the relevant rating (e.g. top 3 credit ratings or smaller rating ratings, or even between S&P, Moody’s and Fitch ratings individually) can be left open, as it does not materially affect the Commission’s assessment.

40 See ESMA Report on CRA Market Share Calculation of 14 December 2020.

41 Replies to question 32 of Questionnaire 4.

42 Replies to question 6 of Questionnaire 10.

43 CRA Regulation, Article 8c.

44 Minutes of a call with a competitor on 19 August 2021, 16:30 CET, paragraph 12.

45 Replies to question 41 of Questionnaire 5 and question 9 of Questionnaire 10.

6.2.1.3. Relevant geographical market

(A) The Commission precedents

(46) The Commission has never defined the relevant geographical market for the supply of credit ratings so far.

(B) The Notifying Party’s view

(47) According to the parties, the scope of the geographic market for credit ratings issuance could be drawn at EU level, given the relevant regulatory requirements and may actually be global to the extent that the regulatory requirements are considered surmountable from a supply and demand-side perspective.

(48) On the other hand, the geographic market of the credit ratings distribution is global in scope.

(C) The Commission’s assessment

(49) Credit rating agencies, and in particular the three largest ones, are active globally. The market investigation also indicates that customers procure credit ratings globally. Most customers seeking credit ratings do procure rating services from credit rating agencies at a worldwide level.

(50) However, there seems to be notable differences in terms of the type of ratings issued and the competitive dynamics across different regions globally, as noted in the Parties’ internal documents. The regulatory regime in the EEA, namely the CRA Regulation and the ESMA supervision, does imply specific requirements for credit ratings agencies issuing ratings, companies requesting ratings, and users of ratings data.

(51) For the purposes of this decision, whether markets for credit ratings are EEA-wide or global in scope can be left open, as it does not materially affect the Commission’s assessment.

6.2.1.4. Conclusion

(52) For the purposes of this Decision, the Commission considers that credit rating issuance and the distribution of credit ratings data do not appear as distinct markets, and that a single market for the issuance and distribution of credit ratings by credit rating agencies can be found. Whether this market can be segmented by (i) type of credit ratings (based on the ESMA classification, between public or private rating, between solicited and unsolicited ratings or between entity and transactions ratings), and/or (ii) based on which credit rating agency issues the relevant rating (e.g. top 3 credit ratings or smaller rating ratings, or even between S&P, Moody’s and Fitch ratings individually) can be left open.

46 Form CO, Chapter on Vertical Relationships, paragraph 3.9.

47 Form CO, Chapter on Vertical Relationships, paragraph 3.58.

48 Replies to question 8 of Questionnaire 10.

49 See EC_00000075, slide 5 or EC_00000075 slide 12.

17

(53) For the purposes of this Decision, the Commission considers that the market for credit ratings is EEA-wide or global.

6.2.2. Company credit risk analytics data

6.2.2.1. Overview and the Parties’ activities

(54) S&P provides a product called the Credit Default Swaps Market Derived Signal Model (“CDS MDS”). CDS MDS uses CDS spreads to provide potential signals of changes in a company’s credit risk. These signals are volatile due to the nature of the input and customers do not generally rely on these signals as a prediction of creditworthiness by themselves, i.e. they are not a substitute for credit ratings.

(55) IHSM is not active in offering products with a similar function, but IHSM is a provider of CDS pricing data, which is an input for CDS MDS.

6.2.2.2. Relevant product market

(A) The Commission precedents

(56) The Commission has not previously considered the product market of company credit risk analytics data. In other previous decisions, the Commission considered discrete data content sets as plausible separate markets, given that they are not substitutable from a demand side perspective.

(B) The Notifying Party’s view

(57) The Notifying Party considers that the market in which CDS MDS is active, could be referred to as the supply of company credit risk analytics data. Suppliers provide this kind of data covering a broad range of companies and using different inputs, but the Notifying Party does not consider it appropriate to segment the market based on the type of company that the data refers to.

(58) According to the Notifying Party, customers are unlikely to consider company credit risk analytics data to be substitutable with other data or market intelligence.

(C) The Commission’s assessment

(59) The Commission understands that the plausible relevant market for company credit risk analytics data is differentiated with S&P itself offering several different kinds of products which could be grouped under this plausible market. From a supply side perspective, it is reasonable to assume that apart from requiring different inputs for different products, the general resources and expertise required for company credit risk analytics are rather similar to those required for providing credit ratings. The Notifying Party stated that “this part of the business seeks to offer insights as to the creditworthiness of entities/issuers using methodology and models originally

50 Commission decision of 19 February 2008 in Case M.4726, Thomson Corporation / Reuters Group, paragraph 44 and Commission decision of 20 July 2018 in Case M.8837, Blackstone / Thomson Reuters F&R Business, paragraph 17.

51 Credit Analytics, Credit Models, PD Model Fundamentals, PD Model Market Signals and RiskGauge Score, which are all products marketed by SPGMI’s Credit Risk Solutions business, see Notifying Party’s response to RFI 21, paragraph 5.4.

18

(52) developed for credit ratings issuance by S&P Global Ratings. Based on this, the market could be considered broader than company credit risk analytics and comprise credit ratings as well.

(60) However, from a demand side perspective, company credit risk analytics data are unlikely to be considered substitutable with credit ratings which serve different purposes, including being required by law in certain circumstances.

(61) Based on the above considerations and given that the market investigation provided no indication that the plausible market for company credit risk analytics data could be narrower or wider, the Commission considers the relevant product market to be the market for company credit risk analytics data.

6.2.2.3. Relevant geographic market

(A) The Commission precedents

(62) The Commission has not previously considered the geographic scope of a plausible company credit risk analytics data market. The Commission considered in a previous decision that markets for discrete financial data content sets are at least EEA-wide or global.

(B) The Notifying Party’s view

(63) The Notifying Party considers the relevant geographic market to be global as most providers are active globally and do not need to be physically situated in a particular location in order to provide data with respect to a particular company. Equally, customer demand is not driven by either customer or supplier location. In any event, the Notifying Party submits that it is not necessary to conclude on market definition as no plausible concerns arise on any basis.

(C) The Commission’s assessment

(64) The Commission has found no evidence to depart from the decisional practice with respect to discrete data content sets and the Notifying Party’s view. As such, the Commission concludes for the purposes of this case that the geographic scope of a plausible market for company credit risk analytics data is at least EEA-wide and likely global.

6.2.2.4. Conclusion

(65) For the purposes of this Decision, the Commission considers that company credit risk analytics data are a plausible separate market from credit ratings, and that this plausible market is at least EEA-wide or global.

52 Notifying Party’s response to RFI 21, paragraph 5.3.

53 Commission decision of 19 February 2008 in Case M.4726, Thomson Corporation / Reuters Groups, paragraph 106.

19

6.2.3. Indices

6.2.3.1. Overview and the Parties’ activities

(66) An index is a publicly available figure, regularly determined (i) by applying a formula or other method of calculation or making an assessment and (ii) on the basis of the value of one or more underlying assets or prices. An index has a numerical value calculated from prices of the instruments at a particular point in time.

(67) Indices are created by the index providers who own the intellectual property rights of the index. However, index providers may cooperate with clients or competitors and collectively own the intellectual property rights of an index. Furthermore, the actual calculation and administration of an index can be out-sourced to a third party that is not the creator of the index or holder of the intellectual property rights.

(68) Fixed income indices are indices that track debt instruments such as government bonds, corporate bonds and bank loans, which provide a fixed stream of income to the holder of the instrument. S&P is active through SPDJI, in the supply of fixed income indices as a relatively small player except in certain segments. IHSM owns the well-known iBoxx (investment grade and high yield corporate debt) index family.

(69) Equity indices are indices that track company shares. SPDJI’s primary offering is its ‘Headline Equity Indices’, namely the S&P 500, S&P MidCap 400, S&P SmallCap 600, Completion/Total Market and the Dow Jones Industrial Average (DJIA). […]% of S&P’s indices revenue comes from equity indices. In 2017, IHSM started to provide equity indices after the purchase of Euromoney indices. ISHM’s offering is limited to four primary sets of equity indices; EMIX Smaller European Companies Indices; EMIX World Indices; EMIX Global Mining Indices; and EMIX Global Gold, Mining and Energy Indices.

(70) SPDJI also supplies Environmental, Social and Corporate Governance (“ESG”) equity indices, which are a type of index reflecting a specific investment strategy (where the variable is ESG scores instead of e.g. company geography or capitalisation or revenue) offering investors exposure to companies according to their ESG profile in the context of country-specific and regional indices.

(71) CDS indices are indices tracking a basket of credit default swaps. They share some characteristics with fixed-income indices. IHSM through CDX in North America and iTraxx in Europe is active in the supply of CDS indices. IHSM is the only provider offering CDS indices. SPDJI does not supply CDS indices.

(72) Multi-asset indices are indices that track a mixture of assets such as equity and fixed-income. Customers typically use these when they seek to diversify investment products or benchmark diversified portfolios. Multi-asset indices measure cross-asset market performance.

(73) S&P is active in the supply of multi-asset indices sourcing indices from different index suppliers. IHSM does not supply multi-asset indices but provides indices as inputs to multi-asset indices providers.

(74) Alternative indices are indices that track alternative investments such as private equity and venture capital.

(75) Leveraged loan indices are a sub-category of fixed income indices that specifically track tradeable syndicated loans. S&P calculates two leveraged loan indices as part of its Leveraged Commentary and Data market intelligence product (“LCD”): S&P European Leveraged Loan Index (the “ELLI”) and the S&P/LSTA Leveraged Loan Index (“LLI”). IHSM is also active in the supply of leveraged loan indices through iBoxx.

(76) In addition to creating and licensing their own proprietary indices, indices suppliers may also supply calculation and administration services on a white-label basis to third parties to help them create and/or maintain their own proprietary indices. Suppliers therefore provide a range of services, depending on customers’ needs, including daily maintenance and calculation of the index, application and treatment of corporate actions, index distribution, and the supply of constituent (calculated) data files to the customer. Some customers may also ask suppliers to perform the administration requirements of their proprietary indices, in addition to calculation. In practical terms, this means that the indices services supplier (administrator) will own the index methodology from an operational and index governance perspective but not from an IP perspective. The administrators will oversee the index methodology and any changes thereto etc., they will own the rulebooks, perform consultations, and essentially run the index as if they were the proprietary owner. Calculation and administration services are usually supplied as an add-on to index licensing activities.

(77) S&P is active in the supply of calculation and administration indices focusing mainly on equity indices. SPDJI is also active in the supply of custom index design services, where it provides consultancy services to customers that want to design their own proprietary index. IHSM is active in the supply of calculation and administration services offering both equity and fixed-income services.

6.2.3.2. Relevant product market

(A) The Commission precedents

(78) In Deutsche Börse/NYSE Euronext, the Commission considered index licensing to be a separate product market, which could be potentially sub-divided by index type. In Intercontinental Exchange/NYSE Euronext, the Commission has also defined separate markets for indices based on the asset class of their constituents and on geography covered, national and regional.

(79) Most recently, in LSEG/ Refinitiv, the Commission concluded that segmentation based on asset class was appropriate, and found in particular separate relevant product markets exist for UK equity indices and for FX benchmarks. In addition, the Commission concluded that plausible separate markets exist for European equity indices, Global equity indices, fixed income indices, convertible bond indices, money market indices, gilt benchmarks, multi-asset indices, ESG indices, real-estate indices, commodities indices, FX indices, and interest rate benchmarks, but left the precise market definition of these open.

(80) Regarding equity indices, the Commission considered that separate relevant markets exist based on geographic coverage at least for UK equity indices, and plausible markets for European equity indices and Global equity indices.

(81) Regarding fixed income indices, the Commission noted that there are some indications of further segmentation by type of instrument and/or geography.

(82) Regarding CDS indices, the Commission has so far not assessed these in previous merger control decisions.

(83) The Commission also considered multi-asset indices to constitute a separate market from equity indices and fixed income indices, since they comprise securities across both asset classes.

(84) The Commission noted that there are some indications of further segments of equity indices and fixed income indices, such as ESG indices and real estate indices. These types of indices provide customers with specific types of exposure based on company sector (real estate) or other company properties (ESG).

(85) As regards benchmark administration and index calculation services, the Commission did not consider these activities to constitute separate product markets from the different asset-class indices but rather to be ancillary activities often performed by the same entities who design the indices.

62 Commission decision of 24 June 2013 in Case M.6873, Intercontinental Exchange / NYSE Euronext, paragraph 65.

63 Commission decision of 13 January 2021 in Case M.9564, London Stock Exchange Group / Refinitiv Business, paragraph 452.

64 Commission decision of 13 January 2021 in Case M.9564, London Stock Exchange Group / Refinitiv Business, paragraph 455

65 Commission decision of 13 January 2021 in Case M.9564, London Stock Exchange Group / Refinitiv Business, paragraph 457

66 Commission decision of 13 January 2021 in Case M.9564, London Stock Exchange Group / Refinitiv Business, paragraph 458

67 Commission decision of 13 January 2021 in Case M.9564, London Stock Exchange Group / Refinitiv Business, paragraph 459

68 Commission decision of 13 January 2021 in Case M.9564, London Stock Exchange Group / Refinitiv Business, paragraph 462

(B) The Notifying Party’s view

(86) In the Notifying Party’s view, financial indices can be sub-segmented (i) by asset class (e.g. equity vs fixed-income), (ii) by the geographical coverage of underlying securities (e.g. EU equities vs US equities) and/or (iii) by the specific rules for selecting index constituents or assigning weights to them (e.g. indices covering companies based on capitalisation or industry).

(87) The Notifying Party notes that from a supply-side perspective, suppliers are able to switch between supplying different types of indices within asset classes, and also switch between asset classes. However, from a demand perspective, indices for different asset classes are not substitutable (e.g. an equity index is not substitutable for a fixed-income index). Similarly, indices for different geographies will not be substitutable where a customer wants to create or benchmark financial instruments covering a specific geography.

(88) The Notifying Party recognises that different customers have different uses for indices and this includes customers licensing indices to create / issue funds and investment products and those licensing indices as a form of market data (performance benchmarking). However, the Parties do not consider that it is appropriate to segment the market by customer use, as from a supply-side perspective the same index may be used for both uses.

(89) The Notifying Party also notes that in recent years suppliers have started supplying ESG indices. As with other types of equity and fixed income indices, ESG indices compete within their relevant asset classes (e.g. ESG equity indices do not compete with ESG fixed-income indices) and the Notifying Party submits that there is not a separate stand-alone product market for all ESG indices irrespective of asset class.

(90) The Notifying Party considers that CDS indices are not substitutable from a demand perspective with indices tracking other securities and financial instruments i.e. they are not substitutable with equities, fixed-income debt, commodities indices etc.

(91) The Notifying Party finally considers that there is a single product market for index calculation and administration services separate from index licensing.

(92) The Notifying Party considers the creation of indices to be a separate market from the distribution of indices. In the Parties’ view, the relevant product market for the distribution of indices might plausibly comprise, at its widest, the distribution of financial markets data overall, reflecting both the supply-side substitutability between the distribution of different financial datasets and the fact that many customers typically consume financial indices alongside other data.

(C) The Commission’s assessment

(93) First, in relation to equity indices, the majority of customers indicate that a further segmentation by region, sector or other attribute (e.g. small capitalization companies vs large capitalization companies) may be appropriate. However, the Commission concludes that the question of further segmentation of equity indices can be left open for this case as no concerns arise regardless of the precise definition.

(94) Second, in relation to fixed income indices, the majority of customers find that segmentation by instrument type, i.e. bond vs loans vs CDS is appropriate, while they are divided regarding further segmentation by region, currency or other attributes including riskiness (high yield debt versus investment grade debt). As such, the Commission finds separate markets exist at least for bond indices, leveraged loan indices and CDS indices.

(95) As regards index calculation and administration services, the market investigation indicates that they might constitute a separate market from index licensing, and that there are further relevant segments based on the index asset class (i.e. index calculation and administration for fixed income indices versus equity indices). There is some evidence that the top index suppliers for index calculation and administration services are not the same as the top index suppliers for index licensing; moreover, the supply of such services for fixed income and equity indices appears to require different capabilities and inputs and thus may point to separate segments. However, index suppliers tend to provide index calculation and administration services to certain clients, along with their usual index licensing activities.

6.2.3.3. Relevant geographical market

(A) The Commission precedents

(96) In Deutsche Börse/NYSE Euronext, the Commission considered whether the geographic scope of the market is national, EEA-wide or global. The precise geographic market definition was eventually left open.

(97) In LSEG/Refinitiv, the Commission considered the geographic scope of index licensing, and in particular of the markets of European equity indices, Global equity indices, fixed income indices, multi-asset indices, ESG indices and real-estate indices to be worldwide.

(B) The Notifying Party’s view

(98) The Parties consider that the markets for the supply of financial indices are global in nature.

(99) The Parties also consider that the geographic market for index calculation and administration services is global.

72 Replies to question 20 of Questionnaire 5.

73 Replies to question 6 of Questionnaire 5.

74 Replies to question 11 of Questionnaire 4.

75 Replies to question 4 of Questionnaire 4.

76 Commission decision of 1 February 2012 in Case M.6166 – Deutsche Börse / NYSE Euronext, paragraph 149.

77 Commission decision of 13 January 2021 in Case M.9564, London Stock Exchange Group / Refinitiv Business, paragraph 465.

78 Form CO, Chapter on indices, paragraph §6.72, dated 3 September 2021.

(C) The Commission’s assessment

(100) The Commission has found no evidence to depart from the precedent and Notifying Party’s view. Indeed, competitors and customers confirmed that the markets are global; customers compare offerings on a world-wide basis and competitors provide their offerings and set prices on a world-wide basis.

6.2.3.4. Conclusion

(101) For the purposes of this decision, the Commission considers that relevant plausible global markets exist for the licensing of at least equity indices (potentially further segmented by region, sector or other attributes), bond indices (potentially further segmented by region, currency or riskiness), leveraged loan indices, multi-asset indices, commodities indices and ESG indices.

(102) For the purposes of this decision, the Commission considers that a plausible global market exists for index administration and calculation services, potentially segmented based on the underlying asset class (i.e. equity or fixed income).

6.2.4. Identifiers and cross-reference services

6.2.4.1. Overview and Parties’ activities

(103) Identifiers used in the financial data and software product markets are often alphanumeric codes to identify entities, securities and loans, in order to record, transmit and exchange data about those entities/assets.

(104) A security identifier is an alphanumeric code that can be used to identify a specific security, with varying levels of uniqueness depending on the type of identifier. Security identifiers may be considered a type of reference data as they are used to identify or retrieve certain information about a financial instrument. S&P is active in security identifiers with CUSIP Global Services (CGS) which operates the CUSIP system under licence from the American Bankers Association (ABA). The ABA owns the underlying intellectual property of CUSIP. CUSIPs are unique nine-digit numbers assigned to securities (e.g. stocks, bonds) issued in the US, Canada and 53 other jurisdictions for which CGS is the substitute national numbering authority.

(105) Derivative identifiers which are relevant in this case are so called Reference Entity Data (RED) identifiers which are used for a specific type of derivative, namely CDS. A RED identifier is used in a CDS context to confirm the relationship between a reference entity and a reference obligation, each of which is assigned a unique (RED) code. IHSM and S&P are active in this market together, S&P as the owner of the intellectual property rights to RED codes (which are based on CUSIPs) and IHSM based on a “[…]” license.

(106) A loan is not a security and hence, loan identifiers are not technically security identifiers. They do, however, serve practically the same purpose, namely to enable recording, transmission and exchange of data on a particular asset, in this case, a loan. IHSM uses its proprietary loan identifiers called LoanX IDs (LXIDs) to track individual loans in their loan pricing and loan reference data products. The LXID is an alphanumeric code and comprises also 10 data fields which include the key terms of the loan for identification purposes (e.g. issuer bank, loan type, maturity date etc.) which are also part of the wider loan reference data of a loan that includes ca. 100 data fields. S&P also owns loan identifiers, namely loan CUSIPs and LCD IDs (identifiers used in S&P’s Loan Commentary and Data (LCD) product).

(107) Cross-reference services allow for the matching and cross-referencing of instruments, entities and/or industries that may be assigned multiple different identifiers across multiple datasets. Cross-reference services aim to provide a mapping service that allow customers to connect identifiers for the same entity, instrument or industry so that they can then access relevant underlying datasets and other information connected to each identifier in the knowledge that they are comparing/reviewing ‘like-for-like’ and corresponding data across multiple and disparate data sources.

(108) S&P is active in the provision of cross reference services. S&P creates its mapping of identifiers by sourcing identifiers from third parties such as RED identifiers and LXIDs from IHSM.

(109) IHSM also provides a cross-reference tool named Entity Link. Entity Link is a service launched by IHSM in 2020 to provide customers with corporate hierarchy information for issuers and for entities that trade in the corporate credit market. IHSM’s product is a data feed-based solution focussed on cross-referencing in respect of its domains of activity, namely evaluated bonds, CDS and fixed income indices. IHSM also has a legacy cross-reference tool, Markit Maps.

6.2.4.2. Relevant product market – Loan identifiers

(A) The Commission precedents

(110) The Commission has not previously considered a separate product market for loan identifiers. However, a number of previous merger and antitrust decisions have considered identifiers used for financial instruments.

(111) In Standard & Poor’sthe Commission considered security identifiers as serving different purposes from other financial information, and particularly being used to identify a security rather than to monitor the markets or evaluate an investment opportunity, and found their distribution distinct from the distribution of other financial information. In that case, the Commission considered the relevant market to

be the market for first-hand electronic distribution and licensing of US International Securities Identification Numbers (ISINs) (records and numbers) via data feeds. US ISINs are based on CUSIPs, which are the security identifiers developed first for the domestic market in the US. In that decision, the Commission justified the limited substitutability between internal identifiers used by data vendors (such as Bloomberg Tickers and Reuters Instrument Codes (RICs)) and “standalone” identifiers such as ISINs with the more limited coverage of RICs and Bloomberg Tickers versus ISINs. In addition, the Commission highlighted that for some specific use cases, ISINs could not be replaced by RICs or Bloomberg Tickers, for example, inter-bank communications, asset and portfolio valuation, clearing and settlement, etc., and “in general all other legally and economically sensitive operations which require the highest degree of security and accuracy.”

(112) In Thomson/Reutersthe Commission considered security identifiers (referred to as “instrument codes” in that decision) separately in the competitive assessment, while not concluding definitively on a separate product market. The decision distinguished two types of security identifiers: (i) identifiers provided by data vendors like RICs for purposes of identification within the vendors’ own systems and (ii) industry-wide codes provided on a stand-alone basis like Stock Exchange Daily Official List (SEDOLs), ISINs and Financial Instrument Global Identifier (FIGIs).

(113) In LSEG/Refinitiv, the Commission considered that the potential relevant markets are the market for all security identifiers, and potential individual product markets for each security identifier (e.g. SEDOLs, RICs, ISINs, CUSIPs, etc.).

(B) The Notifying Party’s view

(114) Regarding IHSM’s LXIDs, the Notifying Party argues that LXIDs are IHSM’s internal identifiers used to track individual loans supporting IHSM’s loan pricing and reference data services.

(115) In the Notifying Party’s view, LXIDs cannot be considered independently from IHSM’s loan pricing data or loan reference data, given that LXIDs are not sold on a standalone basis. Other competitors in loan pricing and reference data also offer their data with their own internal identifier. These loan identifiers are not substitutable, according to the Notifying Party, as they each are only used to identify records in the respective providers’ loan pricing and reference data.

(116) By contrast, loan CUSIPs are standalone universal identifiers assigned to a deal and its underlying facilities in the US corporate loan market, in response to requests from issuers (i.e. entities issuing the debt) or associated participants in the issuance process (e.g. arranger banks), at the point that a loan is issued (or shortly after) that

are not tied to or otherwise associated with any other product or service, according to the Notifying Party.

(117) The Notifying Party submits that while the LXID and loan CUSIP are both loan identifiers in a broad sense, they each have particular characteristics that mean they are fundamentally different in terms of their features, their intended purpose and thus the way in which they can be and are used by customers. As a result, the Notifying Party is of the view that a loan identifier market comprising both LXIDs and loan CUSIPs is not plausible.

(C) The Commission’s assessment

(118) Some loan identifiers are generally sold together with loan pricing and reference data. This is in particular the case for LXIDs and LINs, which are usually commercialised together with IHSM’s and Refinitiv’s loan pricing and reference data respectively. On the other hand, other loan identifiers are sold on a standalone basis, such as S&P’s loan CUSIPs and FIGIs (previously owned by Bloomberg). The question is whether loan identifiers that are usually sold together with loan pricing and reference data, such as LXIDs and LINs, are part of the market for loan identifiers, meaning that they would also compete with other identifiers commercialised on a stand-alone basis.

(119) The market investigation confirmed that certain loan identifiers, such as LXIDs, are generally sold as part of loan pricing and reference data, and that competition between providers in that space primarily takes place at the level of the supply of loan pricing and reference data overall, including loan identifiers. More specifically, the majority of competitors offering loan identifiers answers that loan identifiers are part of the loan reference data offering. This is supported by factual elements. For example, LXIDs consist in an alphanumeric code and are always supplied together with the key terms (10 data fields) of the loan (e.g. issuer bank, loan type, maturity date etc.). Those key terms are also part of loan reference data. All loan identifiers are supplied with key terms and are in that respect comparable. In addition, the large majority of responding customers state that they need to receive loan pricing and reference data with a particular loan identifier. However, while this indicates that LXIDs and some other loan identifiers are generally sold in a bundle with loan pricing and reference data, this does not mean that they are not also part of a separate loan identifier market, competing against inter-alia loan identifiers commercialised on a stand-alone basis. Several elements collected during the market investigation strongly suggest that LXIDs are part of a distinct market for loan identifiers.

(120) Market participants recognize the intrinsic link between the pricing and reference data and loan identifiers and it is conceivable that the value of a loan identifier is derived to a large extent by the universe of loans that can be identified, described and analysed with the help of the identifier. One competitor summarizes it like this: “Identifiers intrinsically don’t have value no matter how good they are. Anyone can gather a universe of data about loans and create their own identifiers. What makes the identifiers “valuable” – what separates something like an LXID from a LIN – is what comes with it? LXIDs bring LoanX pricing and LINs are used by Refinitiv. Though Refinitiv is an older product, LoanX’s platform and delivery mechanism has allowed it to become more firmly embedded in the market. So, today, LXIDs are the market standard for identifiers not because of the identifier itself but because LoanX pricing is a critical need for participants. Furthermore, it connects not just to pricing, but also the whole world of IHS Markit content including the suite of services offered by Wall Street Office.” [emphasis added]. This highlights, however, that LXIDs are used separately from IHSM’s loan pricing and reference data, including by IHSM’s other services in the loan space.

(121) The fact that an identifier derives a large part of its value from data that it refers to does not mean that it cannot be a separate product, in particular if competing identifiers exist that are not intrinsically linked to such data.

(122) First, the market investigation showed that companies active in the loan space have an interest to get a license for using LXIDs stand-alone. More specifically, a number of competitors in loan pricing and/or reference data, other downstream products, and identifiers stated that they have a standalone interest in LXIDs. One competitor and potential customer states: “We have spoken with IHS only about loan identifiers, and IHS packages its LXID with reference data and/or pricing data.” Another market participant states: “[A potential customer] has previously sought to license LXID’s from Markit to enable its downstream businesses to compete with Markit and to meet client needs. But Markit resisted those attempts. It did so by offering [the potential customer] a licensing price that was so high as to signal clear disinterest in reaching agreement (essentially a constructive refusal to supply).” A competitor states: “The only data from IHSM that’s a need-to-have is the LXIDs – the pricing data is a nice-to-have but there are viable substitutes.” A further competitor explains: “(…) customers have asked [competitor] if they could use LoanX IDs. 10 years ago [competitor] made an attempt to get a license for LoanX IDs, but it was very expensive. [Competitor] would consider licensing LoanX IDs if the terms were appropriate to satisfy its customers’ needs.” One competitor highlights: “Without LoanX identifiers, a distributor cannot efficiently concord bank loan transactions or bank loan pricing, or further tie such content to its fundamentals and capital structure content sets.” This also highlights the importance of loan identifiers, and LXIDs in particular, for downstream products, which are discussed separately in this decision (see Section 6.3.3.1).

(123) The above quotes point to the existence of a standalone demand for LXIDs.

(124) Second, the market investigation suggests that competitors in the loan space are using loan identifiers such as LXIDs for specific use cases outside of IHSM’s loan pricing and reference data. A competitor highlights the use cases of loan identifiers from a supply-side perspective: “[Competitor] requires identifiers for two purposes: 1) for structuring data internally and mapping securities and loans to companies and 2) for providing the results of its analyses and data structuring to customers including identifiers that customers require/use.” While it is conceivable that providers of loan pricing and reference data or analytics in that space can work for internal purposes with an in-house identifier, this is not clear for providing that data/analytics to customers, who have their own views on the identifier they require. Another competitor confirms this: “[The competitors’ market intelligence product] has unique [proprietary identifier of the competitor] to support our internal data mapping. We do provide them to clients and they are part of our data sets – but we also have to provide market standard identifiers (LXIDs, CUSIPs) to support broader market data mapping. We have to meet the market requirement – by providing what our clients use – we can’t be competitive by relying upon our own identifiers.”

(125) Third, while LXIDs are generally sold together with loan reference and pricing data, certain companies actually benefit from specific LXIDs display rights. The Commission analysed specifically IHSM’s contracts with [Details of customer contracts] customers called “Alliance Partners” which are allowed to display LXIDs and asset servicers which also have display rights. The fact alone that IHSM has customers which are allowed to display LXIDs and others which are not, already indicates that LXIDs may be a separate product. These LXID display rights are in relation to loan pricing and reference data but also relating to other products which use loan pricing and reference data as an input, confirming that there are separate use cases for LXIDs, not directly related to interrogating IHSM’s loan pricing and reference database. This already contradicts the Notifying Party’s view that LXIDs are used solely for the purpose of identifying loans in IHSM’s loan pricing and reference database.

(126) Analysing contracts of the Alliance Partners, the Commission observes the following: In the earlier contracts ([Details of customer contracts]) to which the Commission had access, LoanX IDs or LXIDs are not explicitly mentioned. In later contracts ([Details of customer contracts]), LXIDs are explicitly defined , [Details of customer contracts]”. These later contracts also introduce the concept of [Details of customer contracts]” The explicit definition of LXIDs and the introduction of more restrictive licensing terms ([Details of customer contracts])

indicate that LXIDs seem to have developed a distinguishable value separate from loan pricing and reference data, even if not explicitly separately charged for.

(127) When comparing average fees charged to customers with display rights and without display rights, it is evident that LXID display rights have a commercial value that translates into higher average revenues from customers who contract for them: Asset servicers with display rights paid an average annual fee of [Parties’ financial results] for loan pricing data and [Parties’ financial results] for loan reference data in 2020. Alliance Partners with display rights paid average annual fees of [Parties’ financial results] for loan pricing data and [Parties’ financial results] for loan reference data in 2020. Compared to that, customers without display rights paid average annual fees of [Parties’ financial results] for loan pricing data and [Parties’ financial results] for loan reference data in 2020. This means that customers with LXID display rights paid on average [Parties’ financial results] times as much for loan pricing data as customers without display rights, and more than [Parties’ financial results] as much on average for loan reference data in 2020. While this difference may to some extent also be related to other factors influencing the amount of fees, such as number of users or scope of data provided, part of the higher fees are clearly related to the ability to display LXIDs. This is evident from the contract provisions mentioned above that include restrictions regarding the use of LXIDs specifically and separately from IHSM’s loan pricing and reference data.

(128) In addition to the Alliance Partners, [Parties’ financial results] asset servicing customers of IHSM have display rights of LXIDs. Together, revenues from the Alliance Partners and asset servicing customers make up [Parties’ financial results] of revenues generated by IHSM’s loan pricing and reference data in 2020. In summary, a sizeable part of IHSM’s loan pricing and reference data revenues seems to be generated by and based on the combined commercialization of IHSM’s loan pricing, reference and identifier data, of which LXIDs are a distinguishable and separate part evidenced by the display rights of LXIDs.

(129) While it could in turn be argued that only a minority of IHSM’s loan pricing and reference data customers (Alliance Partners and the asset servicing customers) seem to attach a separate value to LXIDs as evidenced by the request for display rights of LXIDs, it is evident that these customers provide products/services to other market participants which value and use LXIDs separately from interrogating IHSM’s loan pricing and reference data as evidenced by their decision to purchase those products/services (e.g. [Customer’s product]). Alliance Partners and asset servicing customers are more akin to re-distributors of IHSM’s loan pricing and reference data, including LXIDs, rather than being final end-customers.

(130) Fourth, the market investigation showed that in practice, LXIDs are used by end-customers for specific use cases, outside of IHSM’s loan pricing and reference data. When asked more specifically about how LXIDs are used in practice, half of the responding customers answer that “LXIDs are used alongside IHSM loan pricing and/or reference data, but also have other use cases for which they are used on a standalone basis (unrelated to IHSM loan pricing and/or reference data)” or even that they are used predominantly on a standalone basis (unrelated to loan pricing and/or reference data). Customers explain that there are a variety of use cases, which are not necessarily completely unrelated to loan pricing and reference data use, but require a loan identifier for communication purposes between market participants. One customer states: “Loan identifiers used for MO & BO reces, FO used for pricing (…)”, where MO & BO presumably indicates “middle office” and “back office”, while “FO” stands for “front office”. Front office is the part of a bank’s business which is usually in charge of trading, while middle- and back-office deal with post-trade processing, accounting, reporting, etc.

(131) Another customer mentions: “We have seen examples where we need to prove that we have an agreement with Market for LoanX identifiers to be enabled in a feed.” This is possibly a result of IHSM’s contracts with Alliance Partners and asset servicers, restricting redistribution of LXIDs to end-customers. A further customer mentions as standalone use case: “Use to follow the life of loans when repurchased/repackaged.” These statements suggest that use cases for LXIDs are broader than limited to being used exclusively in relation to interrogating IHSM’s loan pricing and reference data.

(132) Half of the responding competitors are also of the view that LXIDs have other use cases for end-customers which are unrelated to identifying loans in loan pricing and reference databases. One competitor mentions the following standalone use cases: “Mapping of loans, reporting trades and investments, risk management reporting and monitoring, reporting related to trading, settlements and clearing.” Another competitor states: “Loan reference data has a lot of uses in both the buy side and sell side – any data related to identifying and quantifying deals in the market is intrinsically a product. (…).” Another competitor mentions that their proprietary loan identifier is “used to manage and browse our loans database” but also “offered as part of various services that capture and expose loans and credit data.” A competitor explains: “LoanX identifiers are “must have” resources for CLO trustees and other customers. They are used in periodic reporting to CLO trustees to the CLO tranche investor. And investors in turn use the identifiers to map and monitor the underlying loan holdings of their CLO investments. Asset managers can buy and sell the loans that comprise a CLO, so scrutiny of the asset composition is important. Managers seek to profit from discrepancies in the cost of loans and the ratings that apply to pools of such loans.(…)”

(133) Even the Notifying Party acknowledges that LXIDs have separate use cases by stating “LXIDs are also used to identify loans which are processed using IHSM's WSO or ClearPar settlement services (…)”. While LXIDs are not essential for this purpose, as loans can be cleared and settled without an LXID, “Of the facilities processed in ClearPar [IHSM’s loan settlement service] in 2020, c. […] had an associated LXID, c. […] had an associated CUSIP, and […] submitted FIGIs or LINs for use in respect of facilities processed on ClearPar (although it is possible that ClearPar customers use FIGIs or LINs for loans in other use cases).” [emphasis added]

(134) The Commission did not investigate in detail the way in which the different product features (i.e. the fact that LXIDs change over the life of a loan, while loan CUSIPs are issued once and do not change) affect market participants’ use cases. However, given that IHSM provides customers with recommended update files which reconcile changing LXIDs and allow a customer to follow a loan over time to its initial LXID, the effective information available to customers in terms of substance does not seem to differ between LXIDs and other loan identifiers. In any case, no customer mentioned this difference in product features or its effects on the attractiveness of using one identifier versus the other. The only consistent quality criterion mentioned by customers throughout responses to the market investigation was the coverage of loan identifiers, i.e. the proportion of all loans covered by an identifier.

(135) Fifth, the market investigation showed that it is not uncommon for competitors to display other loan identifiers than their own. Half of the responding competitors state that they distribute third-party loan identifiers alongside their own products. This indicates that loan identifiers more generally have a separate value for customers which is independent from the underlying data distributed by each provider.

(136) A European business association also highlights that identifiers, including loan identifiers, are separate products in their members’ views: “Financial market participants rely on identifiers (for a wide range of financial products) in order to be able to identify and process financial products accordingly, notably in a post-trade context (clearing and settlement). These identifiers are also used in relation to loans. The identifiers are notably relevant for access to access certain data (e.g. RED, CUSIP, LoanX…). These identifiers products are purchased for separate and specific purposes by [European business association] members, and should be considered and assessed as a separate market.”

(137) In light of the above, the market investigation answers do suggest that LXIDs serve also other use cases than only for the internal organization of IHSM’s loan pricing and reference data. Even for the narrower use case of receiving loan pricing and reference data, customers have strong views on which identifier they need to receive this data with (also confirmed by the fact that most providers supply third-party loan identifiers as part of their loan pricing and reference data), highlighting the separate value of loan identifiers.

(138) Finally, the market investigation revealed that LXIDs and other loan identifiers sold standalone such as loan CUSIPs compete against each other. From a supply-side perspective, the majority of competitors respond that LXIDs and loan CUSIPs are substitutes for specific use cases or for nearly all use cases. One competitor explains: “It’s not “specific use cases” really – but they are substitutes for the instruments for which there are CUSIPs. But because LXIDs cover a significantly larger share of the market —particularly in the less transparent part of the loan market—they are more commonly used. (…)”[emphasis added] Several respondents echo that LXIDs have a broader coverage than loan CUSIPs. One competitor states: “In our view they are complementary to each other.”

(139) From a demand-side perspective, the majority of customers responds that LXIDs and loan CUSIPs are substitutes for specific use cases.

(140) To the question “Would you consider that any other loan identifier offers a good alternative to LXIDs?” more than half of the informative competitors’ responses was “Yes, but no alternative as close as loan CUSIPs”. More than half of the informative customers’ responses to this question are also the same.

(141) This indicates that LXIDs and other identifiers sold on a standalone basis such as S&P’s loan CUSIPs are considered substitutable by competitors and customers, at least for certain use cases. While the market investigation does not provide exhaustive answers which are those specific use cases, for which the identifiers of the Parties are substitutable, one customer replies for example: “(…), mapping across required.” This indicates, together with comments on the coverage of the identifiers that customers and competitors are possibly trying to maximize their ability to process and compare loan data by mapping several identifiers and data. Another customer explains: “These ID’s are tied back to another database that tracks additional metrics on a given transaction.”

(142) This use case is supported by several answers showing a certain level of complementarity in terms of geographic coverage of LXIDs and loan CUSIPs, e.g. one customer says: “Identifier for EMEA loans is predominantly LXID but some have CUSIPs too. In the US it is almost exclusively CUSIP.” Another customer states their view on LXIDs and loan CUSIPs like this: “(…) EMEA – LoanIX dominant, US, Cusip dominant but overlap between the 2.”

(143) The Commission also notes that there is apparently a lot of overlap between loan identifiers, at least in relation to the use of loan identifiers within IHSM’s traded loan service WSO : “In relation to WSO Services customers (…), […] of loans by volume (and […] by value) had a LXID associated with them; […] of loans by

volume (and […] by value) had a LIN associated with them; (iii) […] of loans by volume (and […] by value) had a CUSIP associated with them; and […] of loans by volume (and […] by value) had a Bloomberg ID associated with them. In the Commission’s view, this indicates that while currently no single loan identifier offers a full coverage of the traded loan universe, different loan identifiers (including LXIDs and loan CUSIPs) are obviously substitutable for the use in IHSM’s traded loan service WSO (which is a market leader in that space, […]).

(144) In addition, IHSM provides a loan mapping service which enables LXIDs to be linked to other loan identifiers including loan CUSIPs, FIGIs, LINs and other identifiers. The fact that IHSM provides a mapping service of its own loan identifiers to third party loan identifiers shows that there is a demand for loan identifiers more broadly which is not directly linked to the loan pricing and reference data of a specific provider. This is also supported by the fact that competitors provide their own loan pricing and reference data and loan market intelligence products including third party loan identifiers (see paragraph (135)). Furthermore, it seems that IHSM is currently the only provider of a mapping service covering all relevant loan identifiers.

(145) Comparing the reasoning in a previous Commission decision in respect of security identifiers, “internal identifiers” used by data vendors and “standalone identifiers” were not considered substitutable. However, loans are not securities and the reasons why the two types of security identifiers were not considered substitutable do not seem to apply with respect to loan identifiers.

(146) Firstly, in terms of coverage, the Commission refers to coverage figures provided by the Notifying Party (see Table 2) which show that in contrast to security identifiers (where internal identifiers were considered to lack coverage compared to standalone identifiers), this would seem to be the opposite with respect to loan identifiers. As also confirmed by the market investigation, LXIDs have a much broader loan market coverage than for example loan CUSIPs.

(147) Secondly, for securities transactions and related reporting, legal requirements mandate the use of ISINs, making them an essential identifier for all entities wishing to trade, clear, settle and report on securities. However, similar legal requirements do not exist for loans.

(148) Thirdly, the distinction that ISINs are created upon request and proprietary identifiers upon discretion of the owner seems to have less relevance for loans. This is because of the current time lag between LXID and loan CUSIP creation (see Section 6.3.2.1), which leads in most cases to LXIDs (the “internal identifier”) being available before loan CUSIPs (the “standalone identifier”). Furthermore, while LXIDs are indeed issued at the discretion of IHSM, it has been in IHSM’s interest to create as many LXIDs as possible, i.e. covering as many loans as possible to increase coverage and relevance of IHSM’s loan pricing and reference data and related identifiers. In terms of incentives to issue loan identifiers, there would hence not seem to be any significant difference between LXIDs and loan CUSIPs.

(D) Conclusion

(149) In summary, the above shows that LXIDs and identifiers sold on a stand-alone basis such as S&P’s loan CUSIPs are considered substitutable for certain use cases from a demand-side perspective. Mapping services of loan identifiers further support the conclusion that a broader market of loan identifiers is plausible, where LXIDs and loan CUSIPs compete.

6.2.4.3. Relevant geographic market – Loan Identifiers

(A) The Commission precedents

(150) The Commission has not previously considered the geographic market for loan identifiers. However, a number of previous merger and antitrust decisions have considered related markets, in the sense that they considered identifiers used for financial instruments.

(151) In Thomson/Reutersthe Commission considered the geographic scope of content sets, including instrument codes, to be at least EEA-wide and probably global.

(152) In Standard & Poor’s the Commission recognised that while often national security identifiers only cover securities issued in the respective countries, in certain exceptions, the National Numbering Agency (NNA) responsible for issuing the security identifiers also attributes identifiers for certain internationally issued securities. Ultimately, the Commission considered the market to be global for the purposes of that decision, as US ISINs were used globally by market players.

(153) In LSEG/Refinitiv the Commission admitted that while certain identifiers have more coverage of specific geographies, including CUSIPs covering North America, the geographic market would still be worldwide or at least EEA-wide in scope.

(B) The Notifying Party’s view

(154) The Notifying Party considers that the relevant product market of loan pricing and reference data and loan identifiers (however defined) is global. This is because suppliers can be situated anywhere globally as input is obtained remotely and distributed digitally to customers in any location. The Notifying Party submits that customer demand is not driven by the customers’ or suppliers’ location, and indeed customers may desire loan pricing or loan reference data in respect of issuers or regions in which they are not physically present.

(C) The Commission’s assessment

(155) The Commission asked market participants in its market investigation at which geographic level they offer and source loan pricing and/or reference data and/or loan identifiers.

(156) The large majority of competitors replied that they offer data/identifiers at a worldwide level, a majority replied that their competitors are active and customers are located globally and half of the competitors said that prices are set at worldwide level. No competitor specified any other relevant geographic scope of their product apart from global or indicated that their response would differ depending on whether loan pricing or reference data were concerned. The responses indicate that from a supply-side perspective, the geographic market of loan pricing data, loan reference data, and loan identifiers is considered global.

(157) The large majority of customers replied that they purchase loan pricing and reference data by comparing offers at a worldwide level. The same large majority confirmed that the same suppliers are active at worldwide level.

(D) Conclusion

(158) For the purposes of this decision, the Commission considers based on the above evidence and reasoning, that loan identifiers are a separate product from loan pricing and reference data and given the at least partial substitutability between LXIDs and S&P’s loan CUSIPs, the relevant product market plausibly consists of loan identifiers more broadly. The Commission considers that this market is global in scope.

6.2.4.4. Relevant product market – CUSIP identifiers

(A) The Commission precedents

(159) The Commission has previously considered the market for security identifiers in three decisions: Thomson/Reuters, LSEG/Refinitiv and concretely in respect of S&P’s security identifier, CUSIP also in Standard & Poor’s.

(160) In Thomson/Reuters, the Commission did not conclude on a separate product market for security identifiers, but did consider them separately in the competitive assessment. Furthermore, the decision described identifiers like RICs used by data vendors for purposes of identification within the vendors’ own systems and industry-wide codes provided on a stand-alone basis like SEDOLs, ISINs and FIGIs.

(161) In LSEG/Refinitiv, the Commission considered that security identifiers constitute a relevant separate market from other financial information content sets. In that

decision, the Commission left open whether all security identifiers constitute a relevant market or whether each security identifier constitutes a separate market.

(162) In Standard & Poor’s the Commission considered the market for first-hand electronic distribution and licensing of US International Securities Identification Numbers (ISINs) (records and numbers) via data feeds. Bloomberg Tickers and Reuters Instrument Codes (RICs) as proprietary identifiers were not considered effective substitutes as their coverage is not the same as for ISINs, and their issuance is at the discretion of the system owner, whereas ISINs are issued by national numbering agencies (NNAs) at the request of issuers.

(B) The Notifying Party’s view

(163) The Notifying Party acknowledges the relevant product market of security identifiers as defined by the Commission in Standard & Poor’s, and considers that the relevant market in this case is the first-hand supply of CUSIP identifiers along with related descriptive data via data feeds.

(C) The Commission’s assessment

(164) The Commission notes that CUSIPs were first conceived and introduced in 1968 for US-issued securities and before the ISIN Standard was developed (1978). In accordance with the ISIN Standard, ISINs build on pre-existing national numbers. This is why US-ISINs are based on CUSIPs, which existed before ISINs. They have been provided by S&P through its CUSIP Global Services Unit and on behalf of the ABA since 1968.

(165) CUSIP numbers consist of nine characters (including letters and numbers) that uniquely identify a company or issuer and the financial instrument. In 1989, a similar system to identify non-US securities (CUSIP International Numbering System or CINS) was introduced. CINS employ the same nine character identifier as CUSIP, but also contain a letter in the first position to signify the issuer's country or geographic region. CGS is the NNA for the USA and Canada, and currently the substitute NNA for 53 other jurisdictions.

(166) CUSIPs were assigned over the years to more and more asset classes (e.g. in 1983 to Certificates of Deposit, in 1990 to commercial paper issues, in 2003 in the form of RED Codes, which are CUSIP-like identifiers, to credit default swaps, in 2004 to syndicated loans, in 2016 to physical precious metals).

(167) Automation in trading, clearing and settlement processes of securities as well as regulatory changes mandating the use of CUSIPs and ISINs in US- and EU-Legislation for settlement and reporting purposes have changed the importance and reliance of market participants on CUSIPs and ISINs. However, their fundamental purpose has not changed over the years and remains the unambiguous identification of securities (and other assets) in communication between market participants, including supervisory authorities, at the national level.

(168) From both a supply-side as well as demand-side perspective, there are effectively no substitutes to CUSIPs/CUSIP-based ISINs for securities issued in one of the jurisdictions for which CGS is the NNA, i.e. there is only one entity who supplies CUSIPs/CUSIP-based ISINs first-hand which is S&P through its CGS Unit. Even if some customers mention SEDOLs, RICs and Bloomberg Codes/Tickers as potential alternatives, this applies only to non-US securities/securities not covered by CGS’s NNA service or for specific environments, e.g. “When using Refinitiv data feeds, the RIC code can be credible alternative (…)”

(169) One customer confirms this: “For US/Canadian Securities there is only ONE source: CUSIP. Even if we use ISIN for US securities we require a license due to 'CUSIP embedded ISIN' topic. In both cases the CUSIP Service Bureau is the owner of the license.” Hence, whoever wants/needs to trade and hold US-issued financial instruments or financial instruments issued in any other jurisdiction for which CGS is the substitute NNA requires access to CUSIPs/CUSIP-based ISINs and related descriptive data.

(170) Customers unanimously answered “No” when asked if they would likely switch security identifier provider in case of a small but significant increase (5-10%) in the licensing fees of CUSIPs.

(171) Based on the market investigation the Commission considers that it has no reason to depart from previous decisions and the Notifying Party’s view that the relevant market for the purposes of this decision is the issuance, distribution and licensing of CUSIP based-ISINs (records and numbers) via data feeds.

6.2.4.5. Relevant geographic market – CUSIP Identifiers

(A) The Commission precedents

(172) In its previous Standard & Poor’s decision, the Commission considered the market for CUSIP identifiers to be global, given that CUSIPs are distributed and used by customers globally.

(B) The Notifying Party’s view

(173) In the Notifying Party’s view, the geographic market for the first-hand supply of CUSIP identifiers is global in scope.

(C) The Commission’s assessment

(174) The Commission notes that the purpose of security identifiers like CUSIPs is to unambiguously identify securities in communication between market participants trading securities. In particular, institutional investors trade securities globally (i.e. independent of their own location) and therefore require globally acknowledged security identifiers like CUSIPs. In addition, the market investigation provided no evidence that the Commission should depart from previous decisions and the Notifying Party’s view. The Commission therefore considers that the market for CUSIP identifiers to be global.

(D) Conclusion

(175) For the purposes of this Decision, the Commission considers the issuance, distribution and licensing of CUSIPs/CUSIP-based-ISINs (records and numbers) via data feeds a separate market which is global in scope.

6.2.4.6. Relevant product market – RED Code identifiers

(A) The Commission precedents

(176) […].[…]

(B) The Notifying Party’s view

(177) The Commission has previously considered the market for RED Codes in its IHSM/CME/JV decision, though did not conclude in that decision whether RED Codes form a distinct market or are part of a larger market comprising other identifiers such as ISINs paired with LEIs.

(178) In addition, the Commission has considered security identifiers, concretely CUSIP-based ISINs and SEDOLs, see Section 6.2.4.4. In those cases, the Commission has considered markets comprising all security identifiers, and plausible individual markets for each identifier.

(C) The Commission’s assessment

(179) Regarding RED identifiers, the Notifying Party considers that there are a range of alternative identifier systems available that could be used as a substitute for RED identifiers, including self-supply and the combined use of legal entity identifiers (LEIs) to identify entities and bond ISIN numbers to identify reference obligations. Therefore, a broader market comprising those alternatives could be considered, according to the Notifying Party. However, the Notifying Party acknowledges that a plausible narrow market only consisting of RED identifiers may exist but submits that no conclusion needs to be drawn regarding the product market as no concerns arise based on a narrower product market definition comprising only RED identifiers.

(180) The Commission notes that while RED Codes may in theory be substitutable with LEIs and ISINs, market participants may be reluctant to consider this as switching identifiers is not generally considered easy and quick. This is confirmed by a number of market participants who attribute a certain market power to IHSM with respect to RED Codes, for example: “Both S&P and IHS Markit have significant market power in relation to market data. For S&P, it’s their indexes and credit ratings, for Markit it’s their fixed income indexes and RED codes.” If RED Codes were easily substitutable, ISHM would likely not be considered to have market power with respect to them.

(181) From a supply-side perspective, RED Codes itself are not substitutable by any potential competitor, […].

(182) Based on the above and no indications in the responses to the market investigation that RED Codes are substitutable with one or more identifiers, the Commission considers the relevant market in this case to consists of RED Codes.

6.2.4.7. Relevant geographic market – RED Code Identifiers

(A) The Commission precedents

(183) The Commission has previously considered the geographic market for RED Codes or a plausible wider market comprising other identifiers that would be able to substitute RED Codes in its IHSM/CME/JV decision as global.

(B) The Notifying Party’s view

(184) The Notifying Party submits that the market for RED Codes is global in scope, given that IHSM distributes RED Codes globally.

(C) The Commission’s assessment

(185) The Commission received no responses to the market investigation suggesting that the geographic scope should be any other than in line with previous decisions and the Notifying Party’s view, i.e. global. The Commission furthermore notes that RED Codes are used to support the trading, clearing and settlement of CDS, which is a global market. Based on these considerations, the Commission sees no reason to depart from its previous consideration that the market for RED Codes is global.

(D) Conclusion

(186) For the purposes of this Decision, the Commission considers that RED Codes form a separate market which is global in scope.

6.2.4.8. Relevant product market – Cross reference tools

(A) The Commission precedents

(187) The Commission has not previously considered a market for cross-reference services or tools.

(B) The Notifying Party’s view

(188) The Notifying Party submits that cross-reference services will likely have a limited demand-side substitutability with other products or services given the specific purpose they serve for customers, i.e. linking different datasets on entity, instrument and industry level. While there could be differentiation based on the focus of the respective cross-reference tool, e.g. IHSM’s Entity Link being more focused on debt-related identifiers, and S&P’s offering being broader, the Notifying Party does not consider it necessary to conclude on this, as no concerns arise.

(C) The Commission’s assessment

(189) The Commission notes that customers may have a distinct view on which identifiers they need access to based on the underlying data they want to link, and hence identifier coverage is likely a very important feature of cross-reference tools. Respondents to the market investigation considered nevertheless that competition takes place at the level of cross-reference tools overall and no further segmentation is relevant. Respondents to the market investigation found that S&P and IHSM are close competitors in the market for cross-reference tools, but along with other competitors. This points to basically substitutable offerings.

(190) The market investigation otherwise did not provide any specific indications that the market for cross-reference tools could be narrower. The Commission therefore considers that the relevant product market for the purpose of this decision is the market for cross-reference tools.

6.2.4.9. Relevant geographic market – Cross reference tools

(A) The Commission precedents

(191) The Commission has not previously considered the geographic scope of the plausible market for cross-reference tools.

(B) The Notifying Party’s view

(192) Regarding cross-reference tools, the Notifying Party considers that a market comprising the global supply of cross-reference services would be appropriate given that those services are supplied on a global basis. Equally, demand is not driven by location of either the customer or the supplier.

(C) The Commission’s assessment

(193) The Commission considers the plausible market for cross-reference tools to be global for the following reasons.

(194) First, identifiers and sector classification schemes, which are inputs to cross-reference tools, are global markets.

(195) Second, respondents to the market investigation unanimously answered that they purchase cross-reference tools by comparing offers at worldwide level and that suppliers are active at this level.

(D) Conclusion

(196) For the purposes of this Decision, the Commission considers that cross-reference tools are a separate market which is global in scope.

6.2.5. Pricing and reference data

6.2.5.1. Overview and Parties’ activities

(197) Pricing and reference data can be related to different asset-classes: (i) corporate and sovereign bond pricing and reference data, (ii) municipal bond pricing and reference data, (iii) loan pricing and reference data (iv) securitized products pricing and reference data, and (v) CDS pricing and reference data.

(198) Pricing data can be based on actual transactions (i.e. price at which a certain asset was bought/sold), on quotes (i.e. firm offers to buy/sell a certain asset) or estimates or even extrapolations based on a small sample of those (“evaluated pricing”). Reference data contains terms and conditions of financial instruments which in certain asset classes, like loans, can change over time.

(199) Pricing and reference data can be supplied together or separately. Both pricing data and reference data can each be supplied as a ‘standard’ package (i.e. the customer gets a full basic dataset) or, depending on the customer’s needs, as a bespoke file containing only certain data elements (e.g. focussing on a specific geography).

(200) Pricing and reference data is always supplied using one or several identifiers to unambiguously distinguish and identify the financial instruments the data refers to. The data is used by a variety of different customers, for example investors, banks, asset servicers, market intelligence providers and credit rating agencies.

(201) IHSM is active in pricing and reference data for several types of financial instruments, namely (i) loan pricing and reference data, (ii) bond pricing and reference data, and (iii) CDS pricing and reference data.

6.2.5.2. Relevant product market – Pricing and reference data

(A) The Commission precedents

(202) The Commission defined pricing and reference data in LSEG/Refinitiv, although in that case, the focus and overlap was in consolidated non real-time pricing and reference data, which was considered a separate market, different from the market of packaged solutions, such as desktop services. In other previous decisions, the Commission considered discrete data content sets as plausible separate markets, given that they are not substitutable from a demand side perspective.

(B) The Notifying Party’s view

(203) The Notifying Party acknowledges that, despite some supply-side substitutability, pricing and reference data should be considered separate markets for different asset classes, due to limited demand-side substitutability. Within asset classes, the Notifying Party submits that no further segmentation is relevant, e.g. into different kinds of bonds (corporate, municipal, sovereign and securitized products), as there is full supply-side substitutability and customers generally tend to demand the full range, though the segmentation may not be entirely irrelevant for customers.

(204) Considering whether pricing and reference data are in general part of the same market, the Notifying Party considers that there is at least some supply side substitutability in the sense that many providers are active in both, pricing and reference data and the customers tend to overlap to a large extent. From a demand side perspective, the Notifying Party considers pricing and reference data to be complementary. In any case, the Notifying Party considers that the Commission does not need to conclude on the exact market definition with respect to an overall pricing and reference data market (for different asset classes) or separate pricing data and reference data markets. The Notifying Party has nevertheless provided market share data based on the most narrow plausible market definition, namely for pricing and reference data separately for all three relevant asset classes (loans, bonds, CDS).

(205) The Notifying Party further considers that customers generally do not distinguish, other than in exceptional circumstances, between evaluated and composite/mark-to-market pricing. According to the Notifying Party, these are two different price calculating methodologies which may be used separately or in combination in relation to some instruments (e.g. bonds and CDS) to ensure accuracy and as full coverage as possible, in particular when observable pricing data is not as readily available or prevalent. The Notifying Party does not consider separate markets for evaluated and non-evaluated pricing plausible.

(C) The Commission’s assessment

(206) The Commission considers that the evidence from the market investigation supports the Notifying Party’s view that pricing and reference data for different asset classes are separate markets. The responses to the market investigation even suggest that there is no supply side substitutability in the short term. For example, no competitor considers it feasible that a provider of bond or CDS pricing data can start producing loan pricing data within a short time such as 6 months or less. In fact, most respondents are either unable to estimate this or state that starting a pricing data service in a different asset class would take between 1 and 2 years and several million Euros of investment to get started. One competitor estimates: “It is our understanding that it would be difficult to begin producing loan pricing data in a short time. Loan markets are in many ways more illiquid than bond markets and ownership of loan data is not easily collected.” Another respondent emphasizes that most customers value data history, so a successful entrant in the loan pricing data market would require several (4-5) years of data history before becoming a relevant competitor. Furthermore, the respondent states: “It is also worth noting that even once the product gets to the marketplace, the new offering would still need to concord to either LXIDs or CUSIP IDs (both controlled by the combined entity) and the cost of obtaining those IDs likely will increase significantly for a competitive product.” In summary, entry into adjacent pricing and reference data markets seem to be difficult and from a demand-side perspective, pricing and reference data for different asset classes is not substitutable.

(207) With respect to the question whether pricing and reference data are part of the same market or separate markets, the evidence in the market investigation is mixed. The majority of responding customers purchase loan pricing and reference data separately, one third of the responding customers purchase loan pricing and reference data together, and a small minority purchases only loan pricing data.

(208) From a supply side perspective, the data sources for loan pricing and reference data are different. Loan pricing data is sourced from actively trading dealer banks and brokers, while loan reference data is sourced from Information Memoranda prepared by the agent banks syndicating a loan. One competitor also notices that both kinds of data present very different challenges: “It is important to delineate between pricing collection and reference data collection. While pricing data collection is much less complicated (and with much less data item collection) than reference data collection; obtaining pricing history is much more difficult than reference data historical collection. This is because there is document history available for about 60-65% of the loan facilities. The training and education component toward collecting the loan reference data is 5 times more difficult than collecting pricing data.”

(209) The Commission notes that given that not all providers provide both types of data, customers do not always buy both types of data or if they do, they might buy them separately, the data sources are different and require different expertise in terms of database management, the evidence would tend to point rather in the direction of separate pricing and reference data markets.

(210) With respect to the specific asset classes (CDS, loans, bonds), a further sub-segmentation could be relevant for bonds depending on the type of issuer (sovereign, municipal, corporate, securitized). However, customers often purchase data on different kinds of bonds together and suppliers’ products are comparable in the sense that they cover all those sub-segments at a similar quality level. On that basis, and given that no concerns arise with respect to any plausible sub-segment, the Commission considers that the exact product market definition with respect to bond pricing and/or reference data can be left open.

(211) The market investigation did not indicate that any narrower separate plausible markets may exist with respect to CDS pricing and/or reference data and loan pricing and/or reference data.

6.2.5.3. Relevant geographical market

(A) The Commission precedents

(212) The Commission considered in a previous decision that markets for discrete financial data content sets are at least EEA-wide or global. In LSEG/Refinitiv, the Commission considered the related market of consolidated non real-time pricing and reference data to be worldwide in scope.

(B) The Notifying Party’s view

(213) The Notifying Party consider that the relevant geographical market is global since suppliers may provide either pricing or reference data wherever they are located. Moreover, input data can be obtained remotely from publicly available sources that can then be digitally distributed to customers in any location. The Notifying Party considers this to apply to pricing and reference data for all asset classes relevant in this case, namely for loans, bonds and CDS.

(C) The Commission’s assessment

(214) The Commission considers the market for pricing and reference data for loans, bonds and CDS to be global in scope for the following reasons.

(215) First, the majority of competitors responding to the market investigation state that their company offers loan pricing and/or reference data at worldwide level, that their competitors are active on a global level too and that customers are located across the world. Some companies indicate that they set prices at a different level than global, but the Commission does not deem this a sufficiently important factor to consider the market narrower than global.

(216) Second, the large majority of customers purchase loan pricing and reference data by comparing offers worldwide, and consider the same suppliers are active worldwide.

(217) Third, the market investigation did not provide any evidence that the geographic scope for pricing and reference data in bonds and CDS should be different from that for loans.

(D) Conclusion

(218) For the purposes of this Decision, the Commission considers that the following plausible separate relevant markets exist: i) loan pricing data, ii) loan reference data, iii) bond pricing data, iv) bond reference data, v) CDS pricing data, vi) CDS reference data. All those plausible markets are global.

6.2.6. Desktop services

6.2.6.1. Overview and Parties’ activities

(219) Desktop services enable individual users to access content including real-time and non-real-time financial data, news and analytics spanning asset classes and geographies via a software container, or “front end”. Desktop services deliver data for viewing by humans on a screen. In addition to displaying financial data for users, desktop services often include decision support tools, workflow tools, and instant messaging capabilities. Desktop services sometimes offer access to trading capabilities and venues. Desktop services can take the form of either a web-delivered solution available on any computer, or a physical ‘desktop terminal’. Desktop services comprise comprehensive and fully integrated desktops (premium products), or more specific desktops with targeted content sets. In both cases, these can be referred to as "workstations" or "terminals". Various content and capabilities are licensed to end users for use either in or alongside integrated desktops (or on laptops or mobile devices) or separately.

(220) Desktop services cater to different customer needs. Such needs vary significantly, for example, between (i) on-trading floor customers (e.g. traders) who use desktop services to inform trading decisions in real-time and (ii) off-trading floor customers (e.g. asset managers) who use desktop services to provide advisory and investment services to institutional investors. While the basic desktop solution offered may be the same, its content can be tailored by customer segment.

(221) S&P supplies desktop services via SPGMI, such as its Market Intelligence platform (previously known as the Capital IQ platform). IHSM does not provide desktop services.

6.2.6.2. Relevant product market

(A) The Commission precedents

(222) In Blackstone/Thomson Reuters F&R Business, the Commission discussed a plausible market for desktop services, which would be separate from consolidated real-time datafeeds (“CRTD”), non-real-time datafeeds (“NRTD”), and discrete content datasets. The relevant product market definition was eventually left open.

(223) In LSEG/Refinitiv, the Commission considered that the market for desktop services is a separate market. Moreover, The Commission considered that the market should not be further sub-segmented by asset class or customer type.

(B) The Notifying Party’s view

(224) The Notifying Party considers the desktop services in relation to the distribution of financial data and indices in particular, arguing that the relevant product market might plausibly comprise, at its widest, the distribution of financial markets data overall, reflecting both the supply-side substitutability between the distribution of different financial datasets and the fact that many customers typically consume financial indices alongside other data. Such a market could be segmented by reference to aggregated desktop and datafeed offerings (consistent with Commission precedent on distribution channel). On a more narrow basis, the Notifying Party considers that a market comprising just the distribution of indices (again including segmented by desktop and datafeed) may also be plausible. The Notifying Party adds that there is no need to conclude on the precise scope of the product market due to the lack of concerns either way.

(C) The Commission’s assessment

(225) The Commission does not consider it appropriate to consider desktop services only in conjunction with the distribution of financial data, whether indices or otherwise. Desktop services can be sold both together with other products (data sets or datafeeds), but the package chosen varies by customer so there is no predominant pairing that could constitute a market. Moreover, desktop services are also sold and purchased separately. Content availability is only one factor among the top factors customers consider when selecting a desktop service; the others are the functionalities of the desktop, the user-friendliness, the messaging/community aspects and price. As such, in line with precedents, the Commission considers desktop services to constitute a separate product market.

(226) As for further segmentation of desktop services, competitors indicate that competitive dynamics do differ to some extent based on the target user (e.g. investment banker, asset manager, trader, etc.). However this question can be left open as no competition concerns arise regardless of further segmentation based on target user.

6.2.7. Non-real time datafeeds (“NRTDs”)

6.2.7.1. Overview and Parties’ activities

(232) Datafeeds, by contrast with desktop solution, deliver data in “raw” format, from which customers can build their own internal applications or portals. The data feed provision market is composed of real-time data feeds and non-real time data feeds.

(233) A real-time datafeed is a virtual pipeline that supplies continually updated financial market information. Real-time data feeds can be used as inputs for applications developed by banks and financial institutions, for example to allow for electronic or automatized algorithmic trading.

(234) There are two types of real-time datafeeds: consolidated and direct. Consolidated real-time data feeds (“CRTDs”) require the aggregation of feeds from various sources including exchanges into a single source. The CRTD is then delivered to the end-customer. Direct real-time data feeds connect an individual exchange or other data source with the end-customer.

(235) Non-real-time data is general financial and economic information such as historical pricing and reference data, macroeconomic data, company data. It may be updated several times a day or daily, but it does not satisfy the real-time data requirements as continually updated financial market information. Non-real-time data is primarily used for research and advisory purposes. It is distributed in (i) non-real-time data feeds, as (ii) discrete content datasets or through (iii) desktop solutions. NRTDs often combine various types of financial and economic information aggregated from various sources into a single stream of data. NRTDs may include delayed trading data, which is being disseminated more than 15 minutes after its generation, and/or historical data from trading venues and index providers.

(236) S&P offers a NRTD product in the form of XpressFeed, which has a market share of approximately [5-10]% in the market. IHSM has an index aggregator business, SOLA. SOLA is an index management tool that takes index data from multiple suppliers and ‘normalises’ them into a consolidated and standardised format for supply via a single datafeed. While it is not clear to what extent this specialized datafeed competes in the NRTD space, its market share would in any case be negligible at <[0-5]% .

6.2.7.2. Relevant product market

(A) The Commission precedents

(237) In Blackstone/Thomson Reuters F&R Business, the Commission discussed a plausible market for NRTDs, which would be separate from CRTDs, and discrete content datasets. The relevant product market definition was eventually left open.

(238) In LSEG/Refinitiv, the Commission considered that NRTDs constitute a relevant market, which is separate from direct real-time datafeeds and CRTDs.

(B) The Notifying Party’s view

(239) The Notifying Party considers datafeeds in relation to the distribution of financial data and indices in particular, arguing that the relevant product market might plausibly comprise, at its widest, the distribution of financial markets data overall, reflecting both the supply-side substitutability between the distribution of different financial datasets and the fact that many customers typically consume financial indices alongside other data. Such a market could be segmented by reference to aggregated desktop and datafeed offerings (consistent with Commission precedent on distribution channel). On a more narrow basis, the Notifying Party considers that a market comprising just the distribution of indices (again including segmented by desktop and datafeed) may also be plausible. The Notifying Party adds that there is no need to conclude on the precise scope of the product market due to the lack of concerns either way.

(C) The Commission’s assessment

(240) The Commission does not consider it appropriate to consider datafeeds only in conjunction with the distribution of financial indices, as they often combine various types of financial and economic information aggregated from various sources. The Commission’s market investigation did not provide indications to depart from the precedent. As such, in line with precedents, the Commission considers NRTDs to constitute a separate product market.

6.2.7.3. Relevant geographical market

(A) The Commission precedents

(241) In Reuters Instrument Codes, the Commission found that the relevant market for CRTDs is worldwide in scope.

(242) In Thomson/Reuters, the Commission considered the geographic scope of relevant markets involving discrete financial data content datasets to be at least EEA-wide and probably global.

(243) In LSEG/Refinitiv, the Commission considered the market for NRTDs to be worldwide in scope.

(B) The Notifying Party’s view

(244) The Notifying Party considers the market for datafeeds to be global in scope.

(C) The Commission’s assessment

(245) The Commission has found no evidence to depart from the precedent and Notifying Party’s view. As such, the geographic scope is considered to be global.

6.2.7.4. Conclusion

(246) For the purposes of this Decision, the Commission considers that a relevant plausible global market exists for NRTDs.

213Commission decision of 20 December 2012 in Case AT.39654, Reuters Instrument Codes, paragraphs 28ff. See also Commission decision of 20 July 2018 in Case M.8837, Blackstone / Thomson Reuters F&R Business, paragraphs 23-25.

214Commission decision of 19 February 2008 in Case M.4726, Thomson Corporation / Reuters Group, paragraph 106.

215Commission decision of 13 January 2021 in Case M.9564, London Stock Exchange Group / Refinitiv Business, paragraph 529.

216Notifying Party’s response to RFI 24, Annex 3a, paragraph 4.62.

6.2.8. Fundamentals data

6.2.8.1. Overview and Parties’ activities

(247) Fundamentals data consists of granular metrics that represent the primary characteristics and financial data necessary to determine the stability and health of a given company such as financial, cash flow and fund flow statements. This information is typically taken from public sources, such as filings, other regulatory publications or other sources such as (transcripts of) earning calls, before being aggregated and cleansed by data providers. The information is used by customers for a wide range of use cases such as research, risk assessment and analytics purposes.

(248) Only S&P is active in the supply of fundamentals data.

6.2.8.2. Relevant product market

(A) The Commission precedents

(249) In Thomson/Reuters, the Commission identified a separate relevant product market for fundamentals data. In LSEG/Refinitiv, the Commission confirmed the market definition.

(B) The Notifying Party’s view

(250) The Parties consider that an overall market for fundamentals data is a plausible market.

(C) The Commission’s assessment

(251) On the basis of the market investigation results and in light of its decisional practice, the Commission concludes that fundamentals data constitute a relevant market, for the following reasons. The market investigation does not provide any evidence that the relevant market definition for fundamentals data should depart from the Commission’s decision in Thomson /Reuters and LSEG/Refinitiv. Data vendors submit that they offer fundamentals data both separately and in a package with other products/services. The fact that there are commercial offers of both types of products from the same data vendor confirms that end-customers also purchase fundamentals data on a standalone basis. The Commission’s market investigation did not provide indications that the relevant market for fundamentals data should be sub-segmented further.

217Commission decision of 19 February 2008 in Case M.4726, Thomson Corporation / Reuters Group, paragraph 44.

218Commission decision of 13 January 2021 in Case M.9564, London Stock Exchange Group / Refinitiv Business, paragraph 415 – 418.

219Form CO on vertical relationships, paragraph 5.50.

220Replies to question 5 of Questionnaire 8.

6.2.8.3. Relevant geographical market

(A) The Commission precedents

(252) In Thomson /Reuters, the Commission considered the geographic scope of content sets, including fundamentals data, to be at least EEA-wide and probably global.

(253) In LSEG/Refinitiv, the Commission considered that the geographic scope of the relevant market for fundamentals data is worldwide or at least EEA-wide in scope.

(B) The Notifying Party’s view

(254) The parties consider the market for fundamentals data to be global in scope.

(C) The Commission’s assessment

(255) The market investigation confirmed the Commission’s decisional practice and the Notifying Party’s view for this market. Indeed, competitors confirmed that the market is global, as competitors’ offerings are available on a world-wide basis and prices are set at that level.

6.2.8.4. Conclusion

(256) For the purposes of this Decision, the Commission considers that a relevant plausible global market exists for fundamentals data.

6.2.9. Economic data

6.2.9.1. Overview and Parties’ activities

(257) Economic data is data describing the state of economies, generally at the country level. This data consists in interest rate information, GDP information and other macroeconomic data and can be historic (i.e. time series) or forecast data.

(258) IHSM supplies a range of economic data, including historic data and forecast data. These products have a wide range of end-uses, typically geared towards supporting strategic and tactical commercial decisions.

(259) S&P does not provide economic data, but is a customer of economic data, which is an input into credit ratings. Also, economic data is distributed via desktops and data feeds, and hence S&P distributes economic data.

221Commission decision of 19 February 2008 in Case M.4726, Thomson Corporation / Reuters Group, paragraph 106.

222Commission decision of 13 January 2021 in Case M.9564, London Stock Exchange Group / Refinitiv Business, paragraph 424.

223Form CO on vertical relationships, paragraph 5.51.

6.2.9.2. Relevant product market

(A) The Commission precedents

(260) In Thomson/Reuters, the Commission identified a separate relevant product market for time series of economic data. In LSEG/Refinitiv, the Commission confirmed the market definition.

(B) The Notifying Party’s view

(261) In the Parties’ view, the relevant market might comprise the supply of economic data at its widest. However, reflecting the lack of clear supply or demand-side substitutability, the Parties consider potential separate markets for the supply of historic, or time series, economic data and forecast economic data. From a supply-side perspective, while some providers of economic data provide both historic data and forecast data, many focus on one area, suggesting a potential lack of significant supply-side substitutability. In particular, the supply of forecast data requires particular capabilities and expertise that are not required of suppliers of historic data. From a demand-side perspective, while some customers may require both historic and forecast data, others will predominantly require only one type and are unlikely to consider the other as substitutable.

(C) The Commission’s assessment

(262) On the basis of the evidence available to it and in light of its decisional practice, the Commission concludes for the purposes of this case that time series of economic data and forecast economic data constitute plausible relevant markets. The market investigation does not provide any evidence that the relevant market definition for time series of economic data should depart from the Commission’s decision in Thomson/Reuters and LSEG/Refinitiv. In terms of supply-side substitutability, the Commission notes that in the top 5 suppliers each in time series of economic data and in forecast economic data, only 2 overlap. In terms of demand-side substitutability, while some customers may need both sets, the use cases of backward looking and forward looking data are different and therefore customers are not likely to consider them substitutable; forecast economic data may inform strategic thinking while time series of economic data may be inputs for modelling and/or backtesting. Nevertheless, the question of whether economic data should be segmented into forecast economic data and time series of economic data can be left open for the purposes of this case as no competition concerns arise regardless of the precise definition.

225Commission decision of 19 February 2008 in Case M.4726, Thomson Corporation / Reuters Group, paragraph 44.

226Commission decision of 13 January 2021 in Case M.9564, London Stock Exchange Group / Refinitiv Business, paragraph 415 – 418.

227Form CO on vertical relationships, paragraph 3.40.

6.2.9.3. Relevant geographical market

(A) The Commission precedents

(263) In Thomson/Reuters and LSEG/Refinitiv, the Commission considered the geographic scope of time series of economic data, to be at least EEA-wide and probably global.

(B) The Notifying Party’s view

(264) The parties consider the market for economic data to be global in scope since suppliers may provide either loan pricing or loan reference data wherever they are located. Moreover, input data can be obtained remotely from publicly available sources that can then be digitally distributed to customers in any location.

(C) The Commission’s assessment

(265) The Commission has found no evidence to depart from the decisional practice and the Notifying Party’s view. The Commission does not need to conclude on the definition for the purposes of this case as no concerns arise regardless of the precise market definition.

6.2.9.4. Conclusion

(266) For the purposes of this Decision, the Commission considers that a relevant plausible global market exists for economic data, potentially segmented further as historic economic data and forecast economic data.

6.2.10. Sector classification schemes

6.2.10.1. Overview and Parties’ activities

(267) Sector classification schemes, or industry taxonomies as they may be called, are systems for categorisation and comparison of companies according to their activities, industry and sector. Each scheme is composed of levels that increase the granularity of the categorisation – e.g. industry, sector or sub-sector level. A company is given an identifying code that determines its category in each level. Sector classification schemes are used in research, portfolio management, asset allocation and index creation.

(268) GICS is a taxonomy system for classifying companies globally into pre-determined sectors, industries and other groupings. It was developed jointly by MSCI and S&P, and is now distributed independently by both entities […].

228Commission decision of 19 February 2008 in Case M.4726, Thomson Corporation / Reuters Group, paragraph 106.

229Commission decision of 13 January 2021 in Case M.9564, London Stock Exchange Group / Refinitiv Business, paragraph 852.

230Form CO on vertical relationships, paragraph 3.41.

6.2.10.2. Relevant product market

(A) The Commission precedents

(269) In LSEG/Refinitiv, the Commission concluded that sector classification schemes constitute a separate relevant product market.

(B) The Notifying Party’s view

(270) In the Notifying Party’s view, the relevant market might comprise the supply of sector classification schemes, but that no conclusion need be drawn as no competition concerns arise on any basis.

(C) The Commission’s assessment

(271) On the basis of the evidence available to it and in light of its decisional practice, the Commission concludes for the purposes of this case that sector classification schemes constitute a separate relevant product market. The market investigation does not provide any evidence that the relevant market definition should depart from the Commission’s decision in LSEG/Refinitiv. From a supply-side perspective, of the schemes available in the market, two are publicly available and all other schemes but one are available to license on a standalone basis.

6.2.10.3. Relevant geographical market

(A) The Commission precedents

(272) In LSEG/Refinitiv, the Commission considered the geographic scope of sector classification schemes to be global.

(B) The Notifying Party’s view

(273) The parties consider the market for sector classification schemes to be global in scope.

(C) The Commission’s assessment

(274) The Commission has found no evidence to depart from the decisional practice and the Notifying Party’s view. As such, the Commission concludes for the purposes of this case that the geographic scope of the sector classification schemes market is global.

6.2.10.4. Conclusion

(275) For the purposes of this Decision, the Commission considers that a relevant plausible global market exists for sector classification schemes.

231Commission decision of 13 January 2021 in Case M.9564, London Stock Exchange Group / Refinitiv Business, paragraph 415 – 418.

232Form CO on vertical relationships, paragraph 5.76.

233Commission decision of 13 January 2021 in Case M.9564, London Stock Exchange Group / Refinitiv Business, paragraph 490.

234Form CO on vertical relationships, paragraph 5.76.

6.2.11. Leveraged loan market intelligence

6.2.11.1. Overview and Parties’ activities

(276) Market intelligence concerns the supply of information, data, news, tools and analytics generally with respect to a particular sector of the economy, e.g. automotive market intelligence, which would provide data, news and analytics with respect to vehicle production, supply chains and sales.

(277) Both IHSM and S&P are active in market intelligence more broadly (including in commodities market intelligence, see Section 7.3.2), though only S&P is active in market intelligence related to financial markets, concretely debt markets. S&P provides leveraged loan market intelligence in the form of its Leveraged Commentary and Data (LCD) product. LCD is a subscription-based product providing news and research on the leveraged loan markets, including close to real-time news on latest developments, details on specific deals (“deal sheet data” incl. indicative pricing, credit ratings and arranger identities for a given loan) and aggregate market trends.

6.2.11.2. Relevant product market

(A) The Commission precedents

(278) The Commission previously considered the supply of (primarily financial) market intelligence products. In Thomson/Reuters, the Commission considered that individual content sets are not substitutable for one another since they respond to different and well defined needs of customers, are often traded separately, and can be considered on a standalone basis. Within individual content sets, News was considered a relevant separate market.

(279) The market for News in LSEG/Refinitiv was considered a separate product market and described as financial and business-related news content that provides investors with information to make investment decisions. News were considered to possibly be comprehensive or limited to certain sectors, countries or asset classes. However, this further plausible segmentation of the News market was not further considered in those previous decisions.

(B) The Notifying Party’s view

(280) The Notifying Party considers that from a supply-side perspective, barriers to entry in debt-related market intelligence is low and there are a number of companies active in this wider space that could quickly enter. However, the Notifying Party acknowledges that from a demand side perspective, more general or other specialized debt-related market intelligence may not be substitutable with leveraged loan market intelligence for customers and therefore agrees that the narrower market for leveraged loan market intelligence may be considered a plausible market for the purposes of this case.

(C) The Commission’s assessment

(281) The Commission considers based on the answers it received to the market investigation, that leveraged loan market intelligence is a plausible separate market for the following reasons:

(282) First, competitors confirm the Notifying Party’s view that while leveraged loan market intelligence is not isolated and entirely independent from other debt-related market intelligence, there is a sufficient distinction compared to other debt-related market intelligence to consider it a separate market. One competitor summarizes it like this: “Loan market intelligence feeds up into the broader debt markets – but loan investors are a segment of that who need a deeper dive into the niche driven, private world of loans.”

(283) Second, customers are more split, but the majority believes that broader debt-related market intelligence and leveraged loan market intelligence are not interchangeable. Customers mentioned the following products/providers as substitutable: Fitch’s LevFin Insights and Capital Structure, S&P’s LCD, Bloomberg, Debtwire, Reorg. When asked which products or from which providers customers purchase leveraged loan market intelligence, several also indicate purchasing Refinitiv’s LPC product. This broadly corresponds to the competitors submitted by the Notifying Party as competing within leveraged loan market intelligence with S&P.

(284) Third, the Commission notes that respondents to the market investigation rate the above mentioned providers as broadly similar in terms of their competitive strength.

6.2.11.3. Relevant geographical market

(A) The Commission precedents

(285) The Commission’s previous decisions considered market intelligence/news markets to be at least EEA-wide and probably global, or in a later decision global or at least EEA-wide.

(B) The Notifying Party’s view

(286) The Notifying Party considers the market for leveraged loan market intelligence to be global. The physical location of customers and providers is not important in terms of the decisions to whom providers of leveraged loan market intelligence offer their products and from whom customers buy.

(C) The Commission’s assessment

(287) The Commission considers based on the answers it received to the market investigation, that the leveraged loan market intelligence market is global for following reasons:

(288) First, the majority of providers offer their products worldwide and state that their competitors are active on a global level too. While not all competitors set prices for their products at global level, the Commission does not consider this a sufficient element that would put into doubt the overall conclusion on the geographic level of this market.

(289) Second, the large majority of customers purchases leveraged loan market intelligence by comparing offers at global level and states that the same suppliers are active on a worldwide level.

(290) Third, even though some providers may offer their products in separate packages targeting specific geographic areas (e.g. “EMEA”, “Asia”), this does not mean that only customers located in those specific areas are interested or buying those products. It would seem that this is a way of categorizing the product and enabling customers to assemble a product specific for their needs.

6.2.11.4. Conclusion

(291) For the purposes of this Decision, the Commission considers that there is a separate relevant market for leveraged loan market intelligence and that this market is worldwide in scope.

6.2.12. Stock selection and strategy services

6.2.12.1. Overview and Parties’ activities

(292) Stock selection and strategy services provide investment strategies, screens and signals, to assist in designing investment strategies and stock selections for investment managers, banks and others. These services are based on corporate, asset and industry-specific information and both Parties are active in this market. S&P is active with its product Alpha Factor Library and IHSM with its product Research Signals.

6.2.12.2. Relevant product market

(A) The Commission precedents

(293) The Commission has not considered the relevant product market for stock selection and strategy services in any previous decision.

(B) The Notifying Party’s view

(294) The Parties considers that for present purposes a market for the supply of stock selection and strategy services could be considered as a distinct product market. However, the Notifying Party also considers that from a supply-side perspective, the barriers to entry are likely to be relatively low such that providers of financial market intelligence (who will already have access to the necessary inputs to provide stock selection and strategy services, such as corporate, asset and industry-specific information) would be able to commence the supply of stock selection and strategy services quickly and easily. Similarly, from a demand-side perspective, there is likely to be a degree of substitutability with adjacent financial market intelligence, since customers can use such intelligence to construct their own stock selection and strategies.

(C) The Commission’s assessment

(295) The market investigation provided no evidence putting into question the Notifying Party’s view. However, the Commission does not need to conclude on the definition for the purposes of this case as no concerns arise regardless of the precise market definition.

6.2.12.3. Relevant geographical market

(A) The Commission precedents

(296) The Commission has not considered the relevant geographic market for stock selection and strategy services in any previous decision.

(B) The Notifying Party’s view

(297) The Parties consider the market for the supply of selection stock and strategy services to be global in scope.

(C) The Commission’s assessment

(298) The market investigation has not presented any evidence putting into question the Notifying Party’s view. However, the geographic scope can be left open as no competition concerns arise in any case.

6.2.12.4. Conclusion

(299) For the purposes of this Decision, the Commission considers that a relevant plausible global market exists for stock selection and strategy services.

6.2.13. Issuer solutions

6.2.13.1. Overview and Parties’ activities

(300) IHSM's Issuer Solutions product, BD Corporate, is a customer relationship management (CRM) or investor relationships platform that helps customers prospect for new investors, prepare for roadshows, track current investors and their holdings, manage investor relations and report on the success of investor outreach efforts. BD Corporate is a multi-asset class platform.

6.2.13.2. Relevant product market

(A) The Commission precedents

(301) The Commission has not considered the relevant product market for issuer solutions in any previous decision.

(B) The Notifying Party’s view

(302) In the Parties’ view, the relevant product market for issuer solutions is the supply of CRM investor relationship platforms to issuers.

(C) The Commission’s assessment

(303) The market investigation provided no evidence putting into question the Notifying Party’s view. However, the Commission does not need to conclude on the definition for the purposes of this case as no concerns arise regardless of the precise market definition.

6.2.13.3. Relevant geographical market

(A) The Commission precedents

(304) The Commission has not considered the relevant geographic market for issuer solutions in any previous decision.

(B) The Notifying Party’s view

(305) The Parties consider the geographic market for the supply of issuer solutions to be global, as most providers are active globally and many customers use such platforms to manage global relationships.

(C) The Commission’s assessment

(306) The Commission has found no evidence putting into question the Notifying Party’s view. However, the Commission does not need to conclude on the definition for the purposes of this case as no concerns arise regardless of the precise market definition.

6.2.13.4. Conclusion

(307) For the purposes of this Decision, the Commission considers that a relevant plausible global market exists for the supply of CRM investor relationship platforms to issuers.

6.2.14. Investor event management solutions

6.2.14.1. Overview and Parties’ activities

(308) IHSM's Investor event management product, Iplanner, helps sell-side institutions manage roadshows and coordinate connectivity between corporates, industry experts and investors,.

6.2.14.2. Relevant product market

(A) The Commission’s previous decisions

(309) The Commission has not considered the relevant product market for investor event management solutions in any previous decision.

(B) The Notifying Party’s view

(310) In the Parties’ view, the relevant product market is platforms and software designed to facilitate communication and connections in the marketplace between issuers, experts, and investors. IHSM's key competitors in this market include Cvent, Aventri, Dealogic, and MeetMax. However, it is not necessary to conclude on the precise product market definition, as irrespective of how the market is defined there are no competition concerns.

(C) The Commission’s assessment

(311) The market investigation provided no evidence that the Commission should depart from the Notifying Party’s view. The Commission does not need to conclude on the definition for the purposes of this case as no concerns arise regardless of the precise market definition.

6.2.14.3. Relevant geographical market

(A) The Commission’s previous decisions

(312) The Commission has not considered the relevant geographic market for investor event management solutions in any previous decision.

(B) The Notifying Party’s view

(313) The Parties consider the geographic market for the supply of investor event management solutions to be global, as most providers are active globally and many customers use such solutions on a global basis.

(C) The Commission’s assessment

(314) The Commission has found no evidence to depart from the Notifying Party’s view. The Commission does not need to conclude on the definition for the purposes of this case as no concerns arise regardless of the precise market definition.

6.2.14.4. Conclusion

(315) For the purposes of this Decision, the Commission considers that the relevant plausible market is the global market for investor event management solutions.

6.2.15. Issuance platforms

6.2.15.1. Overview and Parties’ activities

(316) An issuance platform is a workflow tool used primarily by investment banks and investors in the issuance process of different assets, covering the stages of book-building and price discovery, when investors express their interest, through document management, to the actual issuance. Specifically, a municipal bond platform is a workflow tool specifically for US municipal bond issue, catering largely to US investors and underwriters; a fixed income book-building platform includes workflow tools that are used in the issuance of corporate (as opposed to municipal) fixed income assets, and an equity book-building platform includes workflow tools that are used in connection with the issuance of equities through IPOs.

(317) IHSM provides a municipal bond issuance platform, a corporate bond issuance platform and an equity issuance platform. S&P is not active in issuance of workflow solutions.

6.2.15.2. Relevant product market

(A) The Commission precedents

(318) The Commission has not considered the relevant product market for issuance platforms in any previous decision.

(B) The Notifying Party’s view

(319) The Notifying Party submits that there are separate product markets for issuance platforms for each of (i) corporate fixed income; (ii) municipal bonds; (iii) equities and (iv) syndicated loans. In particular, IHSM competes with a different competitor set within each of these asset classes, and the issuance platforms each have different use cases and customers based on the asset class that they cater for.

(C) The Commission’s assessment

(320) Based on the evidence available to it, the Commission considers issuance platforms to be a product market separate from other platforms and financial workflow solutions. Customers purchase issuance platforms separately from other products/services, and most suppliers in this market are not active elsewhere. There appears to be some self-supply by a few of the larger investment banks, but it does not appear to be a substitute to third-party issuance platforms for customers already using the latter.

(321) Customers indicate that they purchase issuance platforms separately for different asset classes, such as equities, corporate bonds and municipal bonds, as there is differentiation due to the specificities of the asset class and there are different teams at the customer entity dealing with different asset classes. Indeed, apart from IHSM, other suppliers each compete in only one asset class. Further, there is some evidence that customers purchase parts of IHSM’s issuance platforms selectively, and not always as a package. Nevertheless, the Commission considers that the question of further segmentation of issuance platforms (by asset class and/or by different parts of the platform) can be left open for the purposes of this case, as no competition concerns arise regardless of further segmentation.

6.2.15.3. Relevant geographical market

(A) The Commission precedents

(322) The Commission has not considered the relevant geographic market for issuance platforms in any previous decision.

(B) The Notifying Party’s view

(323) The Notifying Party submits that the geographic market definition is global for the issuance platforms in each asset class, with the exception of the market for municipal bond issuance platforms, which is US-wide as municipal bond issuance platforms are used exclusively in connection with the issuance of US municipal bonds. Otherwise, in corporate fixed income, equities and syndicated loans, these platforms are used for global issuances and customers are generally financial institutions that are active on a global scale.

(C) The Commission’s assessment

(324) The Commission has found no evidence putting into question the Notifying Party’s view. Indeed, customers confirmed that the market is global; they compare offers at the world-wide level and suppliers are active at the world-wide level.

6.2.15.4. Conclusion

(325) For the purposes of this Decision, the Commission considers that a relevant plausible global market exists for issuance platforms, potentially segmented further by asset class and/or by different functional parts of the platform.

6.2.16. Loan administration solutions

6.2.16.1. Overview and Parties’ activities

(326) IHSM’s Wall Street Office product (“WSO”) is one of a number of loan administration workflow tools available to managers of loan portfolios that are designed to reduce the burden of the manual processes that arise in respect of loan portfolio administration, and which are otherwise managed through the use of excel, databases, and in-house applications. Specifically, WSO provides a workflow tool for asset management, reporting, and CLO compliance throughout the loan lifecycle i.e. middle- and back- office operations (as opposed to any deal origination and syndication which may be conducted by the front office). Customers of WSO therefore include asset managers, agents and trustees, using the product to manage and report on their complex loan portfolios. Participants can choose either to fulfil these solutions in-house or to outsource the provision of services to a third party. S&P is not active in loan administration solutions.

6.2.16.2. Relevant product market

(A) The Commission’s previous decisions

(327) The Commission has not considered the relevant product market for loan administration solutions in any previous decision.

(B) The Notifying Party’s view

(328) The Notifying Party submits that the appropriate frame of reference is loan administration solutions, comprising both software and services providers as well as self-supply, across all loan asset classes.

(329) Regarding the potential segmentation between software and services, the Notifying Party notes that when customers choose to outsource the provision of loan administration solutions, there are two options available, depending on how much of their back- and middle-office administration they would like to keep in-house: (i) Some customers purchase software only (e.g. WSO Software), and provide their own services; and (ii) Some customers purchase combined software and services from a single provider (e.g. WSO Services). Where these customers choose WSO Services to provide this solution, they will also receive the WSO software module. WSO Software customers make up approximately […] of all WSO customers and […] of WSO revenues; WSO Services customers (which, as noted above would also have WSO software as part of their solution) make up approximately […] of all WSO customers and […] of WSO revenues. Customers are increasingly […]. WSO Services faces competition from competing services providers, including (for the service element) those which may use WSO software. Similarly, its standalone software is constrained both by competing software providers and by other providers of a combined software and services solution. From a supply side perspective it is relatively easy for providers of loan administration software to expand into offering loan administration services, which essentially consists of the provision of people (in particular as in both cases the necessary data is procured and licensed by the client).

(330) Regarding the inclusion of self-supply in the market, the Notifying Party explains that self-supply of loan administration services exercises a significant constraint on its provision of both WSO Software and Services. A number of significant market participants self-supply loan administration solutions and many of the largest customers which currently choose to outsource loan administration services could similarly decide to take these services back in-house (whether using an in-house software solution, a third party software solution or Excel).

(331) Lastly, regarding possible segmentation by loan asset type, the Notifying Party adds that WSO and its competitors provide administration services for different types of loans, in particular private debt and liquid loans: (i) A loan is considered to be liquid when it is often traded and priced broadly by the market participants. A liquid loan is likely to have many lenders. (ii) A loan is considered to be private debt when it is agreed between two or very few parties/lenders, which results in illiquidity, lack of trading, and lack of pricing. Due to the nature of private debt, information about loan sizes, participants and offerings is less readily available. However, the Notifying Party submits that there is not a clear distinction between liquid and private loans; a loan could be issued between a few parties (rendering it private debt) but could subsequently gain liquidity if additional lenders are invited to participate. WSO and its main competitors are active across both private debt and liquid loans. Only minor alterations are required to software originally purposed for liquid loans to also serve private loans and there are no significant differences in competitive positions of providers as between private debt and liquid loans.

(C) The Commission’s assessment

(332) Based on the evidence available to it, for the purposes of the present case, the Commission considers loan administration solutions to be a product market separate from other financial solutions. While one supplier considers loan administration solutions to be part of a wider “credits/alternatives” market, most consider them to be separate from other products. One competitor explains “loan software/services for the bank debt market are marketed as a distinct set of solutions separate from other products/services” (emphasis added).

(333) Regarding segmentation between software and services, another supplier summarizes “There are supplier[s] that offer both software and offer services. There are also some that only offer software and some only services. Customers can choose to buy software from several suppliers and services from several as well or can buy both from one.” Several other competitors state that traded loan software and traded loan services are two separate markets, “though they are very much integrated and serve a common purpose”. Regarding self-supply, there is indication that some customers can outsource, for instance, software while continuing to self-supply services. Nevertheless, the Commission considers that the question of further segmentation (by software vs services or by loan asset type) and the inclusion or exclusion of self-supply can be left open for the purposes of this case, as no competition concerns arise regardless of further segmentation.

6.2.16.3. Relevant geographical market

(A) The Commission’s previous decisions

(334) The Commission has not considered the relevant geographic market for loan administration solutions in any previous decision.

(B) The Notifying Party’s view

(335) The Notifying Party submits that the appropriate geographic market is global, since most providers are active globally and many customers use loan administration solutions to manage their loans globally.

(C) The Commission’s assessment

(336) The Commission has found no evidence putting into question the Notifying Party’s view. Indeed, competitors confirmed that the market is global; they set prices at the world-wide level and suppliers are active at the world-wide level.

6.2.16.4. Conclusion

(337) For the purposes of this Decision, the Commission considers that a relevant plausible global market exists for loan administration services.

6.2.17. Digital design for financial services

6.2.17.1. Overview and Parties’ activities

(338) IHSM's Markit Digital is a Software as a Service (SaaS) that offers solutions for the financial services industry to aggregate, normalise, enhance and display financial information via web services, websites and native mobile apps. These solutions fall into three categories: custom web/mobile development and hosting, UX (user experience)/UI (user interface) design, and digital advertising. Markit Digital provides these solutions in a way that can be easily integrated into clients’ existing offerings.

6.2.17.2. Relevant product market

(A) The Commission precedents

(339) The Commission has not previously considered the relevant market for digital design for financial services.

(B) The Notifying Party’s view

(340) The Notifying Party considers the product market to be the market for custom web/mobile development and hosting services given in particular the lack of demand-side substitutability with other products / services.

(C) The Commission’s assessment

(341) The market investigation provided no evidence putting into question the Notifying Party’s view regarding the further segmentation of digital design for financial services. However, the Commission does not need to conclude on the definition for the purposes of this case as no concerns arise regardless of the precise market definition.

6.2.17.3. Relevant geographical market

(A) The Commission precedents

(342) The Commission has not previously considered the relevant market for digital design for financial services.

(B) The Notifying Party’s view

(343) The Notifying Party considers the geographic market for digital design for financial services as global, on the basis that IHSM and its competitors offers these services to global customers without any differences based on their region, and IHSM's customers are active on global basis such that their demand is not driven by their own location or that of suppliers.

(C) The Commission’s assessment

(344) The market investigation provided no evidence putting into question the Notifying Party’s view regarding the geographic scope of digital design for financial services. However, the Commission does not need to conclude on the definition for the purposes of this case as no concerns arise regardless of the precise market definition.

6.2.17.4. Conclusion

(345) For the purposes of this Decision, the Commission considers that a relevant plausible global market exists for digital design for financial services.

6.2.18. Equities and regulatory reporting (Dividend forecasting services)

6.2.18.1. Overview and Parties’ activities

(346) IHSM’s dividend forecasting service provides independent estimates of the amount and timing of dividend payments, allowing customers to better understand how companies are performing and what their projected dividends are. Dividend estimates for global securities are based on equity research, market announcements and unique quantitative insight, covering over 28,000 stocks. In addition to this research-based methodology, IHSM also applies advanced analytics and predictive modelling to predict company dividends.

6.2.18.2. Relevant product market

(A) The Commission precedents

(347) The Commission has not considered the relevant product market for equities and regulatory reporting (or dividend forecasting services specifically) in any previous decision.

(B) The Notifying Party’s view

(348) The Notifying Party considers the relevant product market to be the market for dividend forecasting services given in particular the lack of obvious demand-side substitutability with other products / services.

(C) The Commission’s assessment

(349) The market investigation provided no evidence putting into question the Notifying Party’s view. In any case, the Commission does not need to conclude on the definition for the purposes of this case as no concerns arise regardless of the precise market definition.

6.2.18.3. Relevant geographic market

(A) The Commission precedents

(350) The Commission has not considered the relevant geographic market for equities and regulatory reporting (or dividend forecasting services specifically) in any previous decision.

(B) The Notifying Party’s view

(351) The Notifying Party considers the geographic scope of the market for dividend forecasting to be global, since IHSM provides forecasts for stocks on a global basis, suppliers more generally do not need to be physically situated in a particular location in order to provide these forecasting services. Moreover, customer demand is driven by quality of service rather than customer or supplier location.

(C) The Commission’s assessment

(352) The market investigation has not presented any evidence putting into question the Notifying Party’s view. However, the geographic scope can be left open as no competition concerns arise in any case.

6.2.18.4. Conclusion

(353) For the purposes of this Decision, the Commission considers that a relevant plausible global market exists for equities and regulatory reporting.

6.2.19. Investor and administration services

6.2.19.1. Overview and Parties’ activities

(354) IHSM’s Investor and Administration services product, Profile Builder, is a tool for producing buy side investor profiles. Profile Builder provides insight into investors by incorporating global ownership data, contacts and biographical content from IHSM's BD (formerly Big Dough) database.

6.2.19.2. Relevant product market

(A) The Commission precedents

(355) The Commission has not previously considered the relevant product market for investor and administration services in any previous decision.

(B) The Notifying Party’s view

(356) The Parties submit that the relevant product market for investor and administration services is the supply of tools for producing buy-side investor profiles. Segmentation by asset class would not be appropriate from a supply-side perspective, as both IHSM and its competitors provide such tools for both fixed income and equity investors (although demand-side substitutability may be more limited). However, it is not necessary to conclude on the precise product market definition, as irrespective of how the market is defined there is no risk of foreclosure for the reasons set out below.

(C) The Commission’s assessment

(357) The market investigation provided no evidence putting into question the Notifying Party’s view. In any case the Commission does not need to conclude on the definition for the purposes of this case as no concerns arise regardless of the precise market definition.

6.2.19.3. Relevant geographical market

(A) The Commission precedents

(358) The Commission has not specifically considered the relevant geographic market for investor and administration services in any previous decision.

(B) The Notifying Party’s view

(359) The Parties consider that relevant geographic market for the supply of investor and administration services is global as most providers are active globally and investor and administration services utilise global data.

(C) The Commission’s assessment

(360) The market investigation has not presented any evidence putting into question the Notifying Party’s view. However, the geographic scope can be left open as no competition concerns arise in any case.

6.2.19.4. Conclusion

(361) For the purposes of this Decision, the Commission considers that the relevant market is the supply of investor and administration services, which is global in scope.

6.2.20. Institutional holdings/investor data

6.2.20.1. Overview and Parties’ activities

(362) IHSM's institutional holdings product, BD Advanced, provides sell-side customers with data and tools for prospecting and targeting buy-side institutions. BD Advanced incorporates cross-asset class data on potential clients' investment focusses, asset allocation and cross asset class holdings.

6.2.20.2. Relevant product market

(A) The Commission precedents

(363) In Thomson/Reuters, the Commission considered that individual content sets are not substitutable for one another since they respond to different and well defined needs of customers and are often traded separately, and can be considered on a standalone basis.

(364) The Commission has not specifically considered the relevant product market for institutional holdings/investor data in any previous decision.

(B) The Notifying Party’s view

(365) The Parties submit that the relevant product market for institutional holdings / investor data is the supply of data for prospecting and targeting buy-side institutions.

The market should not be segmented by asset class as BD Advanced is designed as a cross-asset solution covering equity, fixed income, derivatives and corporate issuer data, and competitor offerings are similarly asset-agnostic. However, it is not necessary to conclude on the precise product market definition, as irrespective of how the market is defined there is no risk of foreclosure for the reasons set out below.

(C) The Commission’s assessment

(366) The market investigation provided no evidence that the Commission should depart from the decisional practice regarding content sets, or putting into question the Notifying Party’s view. In any case the Commission does not need to conclude on the definition for the purposes of this case as no concerns arise regardless of the precise market definition.

6.2.20.3. Relevant geographical market

(A) The Commission precedents

(367) In Thomson/Reuters, the Commission considered the geographic scope of content sets to be at least EEA-wide and probably global.

(368) The Commission has not specifically considered the relevant geographic market for institutional holdings/investor data in any previous decision.

(B) The Notifying Party’s view

(369) The Parties consider that relevant geographic market for the supply of institutional holdings/investor data is global as most providers are active globally.

(C) The Commission’s assessment

(370) The market investigation has not presented any evidence to depart from the decisional practice regarding content sets, or putting into question the Notifying Party’s view. However, the question of whether the scope is EEA-wide or global can be left open as no competition concerns arise in either case.

6.2.20.4. Conclusion

(371) For the purposes of this Decision, the Commission considers that the relevant market is the supply of institutional holdings/investor data, which is at least EEA-wide in scope.

Notifying Party’s reply to RFI 24, Annex 3a; Form CO, Chapter on Vertical Relationships, paragraph 3.114.

Commission decision of 19 February 2008 in Case M.4726, Thomson Corporation / Reuters Group, paragraph 106.

Notifying Party’s reply to RFI 24, Annex 3a; Form CO, Chapter on Vertical Relationships, paragraph 3.115.

72

6.2.21. Managed corporate actions data

6.2.21.1. Overview and Parties’ activities

(372) Corporate actions are events that affect the securities of a given company, such as dividends, stock splits, M&A, disposals or spin-offs. IHSM is active in the supply of managed corporate actions data, which S&P uses as an input into its equity indices. S&P also provides CUSIPs as a (peripheral) input to managed corporate actions data.

6.2.21.2. Relevant product market

(A) The Commission precedents

(373) In Thomson/Reuters, the Commission considered that individual content sets are not substitutable for one another since they respond to different and well defined needs of customers and are often traded separately, and can be considered on a standalone basis.

(374) The Commission has not specifically considered the relevant product market for managed corporate actions data in any previous decision.

(B) The Notifying Party’s view

(375) The Parties consider that a market comprising the supply of managed corporate actions data alone is relevant even though from a supply-side perspective, many suppliers of other types of financial market intelligence (e.g. pricing and reference data, company fundamentals data) are in principle capable of supplying managed corporate actions data.

(C) The Commission’s assessment

(376) The market investigation provided no evidence that the Commission should depart from the decisional practice regarding individual content sets generally being separate markets; as such, considering the specific nature of the data in question and the limited demand-side substitutability, and in line with the Notifying Party’s view, the Commission considers that managed corporate actions data is a plausible product market. The Commission does not need to conclude on the definition for the purposes of this case as no concerns arise regardless of the precise market definition.

6.2.21.3. Relevant geographical market

(A) The Commission precedents

(377) In Thomson/Reuters, the Commission considered the geographic scope of content sets to be at least EEA-wide and probably global.

(378) The Commission has not specifically considered the relevant geographic market for managed corporate actions data in any previous decision.

(B) The Notifying Party’s view

(379) The Parties consider that the supply of managed corporate actions data takes place on a global basis.

(C) The Commission’s assessment

(380) The market investigation has not presented any evidence to depart from the decisional practice regarding content sets or putting into question the Notifying Party’s view. However, the question of whether the markets are EEA-wide or global in scope can be left open as no competition concerns arise in either case.

6.2.21.4. Conclusion

(381) For the purposes of this Decision, the Commission considers that the relevant market is managed corporate actions data, which is at least EEA-wide in scope.

6.2.22. Global securities financing data

6.2.22.1. Overview and Parties’ activities

(382) IHSM supplies global securities financing data, which comprises data tracking short-selling of securities and institutional fund activity. These data, which cover global securities in the lending programmes of institutional funds and details on securities transactions, help customers manage securities lending programmes, manage collateral, optimise trading performance and enhance investment decision making. IHSM uses equity indices as an input into this dataset

6.2.22.2. Relevant product market

(A) The Commission precedents

(383) In Thomson/Reuters, the Commission considered that individual content sets are not substitutable for one another since they respond to different and well defined needs of customers and are often traded separately, and can be considered on a standalone basis.

(384) The Commission has not specifically considered the relevant product market for global securities financing data in any previous decision.

(B) The Notifying Party’s view

(385) The Parties consider a market for the supply of global securities financing data to be a plausible market given the limited supply-side and demand-side substitutability. From a demand-side perspective, customers are unlikely to consider that other forms of dataset are substitutable with global securities financing data, which fulfils a specific purpose and has specific requirements and inputs such as data relating to lending programmes and specific securities-related transactions. Similarly, from a supply-side perspective, while the input data required to construct a global securities financing dataset may be relatively accessible (IHSM, for example, sources the necessary data directly from industry practitioners such as brokers, custodians, asset managers and funds), the competitor set is specific to this particular activity, suggesting a lack of obvious supply-side substitutability with other activities.

(C) The Commission’s assessment

(386) Based on the evidence available to it, the Commission considers global securities financing data to be a separate product market. Responding competitors confirmed the Notifying Party’s view, explaining that global securities financing data can be provided in conjunction with or separately from other offerings, while adding that there is some differentiation among the various products. The question of whether this gives rise to further segmentations of global securities financing data can be left open for the purposes of this case as no competition concerns arise regardless of the precise market definition.

6.2.22.3. Relevant geographical market

(A) The Commission precedents

(387) In Thomson/Reuters, the Commission considered the geographic scope of content sets to be at least EEA-wide and probably global.

(388) The Commission has not specifically considered the relevant geographic market for global securities financing data in any previous decision.

(B) The Notifying Party’s view

(389) The Parties consider that the supply of global securities financing data takes place on a global basis.

(C) The Commission’s assessment

(390) The market investigation has not presented any evidence to depart from the decisional practice regarding content sets or putting into question the Notifying Party’s view. However, the question of whether the markets are EEA-wide or global in scope can be left open as no competition concerns arise in either case.

6.2.22.4. Conclusion

(391) For the purposes of this Decision, the Commission considers that the relevant market is global securities financing data, which is at least EEA-wide in scope.

Notifying Party’s reply to RFI 24, Annex 3a; Form CO, Chapter on Vertical Relationships, paragraph 4.53.

Commission decision of 19 February 2008 in Case M.4726, Thomson/Reuters, paragraph 106.

Form CO, Chapter on Vertical Relationships, paragraph 4.70.

6.2.23. Portfolio valuation tools

(392) IHSM supplies portfolio valuation tools, which comprise post-trade valuations of a range of OTC derivatives and other financial products such as cash securities. These valuations are used by a range of customers, such as auditors, banks, corporates, fund administrators and custodians and fund managers to provide fair values for a range of liquid and illiquid securities.

6.2.23.1. Relevant product market

(A) The Commission precedents

(393) The Commission has not considered the relevant product market for portfolio valuation tools in any previous decision.

(B) The Notifying Party’s view

(394) The Parties consider that a market for the supply of portfolio valuations overall may be appropriate given the lack of substitutability between portfolio valuation and other valuation services. Moreover, the Parties do not consider a sub-segmentation of the portfolio valuation market to be relevant since customers will typically hold a range of different products and will therefore require a range of valuations across product types; and, from a supply-side perspective, all major suppliers of portfolio valuations will typically offer multi-product valuations.

(C) The Commission’s assessment

(395) Based on the market investigation and the evidence available to it, the Commission considers that portfolio valuation tools are a separate market. Competitors confirm that they sell such tools both separately and in conjunction with other products. The fact that there are commercial offers of both types of products from the same data vendor confirms that many end-customers prefer to purchase portfolio valuation tools on a standalone basis. There is evidence for a potential further segmentation by asset class, or the coverage of public assets vs private assets, as competitors confirm that such distinctions are important for customers when purchasing portfolio valuation tools. However, the question of further segmentation can be left open as no competition concerns arise regardless of the precise market definition.

6.2.23.2. Relevant geographical market

(A) The Commission precedents

(396) The Commission has not considered the relevant geographic market for portfolio valuation tools in any previous decision.

(B) The Notifying Party’s view

(397) In the Parties’ view, the market of portfolio valuation is global in scope.

(C) The Commission’s assessment

(398) The Commission has found no evidence putting into question the Notifying Party’s view; competitors confirm that portfolio valuation tools are offered, and prices are set, on a world-wide basis. In any case, the Commission does not need to conclude on the definition for the purposes of this case as no concerns arise regardless of the precise market definition.

6.2.23.3. Conclusion

(399) For the purposes of this Decision, the Commission considers that a relevant plausible global market exists for the supply of portfolio valuation tools, potentially further segmented by asset class, or the coverage of public assets vs private assets.

6.3. Competitive assessment

6.3.1. Affected markets with respect to financial data and infrastructure products

(400) The Transaction leads to the following horizontally affected markets in the area of financial data and infrastructure products:

a) Loan identifiers

b) Leveraged loan indices

c) Equity indices

d) Natural resources sector equity indices

e) US Corporate Bond indices

f) Index calculation and administration services

(401) In addition, the Transaction leads to vertically affected markets with respect to the markets listed in Table 1.

(402) The Commission also assesses potential conglomerate effects with respect to the combination of (i) the Parties’ indices, (ii) IHSM’s issuance platforms and S&P’s desktops, (iii) S&P’s credit ratings and IHSM’s loan administration solutions, and (iv) S&P’s credit ratings and IHSM’s issuance platforms.

Notifying Party’s response to RFI 24, Annex 3a, paragraph 4.76.

Replies to question 22 of Questionnaire 8.

Table 1: Vertically affected markets in relation to financial data and infrastructure products

Upstream market(s) (Party active) Downstream market(s) (Party active)

Loan identifiers (both) Leveraged loan market intelligence (S&P)

Fundamentals data (S&P)

Leveraged loan market intelligence (S&P)

Leveraged loan indices (both) Loan pricing and reference data (IHSM) Company credit risk analytics (S&P)

Credit ratings (S&P)

CDS pricing data (IHSM) Company credit risk analytics (S&P)

Municipal bond pricing and reference data Municipal bond indices (S&P) (IHSM)

RED Codes (IHSM) Cross-reference tools (S&P; IHSM negligible)

Fixed income indices (IHSM; S&P negligible)

CDS indices (IHSM)

Pricing and reference data (IHSM)

Equities and regulatory reporting (IHSM)

CUSIP data (S&P)

Issuer solutions (IHSM)

Investor administration services (IHSM)

Institutional holdings/investor data (IHSM; S&P negligible)

Managed corporate actions (IHSM)

In addition, a number of IHSM products offer access to CUSIP data using a similar bring your own data model. For example, IHSM’s Markit digital is a Software as a Service (SaaS) that offers solutions for the financial services industry to aggregate, normalise, enhance and display financial information via web services, websites and native mobile apps. Subject to having a CUSIP license, the user can, for example, enter a CUSIP to find a specific instrument, and when the solution returns the product in the display, this will include the CUSIP and potentially other identifiers. Other IHSM products that use a bring your own data model with respect to CUSIPs include: Ipreo (issuance platforms), Iplanner (investor event management solutions). These are not discussed further in this Decision.

78

Upstream market(s) (Party active) Downstream market(s) (Party active)

Fixed income indices (both Parties)

CDS indices (IHSM)

Pricing and reference data (IHSM)

Economic data (IHSM)

Credit ratings data (S&P)

Issuer solutions (IHSM)

Investor administration services (IHSM)

Institutional holdings/investor data (IHSM)

Investor event management solutions

Portfolio valuation services (IHSM)

Equity Indices (S&P; IHSM negligible) Global securities financing data (IHSM)

Economic data (IHSM)

Fixed income indices (IHSM; S&P negligible)

Indices (both) Desktop services (S&P)

Non-real time datafeeds (S&P)

Index calculation / administration (both) Indices (both)

Managed corporate actions (IHSM) Equity indices (S&P; IHSM negligible)

Economic data (IHSM) Credit ratings (S&P)

Economic data (IHSM)

Sector classification schemes (S&P)

Stock selection and strategy services (both Parties)

Similarly to credit ratings, a number of IHSM products allow users to incorporate credit ratings data into such products. For example, Wall Street Office (“WSO”) is IHSM’s loan administration workflow tool for managers of loan portfolios, which reduces the burden of the manual processes that arise in respect of loan portfolio administration throughout the loan lifecycle. Subject to having a relevant license, WSO users can populate the WSO tool with credit ratings of the preferred supplier (“bring your own data” model). Given that it remains the user’s responsibility to choose its data supplier and to obtain the relevant license, the Commission does not consider such links as giving rise to vertical relationships. Other IHSM products that use a “bring your own data” model with respect to credit ratings data include Ipreo (issuance platforms). These are therefore not discussed further in this Decision.

79

(C) The Commission’s assessment

Market power

(411) First, while the Commission has no market share information with respect to the plausible global loan identifier market, information provided by the Notifying Party regarding the coverage of their identifiers in the global loan market and the market investigation provided some information with respect to the Parties’ position in that plausible market.

(412) The global loan market is characterized by limited transparency and public reporting, given that loan transactions are private. Hence, there is no public data source enabling a calculation of market shares. Given IHSM’s market position in the loan pricing and reference data market, and IHSM’s estimate that their loan pricing and reference data covers ca. […] of the global loan market (consisting of ca. […] loans in total), at least a ranking of loan identifiers can be approximated from the coverage data in Table 2 and Table 3. Based on the coverage data, LXIDs are the clear market leader in a plausible global market for loan identifiers, followed by LCD IDs and loan CUSIPs which have more or less the same coverage, but both in any case have only around 1/3 of the coverage of LXIDs. ISIN and FIGI coverage is slightly lower than LCD IDs’ and loan CUSIPs’, followed by LINs.

(413) The market investigation confirms this ranking broadly. For example, one competitor states: “LoanX identifiers are “must have” resources for CLO trustees and other customers.(…)” [emphasis added]

(414) Another competitor states: “(…) LXIDs are the market standard for identifiers (…).”[emphasis added] A very limited number of respondents to the market investigation mention other identifiers, such as Bloomberg IDs, FIGIs, loan CUSIPs, LCD IDs and LINs, and practically always in the context of those identifiers providing a lower coverage. For example, one customer mentions: “LXIDs are the de-facto market standard identifier for loans. It is the type of the identifier that other providers are most likely to be able to supply together with their own data. A runner up (at least in the European space) is Bloomberg with their Bloomberg IDs (both the proprietary ones but also the FIGIs).”[emphasis added]

(415) As already mentioned in paragraph (119), the large majority of responding customers state that they need to receive loan pricing data with a particular loan identifier. In the follow-up question, the identifier they require for this purpose is overwhelmingly “LXIDs”. A clear majority says the same with respect to loan reference data.

(416) A competitor confirms this: “[The competitors’ market intelligence product] has unique [proprietary identifier of the competitor] to support our internal data mapping. We do provide them to clients and they are part of our data sets – but we also have to provide market standard identifiers (LXIDs, CUSIPs) to support broader market data mapping. We have to meet the market requirement – by providing what our clients use – we can’t be competitive by relying upon our own identifiers.” [emphasis added]

(417) Customers answer similarly with respect to downstream products in which loan identifiers typically feature, such as leveraged loan market intelligence. A majority of customers state that it is important for them that leveraged loan market intelligence includes LXIDs and CUSIPs.

(418) While the above statements and the coverage figures of the different loan identifiers allow no conclusion on exact market shares, the market investigation results support the conclusion that the Transaction effectively combines the number one and likely dominant provider with the second most important loan identifier provider. The other loan identifiers (FIGIs, LINs, LCD IDs) do not seem to pose any meaningful constraints currently and this is unlikely to change quickly given the characteristics of the market (see further down with respect to network effects, barriers to entry and switching).

Closeness of competition

(419) Second, the merging firms appear to be close competitors, as evidenced by responses with respect to both substitutability and closeness of the two identifiers. As set out in paragraph (141), the majority of informative competitor responses say that LXIDs and loan CUSIPs are substitutes for specific use cases or for nearly all use cases. One competitor explains: “they [LXIDs and loan CUSIPs] are substitutes for the instruments for which there are CUSIPs. (…)” From a demand-side perspective, the majority of customers respond that LXIDs and loan CUSIPs are substitutes for specific use cases.

(420) To the question “Would you consider that any other loan identifier offers a good alternative to LXIDs?” more than half of the informative competitors’ responses and more than half of the informative customers’ responses was “Yes, but no alternative as close as loan CUSIPs”. A competitor states: “The only challengers to LoanX identifiers today are CUSIPs, offered by S&P. However, they are currently only nascent competitors and not yet a scale rival.”

(421) The Commission notes that while none of the other loan identifiers seems to have a coverage that can rival IHSM’s, it is also evident that the standalone loan CUSIPs of S&P have comparatively the highest coverage after LXIDs. The conclusion from those answers is that LXIDs and loan CUSIPs compete and the Parties’ own statistics show that loan CUSIPs are the closest rival to LXIDs in the global loan identifier market, given that coverage seems to be the most important characteristic for customers of loan identifiers. LXIDs and loan CUSIPs are clearly the two loan identifiers with the widest coverage in the market.

Network effects

(422) Third, answers to the market investigation suggest that the loan identifier market is characterized by network effects. Even if these do not seem to be as strong as in other identifier markets (see paragraph (423) below), it is still evident from the responses that both customers and competitors deem the use of specific loan identifiers (and the Parties’ loan identifiers are consistently among the first mentioned) important, because many other market participants use them.

(423) In securities markets with regulation on public reporting, International Securities Identification Numbers (ISINs) have been made obligatory to use in order to comply with reporting requirements. This has created strong incentives for market participants to use ISINs for other purposes as well. A similar obligation does not exist in the loan markets. One competitor states: “The IHSM’s LoanX IDs are widely used across the market but there is no single standard identifier system for leveraged loans. [Competitor] has its own identifier system called [identifier name] which [competitor] uses for its own purposes and therefore it can function without the LoanX IDs. Nevertheless, our customers have to do some extra work (because [competitor] does not uses LoanX IDs) (…)” Another competitor puts it like this: “[The competitor’s product] has unique [proprietary identifiers] to support our internal data mapping. We do provide them to clients and they are part of our data sets – but we also have to provide market standard identifiers (LXIDs, CUSIPs) to support broader market data mapping. We have to meet the market requirement – by providing what our clients use – we can’t be competitive by relying upon our own identifiers.”

(424) These quotes illustrate that while there may be no formal legal requirements or other external reason forcing market participants to use a specific loan identifier, certain loan identifiers, such as LXIDs and to a more limited extent loan CUSIPs, have developed into the default loan identifier option, based on which the market

313Submission of a competitor dated 12 July 2021, paragraphs 11 and 12.

314“Network effects are a special type of externality in which consumers’ utility and/or firms’ profits are directly affected by the number of consumers and/or producers using the same (or a compatible) technology.” Shy, Oz (2011), A short survey of network economics, Review of Industrial Organization (2011) 38:119–149.

315E.g. reply to question 41.1 of Questionnaire 6, supplemented by separate email on 29 September 2021; reply to question 3.2.1 of Questionnaire 6.

316Reply to question 41.1 of Questionnaire 6, supplemented by separate email on 29 September 2021.

317Reply to question 3.2.1 of Questionnaire 6.

84

communicates. This is likely also related to both identifiers being active already since the early 2000s (the first contract in relation to loan CUSIPs between S&P and the LSTA dates from 2002, and IHSM acquired LoanX in 2004). As a result of having been active for a long time, both have developed a certain coverage of the global loan markets. In the case of IHSM’s LXIDs, this is probably also a result of a series of acquisitions of loan-related assets as well as organically grown products (LIBOR replacement data for loans, loan reference data, Market Entity Identifiers), which have led to LXIDs becoming the standard identifier for customers using IHSM’s loan-related products. Practically all of those products seem to have market leading positions in their respective markets.

(425) From the responses to the market investigation, many customers and competitors point to the aspect of “coverage” as a distinguishing feature and highlight that LXIDs and to a more limited extent loan CUSIPs, are deeply entrenched identifiers in the loan market. For example, “Our understanding is that loan CUSIPs have more limited loan coverage than LXIDs.” A competitor explains: “(…) they [LXIDs and loan CUSIPs] are substitutes for the instruments for which there are CUSIPs. But because LXIDs cover a significantly larger share of the market — particularly in the less transparent part of the loan market — they are more commonly used.” Another competitor explains that in order to offer a competitive loan-related product “(…) the existing identifiers (primarily LXIDs and CUSIPs) are already embedded in the market infrastructure. Accordingly, you would need to license the identifiers which are already used in the market (LXIDs) (…)”[emphasis added]

(426) A potential customer explains: “In practice, this results in a manual search for the relevant loan based on reference data provided and allocate a FIGI to that loan, which is a labour-intensive exercise(…) CUSIPs are sometimes available to customers for USD loans depending on the portfolio data source, but CUSIP does not cover all USD loans, so some percentage of any given portfolio must still be manually mapped. For completeness, there is no third party service available for mapping LXIDs or other identifiers. [The potential customer] receives frequent requests from its customers to be able to offer the LXIDs.”[emphasis added]

318Although LINs have possibly been around since the early 1990s, but Refinitiv’s LPC loan pricing data services were launched in 1999, according to the Notifying Party.

319Acquisitions of i) Vichara Technologies Inc (2003), CLO pricing data; ii) LoanX (2004), a global electronic trading platform for syndicated loans including loan pricing data and LXIDs; iii) International Index Company (2007), iBoxx (US Leveraged loan index); iv) JP Morgan FCS Corp including Wall Street Office (WSO) and Notice Manager (2008), traded loan software/services; v) ClearPar Distressed Loan Trade Settlement/Loan Settlement Custodian Services (2010); vi) DTCC Loan/SERV LLC (2016), now Markit Loan Reconciliation and vii) Ipreo including Debtdomain (2018)), see Notifying Party’s response to RFI 35, question 4.

320Notifying Party’s response to RFI 35, question 4.

321Loan Trade Settlement (Clearpar), Loan Reconciliation, Notice Manager, Custodian Services (Docs & Messaging), Debtdomain, WSO Software, WSO Services, Trade Closing Services. The only loan-related product which does not have a “#1” market position, but only “#3” is Agent Services.

322IHSM internal document No. ASH000200, slide 8.

323Reply to question 21.1 of Questionnaire 6.

324Reply to question 21.1 of Questionnaire 6.

325Reply to question 10.1 of Questionnaire 6.

326Reply to question 23.1 of Questionnaire 6, supplemented by separate email on 28 September 2021.

85

(427) A potential customer explains that they have to engage in labour-intensive and costly manual mapping of alternative identifiers, since they do not have access to LXIDs: “In practice, this results in a manual search for the relevant loan based on reference data provided and allocate a FIGI to that loan, which is a labour-intensive exercise(…)” . Another competitor also highlighted that customers have asked for a mapping to LXIDs, but the competitor has not been able to provide this, as the price quoted by IHSM for that input was considered too high by the competitor.

(428) The quotes above also indicate that the current status of LXIDs is not one that can be easily and quickly challenged, based on the competitive dynamics in the loan data market, which shows signs of network effects. As a result, competition in the loan identifier markets is not particularly strong, but to the extent competition exists, it would seem to be taking place mainly between LXIDs and loan CUSIPs and to a weaker extent between LXIDs and other loan identifiers.

Switching

(429) Fourth, customers likely have limited possibilities of switching loan identifier supplier, because switching loan pricing data provider or loan reference data provider (which includes loan identifiers) is very resource intensive. When asked, why switching loan reference data provider was not easy, one respondent commented: “Main reference point/required identifier” Another respondent that is currently switching loan pricing data provider explains: “We are in the middle of switching loan pricing data provider from another provider to IHSM. It will take us 18 months for the whole process.”

(430) Switching loan identifier is not an option at all for many competitors in loan reference data or leveraged loan market intelligence, for example, because they depend on being able to communicate with their clients using market-standard identifiers (LXIDs or loan CUSIPs). In addition, while loan identifiers have a separate value, they derive their competitive strength from their coverage and the ability for customers to use and interpret loan data and related analytics, news, etc.

Countervailing buyer power

(431) Fifth, customers and competitors do not seem to have any significant countervailing buyer power. The merged entity’s position in the downstream markets for loan data, market intelligence and infrastructure products is not constrained by the bargaining strength of its customers. No customer represents such an important percentage of demand that it enjoys substantial countervailing buyer power. Countervailing buyer power of customers could also derive from customers who contribute data to IHSM’s pricing and reference data service, such as banks. While

327Reply to question 23.1 of Questionnaire 6, supplemented by separate email on 28 September 2021.

328Reply to question 41.1 of Questionnaire 6, supplemented by separate email on 29 September 2021.

(432) Equally, a large majority of competitors (who are in most cases also customers of the merged entity, including in relation to loan pricing and reference data and loan identifiers) in the broader pricing and reference data and market intelligence space do not consider that they provide any essential inputs to the merged entity that would award them any negotiating power.

Barriers to entry

(433) Sixth, barriers to entry seem to be very high with respect to offering loan identifiers. This is because a crucial factor for a successful loan identifier offering seems to be coverage, and coverage of the global loan market has historically only been achievable by either providing related data/products, such as loan pricing or reference data (LXIDs), or by building on the brand, reputation and infrastructure of an existing identifier business (loan CUSIPs). FIGIs were developed more recently (2009) and outside a significant identifier business, which shows in their comparatively low coverage of the loan market. Other loan identifiers like LCD IDs (only available in S&P’s LCD product) or Refinitiv’s LINs (only provided as part of Refinitiv’s loan pricing and reference data) are in the market more in a function as potential competitors than actual competitors to LXIDs and loan CUSIPs, it seems. This highlights that even existing providers of loan identifiers face barriers to compete effectively due to the current market environment, where a market-leading loan identifier has developed and switching costs are high. It would likely require a significant investment to try and displace current loan identifiers or even gain some market share for a new entrant.

(434) Taking the loan pricing and reference data market as a proxy, respondents to the market investigation unanimously expressed that entering either those markets (which include loan identifiers) is very challenging and requires significant resources and time. One competitor explains how much it would cost a market participant that is already active in fixed income pricing data, to enter the loan pricing data space: “[Competitor] believes it would take at least a year for a vendor that does supply bond and CDS prices to build the tools, establish the dealer relationships, create the databases, scale the operation, build the API’s, and hire/source the collection and integration staff. First year estimate costs likely would be around 2.5 mm (based on previous experience). But even at that point, the product would not be fully adopted by the market until 4-5 years of pricing history is available – which would have to be collected and stored retrospectively, if even possible without purchasing from [the main competitors] at exorbitant costs. Factoring that in, I would estimate the total go-to-market cost at around 10-12mm (over 4-5 years). It is also worth noting that even once the product gets to the marketplace, the new offering would still need to concord to either LXIDs or CUSIP IDs (both controlled by the combined entity) and the cost of obtaining those IDs likely will increase significantly for a competitive product.”[emphasis added] Another competitor states: “It is our understanding that it would be difficult to begin producing loan pricing data in a short time.(…).”

87

(436) A case demonstrating how difficult it is to successfully enter the loan identifier (and other identifier markets as well, given that FIGI can be used for many different asset classes) is FIGI. Launched by Bloomberg in 2009 (back then under the name Bloomberg Global Identifier), who had significant presence in loan reference data already, FIGIs have been comparatively unsuccessful in gaining market share in the loan identifier market. In conclusion, the Commission notes that entry in the loan identifier market seems to be difficult, and in fact even suppliers that have been active in the market for several years or in some cases decades have not managed to come close in terms of coverage to LXIDs.

Notifying Party’s arguments with respect to difference between LXIDs and loan CUSIPs

(437) With respect to the Parties argument, that their identifiers are different products in terms of their features, use cases and customer bases, and hence do not compete or constrain each other, the Commission notes the following:

(438) In relation to the difference in product features, the Notifying Party emphasizes that LXIDs change over time, when properties of a loan change (e.g. refinancing) while loan CUSIPs are static, i.e. a loan keeps the same loan CUSIP irrespective of any changes in the properties of the loan. Apart from one respondent mentioning this fact (“(…) Restatement/Amendment/Extension event rules differ across all ID providers.”), none of the responses to the market investigation suggested that this difference in the product features has any impact on loan identifiers’ substitutability or whether they compete. The Commission also notes that even though LXIDs are updated and therefore several LXIDs may exist for the same original loan, IHSM provides customers with “recommended update files” that contain the whole history

337Reply to question 7.1 of Questionnaire 7.

338Reply to question 7.1 of Questionnaire 7.

339Reply to question 10.1 of Questionnaire 7.

88

of LXIDs and hence allow linking the different LXIDs to the same loan. To the extent that other identifiers do not track changes in the loan properties, they could indeed be considered as having different product features. However, at least loan CUSIPs are in effect delivered including changes in the loan properties as well. The Notifying Party explains: “As regards loan CUSIPs, these are accompanied by an underlying data file, published daily, containing principal attributes of syndicated loans with associated CUSIPs (…). Where the features contained in this database of summary loan attributes change (for example where the maturity date or facility type changes) then the relevant feature would be updated in the subsequent iteration of the daily data file (although there is not, unlike for LXIDs, any change history feature). Thus users of loan CUSIPs do not have the same visibility as customers of IHSM’s loan reference data (including LXIDs) into changes to a given syndicated loan.”

The conclusion does not logically follow from the explanation of the way in which loan CUSIPs are delivered. In fact, loan CUSIPs seem to be updated in terms of content delivered alongside loan CUSIPs much in the same way as LXIDs. The only difference would seem to be that IHSM marks changes in the loan properties by issuing a new LXID (that is however relatable to previous LXIDs), as well as providing information about the feature that changed. Compared to that, loan CUSIPs stay the same but the loan properties information is updated just like with LXIDs on a daily basis. Any user of loan CUSIPs can therefore easily store those changes in their database, if they require this information, e.g. to create a times series of historical pricing or analytics. The only data field that seems to be different between key terms submitted with loan CUSIPs compared to LXIDs, is that LXIDs contain the relevant industry of the issuer, e.g. “healthcare” or “real estate”). This is clearly a data field that is unlikely to change frequently.

(439) As a result, it would seem that market participants effectively receive the same information with respect to changes in the loan properties with both identifiers. In any case, respondents to the market investigation did not raise this “discrepancy” between LXIDs and loan CUSIPs. The Commission therefore considers that this is not a meaningful difference in terms of product features between LXIDs and loan CUSIPs.

(440) In terms of commercialization, the Notifying Party claims that LXIDs and loan CUSIPs are distinct, because loan CUSIPs are commercialized standalone while LXIDs are not. The Commission has already analysed this question in Section 6.2.4.2, concluding that even though LXIDs may not be explicitly commercialized standalone, they do generate revenues for display rights which are much higher than “simple” access to IHSM’s loan pricing and reference data (which includes LXIDs). Hence, a certain standalone commercial value is attributable to LXIDs. Apart from that, several market participants suggest that the bundling of LXIDs with loan pricing and reference data is a commercial choice of IHSM. The market investigation in any case showed standalone demand for LXIDs which is separate from the demand for an identifier to interrogate IHSM’s loan pricing and reference data.

342Reply to question 21.1 of Questionnaire 6, paragraph 21.2.

343Reply to question 21.1 of Questionnaire 6.

89

(441) As additional evidence for the standalone value of the LXID brand, […] , […]”. The Notifying Party explains that the trademark symbol is typically used to denote unregistered trademark rights based on usage instead of formal registration. The Commission therefore considers that even though the genesis of LXIDs and loan CUSIPs was different and as a result, loan CUSIPs are licensed separately while LXIDs are currently bundled with loan pricing and reference data, this is not sufficient to conclude that they do not compete or constrain each other.

(442) In terms of different customer profiles, the Notifying Party claims that the two identifiers have different customers. While LXIDs are primarily used by IHSM’s customers of loan pricing and reference data, which are asset managers, asset servicers and banks, loan CUSIPs are used by banks, asset managers and information service providers. The Commission notes, that two of three customer categories are the same. The third category that does not overlap are information services providers, which do not license IHSM’s loan pricing and reference data, but only loan CUSIPs.

(443) This is not surprising, given that information services providers are among the main competitors of IHSM in loan pricing and reference data and hence do not require access to this data. Nevertheless, some of those competitors have in the past requested access to LXIDs and were quoted prices that are described by one as amounting to a “(…) constructive refusal to supply (…).” This emphasizes once more IHSM’s market power in loan identifiers.

Countervailing buyer power

(431) Fifth, customers and competitors do not seem to have any significant countervailing buyer power. The merged entity’s position in the downstream markets for loan data, market intelligence and infrastructure products is not constrained by the bargaining strength of its customers. No customer represents such an important percentage of demand that it enjoys substantial countervailing buyer power. Countervailing buyer power of customers could also derive from customers who contribute data to IHSM’s pricing and reference data service, such as banks. While

327Reply to question 23.1 of Questionnaire 6, supplemented by separate email on 28 September 2021.

328Reply to question 41.1 of Questionnaire 6, supplemented by separate email on 29 September 2021.

(432) Equally, a large majority of competitors (who are in most cases also customers of the merged entity, including in relation to loan pricing and reference data and loan identifiers) in the broader pricing and reference data and market intelligence space do not consider that they provide any essential inputs to the merged entity that would award them any negotiating power.

Barriers to entry

(433) Sixth, barriers to entry seem to be very high with respect to offering loan identifiers. This is because a crucial factor for a successful loan identifier offering seems to be coverage, and coverage of the global loan market has historically only been achievable by either providing related data/products, such as loan pricing or reference data (LXIDs), or by building on the brand, reputation and infrastructure of an existing identifier business (loan CUSIPs). FIGIs were developed more recently (2009) and outside a significant identifier business, which shows in their comparatively low coverage of the loan market. Other loan identifiers like LCD IDs (only available in S&P’s LCD product) or Refinitiv’s LINs (only provided as part of Refinitiv’s loan pricing and reference data) are in the market more in a function as potential competitors than actual competitors to LXIDs and loan CUSIPs, it seems. This highlights that even existing providers of loan identifiers face barriers to compete effectively due to the current market environment, where a market-leading loan identifier has developed and switching costs are high. It would likely require a significant investment to try and displace current loan identifiers or even gain some market share for a new entrant.

(434) Taking the loan pricing and reference data market as a proxy, respondents to the market investigation unanimously expressed that entering either those markets (which include loan identifiers) is very challenging and requires significant resources and time. One competitor explains how much it would cost a market participant that is already active in fixed income pricing data, to enter the loan pricing data space: “[Competitor] believes it would take at least a year for a vendor that does supply bond and CDS prices to build the tools, establish the dealer relationships, create the databases, scale the operation, build the API’s, and hire/source the collection and integration staff. First year estimate costs likely would be around 2.5 mm (based on previous experience). But even at that point, the product would not be fully adopted by the market until 4-5 years of pricing history is available – which would have to be collected and stored retrospectively, if even possible without purchasing from [the main competitors] at exorbitant costs. Factoring that in, I would estimate the total go-to-market cost at around 10-12mm (over 4-5 years). It is also worth noting that even once the product gets to the marketplace, the new offering would still need to concord to either LXIDs or CUSIP IDs (both controlled by the combined entity) and the cost of obtaining those IDs likely will increase significantly for a competitive product.”[emphasis added] Another competitor states: “It is our understanding that it would be difficult to begin producing loan pricing data in a short time.(…).”

87

(451) In this example, loan CUSIPs seem to be in higher demand than LXIDs. This is because loan CUSIPs have a higher coverage in the US loan market, while LXIDs have a higher coverage in the European loan market (while also having coverage of the US market). For example, while [...] of dollar-denominated loans that have an LXID also have a CUSIP, that percentage is just [...] for euro-denominated loans.

(452) While this suggests a certain complementarity between LXIDs and loan CUSIPs in terms of geographies covered, the market for loan identifiers is clearly considered global and customers do not distinguish providers by location. The different geographic focus of both companies is likely due to their historic roots (IHS Markit being headquartered in the UK and S&P in the US).

Notifying Party’s arguments with respect to ability and incentives of the merged entity

(453) The Notifying Party further argues that the merged entity would have no ability or incentive to worsen LXID provision post-merger to benefit loan CUSIPs, or vice versa. With respect to worsening LXID provision the Notifying Party claims that worsening LXID supply would undermine its core loan pricing and reference data product and result in lost sales.

(454) The Commission does not consider it likely that the merged entity would lose sales of loan pricing and reference data if it increased prices for LXIDs, for example. This is mainly, because IHSM is a market leading provider with significant market power in both loan pricing and reference data and loan identifiers. All those markets are characterized by high switching costs and high barriers to entry. The combination with the second most important loan identifier would likely strengthen the merged entity’s position not only in the loan identifier market, but also in the related loan pricing and reference data markets. This is because switching loan pricing and reference data provider is not only costly and takes a long time, but also, switching customers would then have to find a way to replace LXIDs. As some quotes above suggest, this is not impossible, but it is possibly resource-intensive (see quotes of competitors in paragraphs (434) and (435) above). In fact, all respondents to the market investigation that have switched loan pricing and reference data provider, switched to IHSM, not away.

(455) The Commission also disagrees that the most likely result of increasing prices for LXIDs (or IHSM’s loan pricing and reference data generally) would be an increased adoption of Refinitiv’s loan pricing data and associated LINs. It is evident from the Parties’ own submission that Refinitiv’s loan pricing data covers a much lower share of the global loan market than IHSM’s loan pricing data. In addition, based on the

responses to the market investigation, LINs do not seem to be widely adopted or used outside Refinitiv’s data (contrary to loan CUSIPs). Finally, LINs have a much lower coverage than both IHSM and loan CUSIPs.

(456) A similar reasoning applies with respect to the Notifying Party’s argument that if the merged entity were to increase prices for LXIDs (if they were sold separately), customers would be more likely to switch to FIGI instead of loan CUSIPs. Given FIGIs lower acceptance in the loan identifier market, this is not credible.

(457) Apart from that, the Notifying Party argues that the merged entity would have no incentives to increase prices for LXIDs (or loan pricing and reference data more generally), [Details of contract with partner]. However, in a market where the [Details of contract with partner], the argument on the lack of incentives does not seem very credible.

(458) The Notifying Party further argues that the merged entity would have no incentive to increase prices or otherwise worsen the provision of loan CUSIPs to benefit LXIDs, because LXIDs are not a substitute for loan CUSIPs, [summary of confidential contractual arrangements concerning CUSIP’s relationship to the rest of the S&P business]. The Commission considers none of those arguments convincing based on the results of the market investigation and other information available to it.

(459) First, the results of the market investigation have shown substitutability between LXIDs and loan CUSIPs from a customer perspective. In addition, the Commission outlined above, why product features, use cases and customers of the two loan identifiers are ultimately not materially different.

(460) Second, based on the current contract between the ABA and S&P on CUSIPs (which covers loan CUSIPs), the ABA’s and S&P’s interests with respect to increasing prices for loan CUSIPs are aligned [details of contract with partner]. It is therefore plausible that the ABA would not object to a price increase of loan CUSIPs provided it did not breach other clauses of the agreement.

(461) Third, while the CUSIP business may be [details of contract with partner], [details of contract with partner]. In terms of actual practical ring-fencing arrangements, the Commission notes that there is [details of contract with partner].

(462) The Commission considers it likely that the Transaction will provide the merged entity with the ability and incentive to raise prices for one or both loan identifiers, given the market power that the combination of those two identifiers affords the merged entity. The market investigation supports the conclusions that a) LXIDs already have a very strong market position in the loan identifier market, b) the loan identifier market is characterized by network effects, high switching costs and high barriers to entry, leading to the competitive constraints on the market leader LXID to be limited already pre-Transaction, and c) loan CUSIPs seem to be the closest competitor to LXIDs. One competitor confirms with respect to the impact of the

Transaction on the loan identifier market: “(…) This would make the combined entity a dominant provider.”

(463) A competitor sees the following impact as a result of combining the two loan identifiers: “In addition, the merged entity will control both the primary and secondary identifier in the loan segment, the LXID and the CUSIP respectively (…). The merged entity will be in a unique position to offer customers a “one-stop-shop” for loan related products and services and will be incentivized to further entrench its position in the form of price based and non-price based conduct,(…) or by tying or bundling distinct loan products with the aim of foreclosing rivals. As a result, this will reduce competitors’ incentives to invest in related loans businesses and will further reduce innovation in the area, thereby restricting customer choice.”

(464) A customer summarizes the impact of the Transaction on the loan reference data market like this: “reduction in competition, less choice, higher prices.” Another customer states: “(…) the price is most likely to increase and our bargaining power should decrease.” One customer also expects a “negative impact [but only] if this enables Markit/S&P or other providers to increase prices due to lower competition in the space.”

(465) Overall, responses of competitors with respect to the impact of the Transaction on loan pricing data and loan identifier markets are unanimously negative (not counting uninformative responses). The majority of responses from competitors is also negative with respect to the impact of the Transaction on loan reference data and loan identifiers markets. Almost all of the companies categorized as competitors here are simultaneously also customers or potential customers of the Parties in respect of loan identifiers.

(D) Conclusion

(466) Based on the market investigation and other evidence available to the Commission, the Commission considers that the Transaction raises serious doubts as to its compatibility with the internal market and the functioning of the EEA Agreement with respect to the overlap between the Parties’ activities in the plausible global market for loan identifiers by creating or strengthening a dominant position.

6.3.2.2. Indices

(467) Within financial indices, affected markets arise in relation to fixed income indices (but only for the segment of (i) leveraged loan indices and (ii) US corporate bonds),

(A) Leveraged loan indices (horizontal overlap)

(468) The Commission assesses below whether the Transaction raises serious doubts with respect to its compatibility with the internal market by eliminating important competitive constraints in the plausible global market of leveraged loan indices, leading to the creation or strengthening of the merged entity’s dominant position.

(A.i) The Parties’ activities

(469) The Parties are both active in leveraged loan indices. S&P provides the European Leveraged Loan Index (ELLI) as part of its leveraged loan market intelligence product LCD and not standalone. The loan pricing data for the ELLI is sourced from IHSM. S&P also provides US leveraged loan indices (US LLI) which it licenses to customers for use in the construction of financial instruments or in the form of market data, to be used as a benchmark for active funds. The loan pricing data for the US LLI is sourced from [Name of customer].

Table 5: Market shares in the global market for leveraged loan indices

Index Provider 2020

AUM €m %

S&P Global [confidential] [90-100]%

IHS Markit [confidential] [0-5]%

Combined [confidential] [90-100]%

Credit Suisse [confidential] [0-5]%

Total market [confidential] 100,0%

Source: Form CO, Chapter on Vertical Relationships, Annex D.6

(A.ii) The Notifying Party’s view

(470) The Notifying Party considers that the category in the Morningstar database (“Level 3”), based on which the Transaction will lead to a merger to monopoly is too narrow to represent a plausible market. According to the Notifying Party, there is supply-side substitutability among so-called floating rate debt indices and leveraged loan indices. The same competitors provide both indices, including Bloomberg, ICE, JP Morgan and Credit Suisse.

(471) Furthermore, on the demand-side, the Notifying Party deems that leveraged loan indices and other floating rate debt indices are substitutable for customers, though acknowledges that substitutability depends on the use case. The Notifying Party also suggest that price and quality are important factors which would lead customers to look for the many available alternatives if they were not satisfied.

(472) Even if the leveraged loan index market was considered to be a relevant market, the Notifying Party submit that each S&P and IHSM have only [Number of customers], which cannot represent a market of their own. Moreover, the Morningstar database does not capture OTC products (which is how leveraged loans are primarily traded), so a focus on funds only underrepresents the relevant competitor set, in the Notifying Party’s view.

(473) In addition, barriers to entry are low, as relevant data is readily available from many suppliers and public sources. The Notifying Party therefore considers it easy for other fixed income index providers to provide leveraged loan indices.

(474) The Notifying Party claims that funds account for [0-5]% or less of the investible leveraged loan market, the majority of which is made up of bank holdings and institutional separate accounts, who do not require indices for the creation of funds, but only indices for the purposes of benchmarking. When considering the rest of the market Credit Suisse’s and JP Morgan’s position is much stronger, with [80-90]% and [20-30]% respectively, according to the Notifying Party.

(A.iii) The Commission’s assessment

(475) The Commission considers that none of the Notifying Party’s claims are supported by the results of the market investigation or additional submissions by the Parties.

(476) First, several respondents to the market investigation considered “Morningstar Level 3” the relevant level at which to choose indices, and no respondent replied “Level 1 or “Level 2”. Several respondents also answered “It depends”. One customer explains: “Each level is relevant depending on the product category (e.g. for corporate High Yield the Level 3 is more relevant while for global products Level 1 is equally relevant).” Leveraged loans would tend to rather fall in the category “corporate high yield”, i.e. according to this customers view, level 3 would be the relevant category for leveraged loan indices.

(477) A market participant with specific knowledge of the leveraged loan market is of the view that “the Transaction combines the two predominant suppliers of loan indices,

namely the US S&P LSTA leveraged loan index and the IHSM iboxx US loan index. The main reason why there are no relevant competitors in the market of providing leveraged loan indices for the creation of funds is that the Parties own the only IOSCO compliant loan indices globally. In the market participant’s view, being IOSCO compliant is an important criterion for an index fund/ETF. This is also evident from the relevant EU-Regulation, which effectively implements the IOSCO principles and requires compliance with IOSCO principles as a criterion to enable endorsement of benchmarks (i.e. indices used for the creation of financial instruments) provided from an administrator located in a third country.

(478) According to this market participant, Credit Suisse does not compete directly with the Parties loan indices, even though it is the longest established loan index, for that very reason, i.e. that it is not IOSCO compliant. In addition to the IOSCO compliance, listed funds require specific high quality loan pricing data, so-called mark-to-market data. This data is based on actual trades of loans, whereas evaluation pricing is based on models. The market participants stressed that “for mark-to-market loan pricing data, there are only two main suppliers: IHSM and Refinitiv.”

(479) When asking competitors who are the top providers of leveraged loan indices, the majority answer “S&P” without mentioning others. Some respondents also mention IHSM and only one also mentions Credit Suisse. This contradicts the Notifying Party’s view that the other providers of leveraged loan indices (Credit Suisse and JP Morgan) compete closely with S&P and IHSM.

(480) According to the Notifying Party, leveraged loan index providers compete across different products for which those indices are used, i.e. funds/ETFs, OTC products (such as total return swaps), bank holdings and institutional separate accounts. The Commission notes that IHSM and S&P both do not commercialize their leveraged loan indices to providers using them in OTC products (which accounts according to the Parties for [90-100]% of the leveraged loan index market). The Notifying party claims this is because investment banks are much better placed to develop indices for use in OTC products (as they structure, issue and trade those products), than the Parties with their focus on more liquid assets.

(481) The Notifying Party argues that indices used for the creation of funds compete with indices used for the benchmarking of active investments, and states “As regards the inputs required for selling OTC products as opposed to selling ETFs or mutual funds, in each based on leveraged loan indices, as far as the Parties are aware the same or very similar inputs are likely to be required.” Based on the views of the main industry association in this area, it is however not the case, that “the same or very similar inputs” are used for funds/ETFs and OTC products. The Commission therefore considers that the competitive constraints, if any, posed by leveraged loan index providers licensing their indices for use in OTC products to leveraged loan index providers licensing their indices for use in funds/ETFs, seem to be minimal.

(482) Second, and reinforcing the above reasoning, for the majority of loan index (fund) customers it is important which loan pricing and reference data is used to calculate the loan indices they license. When asked whether loan data of IHSM, ICE, Refinitiv and Bloomberg are substitutable, half the respondents replied “No”, with the other half stating that it depended on the exact use. This highlights, that the substitutability for customers seems to be highly use case dependent and in any case, does not confirm the Notifying Party’s view that leveraged loan indices are easily substitutable from a customer perspective.

(483) Third, with respect to barriers to entry, most index providers generally do not think they are able to enter the loan index market quickly. One potential competitor in this space explains: “[Potential competitor] is currently considering various business cases for the fixed income index space, which may include leveraged loan indices. Access to the necessary data, expertise, and technology is already difficult due to opaqueness of the market.” Another competitor states: “[Competitor] has been trying for three or four years to enter the market of loan indices (including US Fund Bank Loan indices) and has so far not been successful.

(484) The competitor explains that the main loan index providers are also at the same time providers of the input data and considers that the prices quoted by existing loan index providers for their data amounts to “a constructive refusal to supply this data.” The competitor further explains that the data is not easy to source alternatively, i.e. directly from loan issuing banks “as that would require significant investment (in terms of time and costs), essentially equivalent to establishing a new data vendor, as it would require connecting to a large number of issuers.”

(485) Fourth, given that Credit Suisse and JP Morgan do not seem to pose significant competitive constraints on the Parties, it is of limited relevance that […]. In any case, given that IHSM is the market leader in loan pricing data, the threat to […] is not likely to amount to significant bargaining power.

(486) The Commission further considers that JP Morgan may not fully compete with IHSM, even when considering for the sake of argument, that leveraged loan indices licensed for the use of different financial instruments compete. […]. […]. Given that this aspect cannot be assessed further, the Commission considers it at least uncertain that JP Morgan competes entirely independently from IHSM.

(487) Based on the above considerations, the Commission considers that the market investigation and other evidence available to it do not confirm the Notifying Party’s views with respect to competitive dynamics in the leveraged loan index market. The Parties monopoly position Post-Transaction is further strengthened by IHSM being the market leading supplier of essential input, i.e. loan pricing data.

equity indices, natural resources sector equity indices and US Corporate Bond indices.

(496) First, IHSM’s increment in equity indices and natural resources sector indices, and S&P’s increment in US corporate bond indices are both virtually nil, at [0-5]%, with the delta in the HHI being less than 150. Since S&P is a stronger provider equity indices but a small player for supplying such services for fixed income indices, while the reverse is true for IHSM, they are not currently close competitors. Indeed, IHSM is not named as S&P’s closest competitor in these markets (or vice versa) by suppliers and customers.

(497) Second, strong competitors remain in each of the segments; in equity indices FTSE Russell and MSCI have market shares above [10-20]%, with a long tail of smaller competitors with market shares still larger than IHSM’s increment. In natural resources sector equity indices, MSCI, Morningstar and Solactive each have market shares of [10-20]% or above, again with a long tail of smaller competitors with market shares still larger than IHSM’s increment. Similarly, in US Corporate Bond indices, Bloomberg is larger than the combined entity, ICE and Nasdaq have market shares above [5-10]%, and FTSE Russell has a market share larger than S&P’s increment.

(498) Third, barriers to entry into specific segments such as natural resources sector equity or US corporate bond indices appear low for existing equity and fixed income index suppliers respectively. Therefore, any supplier active in index licensing but not currently active in the supply of calculation and administration services could enter, or if they are already active with a small share, could expand their supply without significant additional investment.

(499) Finally, the market investigation did not provide any indication that post-Transaction there would not remain enough competition in these segments. Indeed, no respondent raised concerns with regard to these segments.

(B.iv) Conclusion

(500) For the above reasons, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market in relation to the putative global markets for equity indices, natural resources sector equity indices and US Corporate Bond indices.

6.3.2.3. Index calculation and administration services (horizontal overlap)

(A) The Parties’ activities

(501) Each Party is active in the supply of calculation and administration services for (equity and fixed income) indices. Table 7 below shows the market shares of S&P,

390 As per paragraph 20 of the Horizontal Merger Guidelines, the Commission is unlikely to identify horizontal competition concerns in a merger with a post-merger HHI between 1 000 and 2 000 and a delta below 250, or a merger with a post-merger HHI above 2 000 and a delta below 150, except where special circumstances apply. The MI did not indicate that any of the special circumstances mentioned apply.

391 Replies to questions 15 and 17 of Questionnaire 4, and questions 14 and 23 of Questionnaire 5.

392 Replies to questions 59, 64 of Questionnaire 4 and questions 56, 58 of Questionnaire 5.

from [20-30]% to [20-30]%. Bloomberg, DAX-STOXX/Qontigo (Deutsche Börse) and FTSE Russell (LSEG/Refinitiv) all have market shares exceeding [5-10]% and there is a large tail of smaller competitors who could easily expand their position. This tail of competitors includes large indices providers (e.g. MSCI, ICE) who are well positioned to provide these add-on services.

(505) Fourth, barriers to entry for the supply of calculation and administration services are very low. It is very easy for small start-ups to start supplying calculation and administration services, which is reflected by the successful growth of Solactive in recent years and the entry by suppliers such as Moorgate Benchmarks. Solactive’s current estimated share of index calculation and administration services is [20-30]%. However, the Parties consider that this underrepresents Solactive’s competitive impact. Indeed, Solactive provides a low-cost and flexible service, exerting significant pricing pressure on other suppliers and the Parties expect Solactive to continue to expand its share of the calculation and administration services market going forward. The Notifying Party adds that it is also very easy for suppliers active in index licensing to supply third parties with index calculation and administration services. No material investment is required to do so, as they will already have in-house calculation and administration services. Therefore any supplier active in index licensing is a potential entrant to the supply of calculation and administration services.

(C) The Commission’s assessment

(506) Based on the evidence available to it, the Commission finds that the Transaction is unlikely to give rise to competition concerns in the putative global markets for index calculation and administration services, or for such services segmented by equity indices and fixed income indices.

(507) First, while S&P is a stronger provider of index calculation and administration services for equity indices, it is a small player for supplying such services for fixed income indices, while the reverse is true for IHSM. As such, S&P and IHSM are not currently close competitors. Indeed, IHSM is not named as S&P’s closest competitor in these markets (or vice versa) by any responding suppliers, and only as the second closest competitor by one responding supplier.

(508) Second, strong competitors remain in each of the segments; in the overall index calculation and administration services Solactive’s market share ([20-30]%) is very close to that of S&P ([20-30]%), and has been increasing, and there are three other providers with market shares above [5-10]% ([10-20]%, [5-10]% and [5-10]%). In index calculation and administration services for equity indices, Solactive and DAX both have market shares well above that of IHSM’s increment ([20-30]%, [10-20]% and [5-10]% respectively). Similarly, in index calculation and administration services for fixed income indices, Solactive is larger than the combined entity, and there is a long tail of smaller competitors.

(509) Third, barriers to entry or expansion appear low, as large index providers already have in-house calculation and administration services for their own indices. Therefore, any supplier active in index licensing but not currently active in the supply of calculation and administration services could enter, or if they are already active with a small share, could expand their supply without significant additional investment. Calculation services include daily maintenance and calculation of the index, application and treatment of corporate actions, index distribution, and the supply of constituent (calculated) data files to the customer. Administration services include overseeing the index methodology and any changes thereto etc., i.e. owning the rulebooks and performing consultations for changes. A supplier with the requisite data, infrastructure and processes in place can enter or expand without a large incremental burden.

(510) Finally, while the combined market share in certain plausible segments is not low, the market investigation did not provide any indication that post-Transaction there would not remain enough competition in these segments. Indeed, the majority of responding index competitors and customers expect a neutral or positive impact on the index calculation and administration services markets.

(D) Conclusion

(511) For the above reasons, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market in relation to the global market for index calculation and administration services, or any narrower plausible segment.

6.3.3. Affected markets – vertical relationships

6.3.3.1. Loan identifiers (upstream) – Leveraged loan market intelligence (downstream)

(512) The Commission assesses below whether the merged entity would have the ability and incentives to foreclose access to LXIDs to S&P’s competitors in leveraged loan market intelligence as well as the likely impact of such foreclosure.

(A) The Parties’ activities

(513) Upstream, both Parties are active in loan identifiers (see Section 6.3.2.1). There are no independent public sources with respect to market shares in the loan identifier market. However, the Commission considers that the market coverage of loan identifiers as compared by IHSM within its loan pricing and reference data would seem to be representative of the whole market based on the responses to the market investigation.

(514) Downstream, S&P is active in leveraged loan market intelligence with its Loan Commentary and Data (LCD) product. IHSM is not active in this market.

(C) The Commission’s assessment

(C.i) Input foreclosure

Ability

(519) All responding competitors in leveraged loan market intelligence consider loan pricing and reference data, including loan identifiers to be a very important input. One competitor states: “The only data from IHSM that’s a need-to-have is the LXIDs – the pricing data is a nice-to-have but there are viable substitutes.” Another competitor explains: “Leveraged loan market participants will be interested in at least some reference data to understand the borrower and characteristics of the loan. Participants who transact in leveraged loans, or who are looking to derive certain types of credit metrics, will need access to loan pricing.”

(520) In response to the question whether not having access to IHSM’s data would degrade their product to an extent that customers would stop buying leveraged loan market intelligence of LCD’s rivals, the majority of competitors answers “It depends”. However, this is mainly driven by the fact that one competitor is currently in negotiations with IHSM with respect to LXIDs. Two other competitors or potential competitors have tried sourcing LXIDs previously but were quoted a price that was considered too high. One competitor explains: “(…)customers have asked [competitor] if they could use LoanX IDs. 10 years ago [competitor] made an attempt to get a license for LoanX IDs, but it was very expensive. [Competitor] would consider licensing LoanX IDs if the terms were appropriate to satisfy its customers’ needs.” The high price quoted by IHSM for access to its LXIDs was likely in that case a result of the requesting company competing with IHSM in other downstream markets. The Commission notes, that this past behaviour clearly evidences the ability of the merged entity to foreclose access to LXIDs. The Transaction will increase those incentives and extend them to leveraged loan market intelligence, given the presence of S&P in the downstream market.

(521) The remaining competitors in leveraged loan market intelligence are using LXIDs and have contracts with IHSM. One competitor states: “In order to link its information to the information in the form used by customers, as a practical matter, [competitor] must provide identifiers. For the loan market, although this includes LoanX IDs (LXIDs) and CUSIPs (US and Canada), LXIDs are the de facto identifier used.”

(522) Customers confirm this view. The majority of customers considers receiving LXIDs and loan CUSIPs as part of their leveraged loan market intelligence as important. This is because customers require loan identifiers in order to map information, analysis and news in LCD to their loan trading activities which include the monitoring of a portfolio of loans, loan settlement and cross-referencing to other information, and they consider LXIDs the market standard loan identifier, with loan CUSIPs being often referred to as the second best alternative.

(523) Furthermore, S&P’s LCD product also includes LXIDs in one part of the product, namely the European Leveraged Loan Index (ELLI) calculated as part of LCD. Some customers consider the ELLI a “very important” part of LCD.

(524) Based on the responses to the market investigation, the Commission considers that loan identifiers and in particular LXIDs appear to be a critical input to leveraged loan market intelligence.

(525) The fact that leveraged loan market intelligence products are differentiated and cater to different customer demands does not change the importance of LXIDs as an input. One of the competitors downstream considers that the use of LXIDs contributes to the competitiveness of their product. The competitor is constrained in their choice of loan identifier by the input data received from another provider, which uses LXIDs: “[Competitor] receives CLO data from a business partner with LXIDs and hence requires loan pricing data with LXIDs to be able to map end-of-day loan prices to its already existing database. [Competitor] considers its mapping of CLO data to companies to be an important product feature.” This demonstrates again the network effects of loan identifiers, which benefit the market leading provider IHSM. With respect to the question whether other loan identifiers are effective substitutes for LXIDs, the Commission understands from the responses to the market investigation as well as from the coverage figures provided by the Notifying Party, that other loan identifiers do not offer the requisite coverage. Coverage of all loan identifiers competing with the Parties’ loan identifiers is significantly lower. As set out in the chapter assessing the horizontal overlap in loan identifiers above, the competition between loan identifiers is already not very strong pre-Transaction. Furthermore, access to other identifiers is not a sufficient substitute for competitors, who are reacting to customer demands for LXIDs. A competitor states with respect to other loan identifiers: “[Competitor] notes that Bloomberg and Refinitiv also provide loan identifiers but does not see those identifiers as an alternative to LXIDs or CUSIPs. According to [the competitor’s] observations, loan information customers only primarily use LXIDs for content mapping, followed by CUSIPs. The usage of LXIDs includes mapping data sets internally as well as to external data feeds, supporting settlement and trading and secondary mark-to-market pricing.” As a consequence, the Notifying Party’s view that S&P’s LCD rivals could easily substitute LXIDs with another loan identifier is already not credible for this reason. Furthermore, the only other loan identifier which is considered as a close substitute is S&P’s loan CUSIP. With the addition of loan CUSIPs, the Transaction strengthens the already strong position upstream of IHSM in loan identifiers.

(526) The above assessment of the horizontal overlap in loan identifiers already showed that switching costs are high and that customers would unlikely change provider in case of partial price-based foreclosure of LXIDs.

(527) In addition, S&P is currently a strong supplier of leveraged loan market intelligence, with the largest global market share, and could easily address demand of new customers from the foreclosed competitors.

(528) In summary, based on the evidence available to it, the Commission considers that the merged entity would have the ability to either partially or fully foreclose access to LXIDs to rival leveraged loan market intelligence providers.

Incentives

(529) The Notifying Party argues that the merged entity would have no incentives to foreclose access to LXIDs since it would have to refuse access to its loan pricing and reference data which would reduce its profits not only for LXIDs, and it would not be likely that the diversion downstream would make up for that loss.

(530) However, even taking into account the sales of loan pricing and reference data overall, sales of the upstream product to the downstream market’s rivals of the merged entity are relatively low (less than USD […], i.e. ca. […]% of revenues). As a result, foreclosure would be profitable even with very low customer switching or customers adding LCD (below 1%) .

(531) In order to estimate the profits at risk of being lost if the merged entity forecloses its upstream product, the Commission considered the profit margin of loan pricing and reference data upstream (given that no separate revenues/margins are available for LXIDs, as IHSM currently does not commercialize them separately from its loan pricing and reference data). The EBITDA margin of the pricing valuation and reference business segment of IHSM which includes LXIDs stood at […]% according to the Notifying Party. Applied to the sales of loan pricing data to the downstream market of leveraged loan market intelligence providers of USD […] this results in profits potentially at risk of USD […].

(532) These sales of USD […] have been identified by the Notifying Party as total sales of loan pricing data, including LXIDs to the downstream market. However, USD […] thereof are sales to [Customers] who do not compete with S&P’s LCD in the downstream market. A possible foreclosure of loan pricing data including LXIDs only to S&P’s downstream competitors would therefore correspond to much fewer profits at risk upstream and would be even more profitable from a financial incentives perspective.

(533) The Commission compares profits lost upstream to the potential increase in profits downstream from additional sales of S&P’s LCD. The downstream revenue of S&P amounted to USD […] The downstream EBITDA margin of the business segment covering S&P’s leveraged loan market intelligence stood at […]% in 2020 according to the Notifying Party. Therefore the operating profits downstream derived from the existing market share of S&P amounted to USD […].

(534) The profits which could be gained from a foreclosure strategy have to be assessed taking into account the existing market share of S&P downstream. The merged entity would only be able to increase its market share capturing customers that do not already purchase S&P’s market intelligence products. To take account of this, the Commission reduced profits to be potentially gained by [20-30]% which is the current market share of S&P downstream according to the Notifying Party.

(535) The proportion of the market purchasing loan pricing data, including LXIDs upstream and leveraged loan market intelligence downstream which is susceptible to switch or add S&P as a result of the foreclosure is indeed not 100%. It is corrected for the market participants which do not purchase loan pricing data, including LXIDs upstream and corrected for market participants which already purchase the S&P product downstream and would therefore not need to add an S&P downstream solution. The Commission considers for this purpose that between 35.8 and 57% of customers using loan pricing data including LXIDs upstream and leveraged loan market intelligence downstream are susceptible to switch or add S&P solutions downstream following a foreclosure. This figure is calculated by multiplying IHSM’s market share upstream ([50-60]% when IHSM’s market share in loan pricing data is considered, [80-90]% when IHSM’s market share in LXIDs is considered) by [70-80]%, the combined market share of competitors of S&P downstream.

(536) Currently S&P derives an operating profit of around USD […] for each percentage point of market share downstream. Therefore, the operating profits downstream of S&P could in theory increase by up to USD […], if S&P could capture the full […]% of the market which is susceptible to switch in the case of a foreclosure strategy.

(537) The Commission calculated the critical switching rates for which the operating profits lost upstream would be outweighed by operating profits gained downstream. As IHSM is not deriving large revenues upstream currently from LXIDs compared to the large revenues derived by S&P downstream, the switching rates sufficient to create incentives are […]. From a financial incentives perspective it would (conservatively taking into account the higher figure of upstream sales to the downstream market of USD […]) be sufficient that 3-5% of customers upstream add the S&P downstream product. More than 3% of respondents to the market investigation considered that they would add the S&P downstream product if LXIDs would no longer be provided to downstream competitors.

(538) In the alternative, where a foreclosure strategy would only exclude direct downstream competitors of the S&P product from distributing and accessing loan pricing data including LXIDs, the critical switching rate would be below 1%.

(539) As explained above, S&P is the market leader downstream with its LCD product which points to the attractiveness of this product for customers and potential to switch. The Notifying Party argues that rival products downstream are no perfect substitutes to LCD as the content of all products competing in that space is differentiated. However, as explained above in Section 6.2.11, all leveraged loan market intelligence products pertain to a single market and are substitutes at least to some extent. The differentiated nature of certain offerings appears also insufficient to dismiss the existence of incentives knowing that a switching of less than 1% of customers would already render the foreclosure strategy profitable.

(540) This increases incentives to foreclose, as the merged entity could be relatively certain that customers will not easily substitute LXIDs with another loan identifier.

(541) Past behaviour of IHSM vis-à-vis two downstream competitors in other markets using LXIDs as an input also shows that IHSM had the ability and incentives to foreclose in the past, and effectively partially foreclosed access to LXIDs. Two competitors in a downstream market requested access to LXIDs in the past, but were quoted very high prices, amounting to, as one of them put it “a constructive refusal to supply”. The downstream market concerned by this past behaviour is relatively concentrated, possibly also as a result of the foreclosure of an upstream product (LXIDs) that had developed network effects over time, and in which IHSM is the strongest supplier.

(542) Finally, as already mentioned above, the large majority of competitors in the broader pricing and reference data and market intelligence space do not consider that they provide any essential inputs to the merged entity that would award them any negotiating power. The Notifying Party’s argument that the merged entity will lack incentives due to retaliation risks from S&P’s competitors therefore does not seem credible.

Impact

(543) Overall, responses of competitors with respect to the impact of the Transaction on loan pricing data and loan identifier markets are unanimously negative (not counting uninformative responses). One competitor expresses its concerns in the following terms: “[Competitor] is concerned that the merged entity might refuse access to LXIDs or offer them on worse terms post-transaction, e.g., unreasonable price increases, or refusal to allow a license for redistribution. [Competitor] estimates that following the merger, S&P could decide to apply similar terms as currently apply to CUSIP also to LXIDs (i.e. increase prices, e.g. by defining many different use cases, restrict use of LXIDs when re-distributing to end-customers who are not licensed directly by S&P).”

(544) Customers’ views on the impact of the Transaction on the leveraged loan market intelligence market are also not positive, with several negative views.

(545) Such foreclosure strategy could not only increase prices for end-customers, but also impact choice and innovation. Competitors highlight that access to LXIDs enable them to provide competitively promising products. One competitor explains: “[Competitor] considers its mapping of CLO data to companies to be an important product feature.” and adds: “[Competitor] requires identifiers for (…) for providing the results of its analyses and data structuring to customers including identifiers that customers require/use. (…) For loans, this would be predominantly LXIDs and to a lesser extent FIGIs and loan CUSIPs as their coverage is more limited.” Another competitor states: “In order to link its information to the information in the form used by customers, as a practical matter, [competitor] must provide identifiers. For the loan market, (…) this includes LoanX IDs (LXIDs) and CUSIPs (US and Canada).”

(C.ii) Customer foreclosure

(546) Customer foreclosure in relation to those markets is not assessed in detail given that S&P’s market share downstream is [20-30]% which means that S&P would likely not have sufficient market power to effectively foreclose market access to LXIDs rivals upstream. Furthermore, loan identifiers are used in many different downstream products apart from leveraged loan market intelligence (e.g. loan pricing and reference data, leveraged loan indices, credit ratings, corporate fundamentals data and company credit risk analytics). In addition, no respondent to the market investigation expressed concerns with respect to possible customer foreclosure.

(D) Conclusion

(547) Based on the market investigation and other evidence available to the Commission, the Commission considers that the Transaction raises serious doubts as to its compatibility with the internal market and the functioning of the EEA Agreement with respect to the vertical relationship between the Parties’ activities in the plausible global market for loan identifiers (upstream) and the plausible global market for leveraged loan market intelligence (downstream) by providing the ability and incentives to fully or partially refuse access to loan identifiers to the merged entity’s rivals downstream. This would lead to rivals’ being at least partially foreclosed and competing less effectively against the leading player S&P, to the detriment of choice and quality for end-customers.

(551) In any case, the Notifying Party claims that the merged entity would have neither ability nor incentives to foreclose loan reference data, given a) its modest presence upstream, b) the low importance of the input for the downstream product, c) foreclosure would not be profitable as the merged entity would give up profits upstream with no realistic diversion of customers downstream, and d) downstream competitors are also important input providers to the merged entity and foreclosure would risk retaliation.

(C) The Commission’s assessment

(552) Several downstream competitors responding to the market investigation do report that they use loan reference data and loan identifiers as an input to their company fundamentals data offering. As noted in the above chapter on the horizontal overlap in loan identifiers, the Parties’ position upstream is very strong and likely dominant even with IHSM’s LXIDs only. The Commission therefore considers that the merged entity would in principle have the ability to foreclose access to LXIDs for competing fundamentals data providers. However, the Commission notes that on the downstream market, S&P is currently only the 3rd or 4th player, behind Refinitiv and FactSet, and of comparable size to Bloomberg, and hence may not have the ability to fully capture demand lost by competitors.

(553) A market participant expressed specific concerns in relation to LXIDs as an input: “[Competitor] is concerned about a potential full foreclosure of LXID identifiers post-Transaction, which would prevent [competitor] from properly matching instruments with issuers, decreasing its content quality in the Fundamentals data market. For customers, the ability to link loan pricing and reference data with fundamentals and ratings data is particularly important, because it supports full company level capital structure and credit analysis.” However, the competitor complaining about potential foreclosure of LXIDs does not expect the Transaction to have a negative impact on the fundamentals data market. In addition, no other providers in the downstream market raised possible concerns. Other providers in the downstream market did not raise this concern and currently operate successfully without LXIDs as an input, with a combined market share of [50-60]-[60-70]%. This points to LXIDs potentially not being essential in order to compete or the merged entity not having the ability to foreclosure the downstream market.

(554) With respect to incentives, the Commission notes that S&P’s market share is relatively limited and hence the market share based on which the merged entity would benefit from a higher price level as a result of raising rivals’ costs would be limited. Second, there are several competitors downstream, most of which operate their fundamentals data business while not having access to LXIDs or wishing to have access to LXIDs, including some with higher market shares than S&P.

(555) In terms of impact of possible foreclosure, several downstream competitors highlight that they currently do not source this input from IHSM and do not expect an impact of the transaction on their businesses or on competition.

(556) From a customer perspective, views are mixed, with some customers considering it important to be able to match bank loan transactions/pricing with other fundamentals data, but others not. Ultimately, as indicated above, the competitor complaining about potential foreclosure of LXIDs does not expect the Transaction to have a negative impact on the fundamentals data market.

(D) Conclusion

(557) Based on the market investigation and other evidence available to it, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market and the functioning of the EEA Agreement with respect to the vertical relationship between the Parties’ activities in the plausible global market for loan identifiers (upstream) and the plausible global market for fundamentals data (downstream).

6.3.3.3. Loan pricing and reference data (upstream) – Leveraged loan market intelligence (downstream)

(558) The Commission assesses below whether the merged entity would have the ability and incentives to foreclose access to IHSM’s loan pricing and reference data to S&P’s competitors in leveraged loan market intelligence as well as the likely impact of such foreclosure.

(A) The Parties’ activities

(559) Upstream IHSM is active in loan pricing data with a market share of ca. [50-60]% and loan reference data with a market share of ca. [5-10]% (the market share in an overall market of loan pricing and reference data would be ca. [30-40]%). IHSM collects loan pricing data mainly from financial institutions trading in loans (market-making desks primarily but also information from the buy-side) in the form of so-called “dealer runs” (quotes provided by banks to their customers on the terms on which they are willing to buy or sell a loan, which are provided to IHSM electronically) and messages. The data covers bid-offer pricing analytics and liquidity measures for over […] leveraged loan facilities globally, as well as some private or “club” loans.

(560) Downstream, S&P is in the leading provider of leveraged loan market intelligence with a market share of [20-30]% (see Section 6.3.3.1).

(561) All competitors responding to the market investigation are of the view that loan pricing and/or reference data are an important input into leveraged loan market intelligence. One competitor states: “Loan pricing and reference data are key inputs for leveraged loan market intelligence, without which such market intelligence would have little / no content and therefore be of little value.” However, another competitor states: “The only data from IHSM that’s a need-to-have is the LXIDs – the pricing data is a nice-to-have but there are viable substitutes.” Another competitor explains: “Participants who transact in leveraged loans, or who are looking to derive certain types of credit metrics, will need access to loan pricing.” While that answer is not entirely clear with respect to whether loan pricing and reference data is a necessary input to leveraged loan market intelligence, it highlights that the need for this input may vary depending on the customer type.

(565) Competitors were generally of the view that several of the other loan pricing data providers (Refinitiv, Bloomberg, Solve Advisors) provide a loan pricing data product that is a credible alternative to IHSM’s loan pricing data. Best Credit Data and Advantage are not considered credible alternatives, while Solve Advisors, who are not figuring in the above market shares, was considered an alternative. The same is true for most of the other loan reference data providers (Refinitiv, Bloomberg).

(566) The Commission notes that even though loan pricing and reference data would seem to be considered an important input by most competitors in leveraged loan market intelligence, all competitors agree that sufficient alternative providers of loan pricing and reference data will remain on the market post-Transaction. No provider of leveraged loan market intelligence indicates that not having access to IHSM’s loan pricing and reference data would degrade their product.

(567) If a wider market upstream was considered, i.e. a market comprising both loan pricing and reference data, the Commission notes that IHSM’s market share in the loan reference data market (ca. [5-10]%) is significantly smaller than in loan pricing data (ca. [50-60]%), so that IHSM’s market share upstream would be lower (ca. [30-40]%) in a broader market. While the number of competitors able to provide both loan pricing and reference data is smaller (i.e. effectively limited to Bloomberg and Refinitiv), those two other competitors have significant market shares ([20-30]% and [20-30]% respectively in the broader market comprising both loan pricing and reference data), and are considered credible alternatives by market participants.

(568) Second, in terms of incentives, sales of the upstream product to downstream competitors are very low (just USD […]), and hence could speak rather in favour of incentives, given that the profit lost upstream would be very limited. Given that there are alternative providers the incentives to fully foreclose may be limited, as downstream competitors could not effectively be fully foreclosed by the merged entity. A partial foreclosure in the form of price increases could however be profitable given that less profit would be lost upstream and switching loan pricing data provider for downstream competitors is not fast and easy. In any case, there is no need to conclude on whether the merged entity would have incentives to foreclose downstream rivals, as the Parties have offered to divest S&P’s leveraged loan market intelligence business to remedy the concerns raised in relation to the vertical link between loan identifiers upstream and leveraged loan market intelligence downstream, thus removing any possible incentive.

(569) The Commission also does not consider that the retaliation risk from […] in case of a foreclosure strategy in regard to loan pricing data is particularly strong, given that S&P is also active in […], including with much higher market share and coverage than […], and could in turn restrict access to its credit ratings.

(570) However, in any case, no competitor expressed concerns with respect to customers switching away from their leveraged loan market intelligence product, if they no longer had access to IHSM’s loan pricing and/or reference data. Based on the above responses to the market investigation, the Commission concludes that the merged entity would likely not have the ability or incentives to foreclose access to IHSM’s loan pricing data post-Transaction, and that therefore, there is unlikely to be a significant impact of the Transaction on competition in those markets as a result of the Transaction.

(571) Customer foreclosure in relation to these markets is not assessed in detail given that S&P’s market share downstream is [20-30]% which means that S&P would likely not have sufficient market power to effectively foreclose market access to IHSM’s rivals upstream. Furthermore, loan pricing data is used in many different downstream products apart from leveraged loan market intelligence (e.g. leveraged loan indices, credit ratings and company credit risk analytics). In addition, no respondent to the market investigation expressed concerns with respect to possible customer foreclosure.

(D) Conclusion

(572) Based on the market investigation and other evidence available to the Commission, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market and the functioning of the EEA Agreement with respect to the vertical relationship between the Parties’ activities in the plausible global market for loan pricing and/or reference data (upstream) and the plausible global market for leveraged loan market intelligence (downstream).

6.3.3.4. Loan pricing and reference data (upstream) – Company credit risk analytics (downstream)

(573) The Commission assesses below whether the merged entity would have the ability and incentives to foreclose access to IHSM’s loan pricing and reference data to S&P’s competitors in company credit risk analytics as well as the likely impact of such foreclosure.

(A) The Parties’ activities

(578) In terms of incentives, given the high margins of IHSM’s loan pricing data upstream ([…]%), giving up revenue upstream to potentially recoup revenues through increased margins or a higher market share downstream seems unlikely to be profitable. In particular, there are unlikely to be incentives for full foreclosure as this is unlikely to be effective, given the presence of other providers upstream that are considered credible alternatives. This is even more the case, given that the loan pricing data being used as an input in the downstream product is one of many inputs and it is unlikely to make a distinguishable difference for customers of the downstream product, from which upstream provider this data is sourced.

(579) Furthermore, the Commission notes that none of the respondents to the market investigation (including S&P’s main competitors in the downstream market) expressed foreclosure concerns with respect to this vertical link.

(580) Customer foreclosure in relation to these markets is not assessed in detail given that S&P’s market share downstream is less than [5-10]% which means that S&P would likely not have sufficient market power to effectively foreclose market access to IHSM’s rivals upstream. Furthermore, loan pricing data is used in many different downstream products apart from company credit risk analytics (e.g. leveraged loan market intelligence, leveraged loan indices, credit ratings). In addition, no respondent to the market investigation expressed concerns with respect to possible customer foreclosure.

(D) Conclusion

(581) Based on the market investigation and other evidence available to it, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market and the functioning of the EEA Agreement with respect to the vertical relationship between the Parties’ activities in the plausible global market for loan pricing and/or reference data (upstream) and the plausible global market for company credit risk analytics (downstream).

6.3.3.5. Loan pricing and reference data (upstream) – Leveraged loan indices (downstream)

(A) The Parties’ activities

(582) Upstream IHSM is active in loan pricing data with a market share of ca. [50-60]% (see chapter 6.3.3.3) and loan reference data with a market share of ca. [5-10]% (the market share in an overall market of loan pricing and reference data would be ca. [30-40]%). Downstream, S&P and IHSM are both active in leveraged loan indices, S&P with a market share of [90-100]% and IHSM with a market share of [0-5]% (see chapter 6.3.2.2).

Margin data is only available at the level of pricing data overall ([…]% for pricing data in 2021 based on […]), see Notifying Party’s response to RFI 19, Annex 1.

(B) The Notifying Party’s view

(583) According to the Notifying Party, the merged entity would not have the ability to foreclose access to loan pricing data as access to IHSM’s loan pricing data is not necessary to compete in the supply of leveraged loan indices. S&P itself does not source its loan pricing data from IHSM but from [Names of S&P's suppliers]. Apart from that, the Notifying Party submits that ICE and Bloomberg are credible alternative suppliers of this data.

(584) In terms of incentives, the Notifying Party notes that foreclosure would not result in customers diverting to S&P’s leveraged loan indices as alternative loan pricing data input is available and leveraged loan indices are in any case substitutable with other indices tracking floating-rate fixed income securities (some leveraged loan indices benchmark against Bloomberg or ICE fixed income indices).

(C) The Commission’s assessment

(585) In terms of possible input foreclosure, the Commission notes that in its assessment of the horizontal overlap in leveraged loan indices, some evidence pointed to IHSM being one of only two providers of the relevant loan pricing input data for leveraged loan indices. This is because index providers require the “highest quality” pricing data in the form of actual prices or quotes submitted when trading loans. A majority of index customers responded that it is important/relevant for them which loan pricing and reference data is used to calculate loan indices they license. The respondents mention as possible suppliers mostly IHSM and Refinitiv, supporting the evidence cited above. When asked whether IHSM, ICE, Refinitiv and Bloomberg are substitutable in terms of the loan data they provide, most customers reply that this depends on the product for which that data is used.

(586) In any case, the only downstream competitors are currently S&P and IHSM, so only potential competitors could be foreclosed.

(587) From a customer foreclosure perspective, the merged entity would have the ability to stop purchasing from upstream competitors of IHSM. However, the upstream competitors also sell their loan pricing and reference data to other customers and not only to providers of leveraged loan indices. The proportion of IHSM’s total sales in the upstream loan pricing and reference data market to the downstream leveraged loan index market are [0-5]% and [20-30]% respectively. Hence, a foreclosure of IHSM’s upstream rivals is unlikely to have any impact on their ability to compete upstream.

(588) In any case, given the serious doubts raised in relation to the horizontal overlap in leveraged loan indices, for which the Notifying Party has offered S&P’s downstream business as a commitment, see chapter 6.4.4, the Commission does not need to conclude its assessment on incentives and impact, as the divestment eliminates possible incentives (by removing the overlap resulting from the Transaction) and hence, no concerns can plausibly arise.

(C) The Commission’s assessment

(594) The Commission’s market investigation confirms the Notifying Party’s views with respect to this link. Competitors do not consider IHSM’s loan pricing data as particularly unique or indispensable as an input to their credit ratings. A competitor downstream explains: “[Competitor] particularly does not use any IHSM data directly as input to its rating analysis. [Competitor] do use S&P’s SNL product as an ancillary input but could, if needed, also produce those data by itself with some time.” Several competitors remain upstream that are considered equally credible. Therefore, the Commission considers that the merged entity will have no ability to foreclose loan pricing and/or reference data to competitors of S&P in credit ratings.

(595) In terms of incentives, customers do not choose a credit rating agency based on the loan pricing and/or reference data they use. Hence, a foreclosure of this data would (a) not have any significant effects on rival credit rating agencies and (b) not lead to diversion of customers because of the foreclosure.

(596) Finally, several competitors in credit ratings of different sizes are not concerned with respect to foreclosure of loan pricing or reference data.

(D) Conclusion

(598) Based on the market investigation and other evidence available to it, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market and the functioning of the EEA Agreement with respect to the vertical relationship between the Parties’ activities in the plausible global market for loan pricing and/or reference data (upstream) and the plausible global market for credit ratings (downstream).

6.3.3.7. CDS pricing data (upstream) - Company credit risk analytics (downstream)

(599) The Commission assesses below whether the merged entity would have the ability and incentives to foreclose access to IHSM’s CDS pricing data to S&P’s competitors in company credit risk analytics as well as the likely impact of such foreclosure.

(C) The Commission’s assessment

(604) The Commission notes that a number of market participants believe that the Transaction will have a negative impact on the CDS pricing data market, but none of the respondents are competitors in company credit risk analytics. Some of the respondents explain their negative views with a (not further specified) risk of bundling, but do not substantiate their views. Apart from that, none of the respondents to the market investigation raised concerns in relation to CDS pricing as an input to company credit risk analytics. Respondents to the market investigation confirm the Notifying Party’s view that there are a number of credible alternatives for IHSM’s CDS pricing data. The Commission therefore considers that the merged entity would not seem to have the ability to foreclose its CDS pricing data to downstream rivals of S&P.

(605) The Commission received one complaint with respect to possible input foreclosure in relation to CDS standard rate curves. One competitor claims that there is an input relationship between IHSM’s standard rate curves and S&P’s Credit Default Swaps Market Derived Signals (CDS MDS). The Notifying Party however confirms that S&P’s product uses CDS pricing data as an input, but not the CDS standard rate curve. Even if competing downstream products were to use the CDS standard rates curve as an input (of which IHSM is the only provider globally), the addition of a [5-10]% market share downstream does not appear likely to change incentives for the merged entity significantly. No other market participants expressed any concerns with respect to this link. The Commission further notes that the relevant competitor provides a number of inputs to the merged entity that would seem to award certain bargaining power. Hence, the Transaction is unlikely to have an effect on competition in the company credit risk analytics market, be it as a result of the upstream link with CDS pricing data or the potential market for a CDS standard rate curve.

(606) Customer foreclosure in relation to these markets is not assessed in detail given that S&P’s market share downstream is ca. [10-20]% which means that S&P would likely not have sufficient market power to effectively foreclose market access to IHSM’s rivals upstream. Furthermore, CDS pricing data is used in many different downstream products apart from company credit risk analytics (e.g. leveraged loan market intelligence, leveraged loan indices, credit ratings). In addition, no respondent to the market investigation expressed concerns with respect to possible customer foreclosure.

preclude S&P from hypothetically favouring SPDJI’s indices operations. And (ii) S&P licenses CUSIPs on a FRAND basis and would suffer reputational and commercial damages reneging on this commitment. Moreover, the Notifying Party submits that in this case the ABA would almost certainly remove S&P’s licence as the administrator of the CUSIP system, since the ABA has an interest in making CGS’s data as broadly available, accessible and adopted in the financial industry as possible; any commercial practice that may restrict access to those identifiers would undermine their status as market standards, with a negative impact on the value of the ABA’s intellectual property.

(b) S&P would also have no incentive to foreclose access to CUSIP data as, (i) There is already a vertical link with S&P’s pre-existing indices business, but CGS has never restricted access to CUSIPs to support SPDJI’s established equity indices business or to encourage the entry / expansion of the fixed income indices business. There is no reason why it should attempt to do so post-Transaction, in particular given the limited change downstream (the Transaction will have a negligible impact on S&P’s equity indices business and the combined fixed income indices business [S&P revenue information] than S&P’s existing equity indices business where, as above, S&P has never sought to foreclose access to CUSIP data). Indeed there is no plausible basis on which S&P would wish to do so given the potentially significant ramifications on its wider CUSIP business, as detailed above. And (ii) it would face retaliation risk from downstream indices competitors, in particular from companies like [names of S&P's suppliers], all of whom supply important inputs to S&P currently across a range of product areas.

(C) The Commission’s assessment

(629) The Commission’s investigation indicates that the Transaction is unlikely to give rise to input foreclosure.

(630) First, the Commission notes that the governance of the CUSIPs and the oversight by the ABA provide a certain comfort that S&P would not be able to favour its own downstream businesses to the detriment of rivals. The ABA confirms that they focus on this aspect and have not observed such behaviour by S&P in the past; “One of the most important factors that gives ABA confidence that CUSIPs are managed as agreed is that [summary of contractual arrangements between CGS and the rest of the S&P business] no S&P business receives preferential treatment relative to third parties…ABA considers that these restrictions have in the past been respected.” In terms of the operational governance, the ABA confirms that it “is in frequent contact with CGS, and regularly monitors all aspects of the agreement between ABA and CGS.” Lastly, the S&P would be deterred from anti-competitive behaviour in breach of its understanding with the ABA, since while the ABA “has never in the past

(631) Second, and relatedly, the Commission notes that S&P has a public FRAND commitment, which, if effective, would deprive the company from the ability to engage in input foreclosure. CGS’ statement on its website reads that “CGS seeks to charge fair, reasonable and non-discriminatory license fees for providing the convenience and functionality of direct or indirect access to and benefit of CGS Data…”. However, this commitment results from a voluntary decision of S&P which the company could formally decide to overturn one way or another, even if it led to some reputational cost, as mentioned by the Notifying Party. In addition, there is no formal mechanism set up for customers to enforce this commitment. The extent to which the FRAND commitment is effective is uncertain, in particular in relation to the fairness of prices and licensing practices for CUSIPs. With regard to CUSIPs, one respondent replied “We have not experienced massive price increases on this segment of data” while another respondent referred to the “charging of high monopoly rents” from CUSIPs and “exploitative” behaviour. While, for the purposes of this decision, the Commission is unable to comment on the fair or excessive nature of CUSIP prices, this trend of continuous price increase may indicate that the FRAND commitment is partially ineffective. However, the Commission also notes that the “ABA sometimes receives enquiries by their members on whether CGS could offer them discounts given their ABA membership, but ABA’s response is always that all CGS’s relationships are on an arm’s length basis and no one can receive preferential treatment. Other than that, the ABA does not receive many complaints from the market place (CGS needs to inform the ABA of any complaints it receives) relating to the FRAND basis of CUSIPs” (emphasis added). Moreover, S&P’s FRAND commitment could deter de facto total foreclosure strategies which would be more easily detectable, as they would amount to a refusal to supply.

(632) Third, the Commission notes that past behaviour speaks against ability and incentive for input foreclosure of CUSIPs; indeed, S&P is already active downstream of CUSIPs in several different markets, including equity indices, which are larger than the markets considered here. It would not be likely that the combined entity would have the ability or incentive post-Transaction to implement input foreclosure for smaller potential gains downstream.

(633) Fourth, and last, no respondent to the market investigation raised concerns in relation to the impact of the Transaction in any of these markets due to an input foreclosure of CUSIPs.

(D) Conclusion

(634) For the above reasons, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market in relation to the vertical relationships arising between CUSIPs (upstream) and indices downstream (regardless of the precise segmentation).

6.3.3.11. CUSIPs (upstream) – Other markets (downstream)

(A) The Parties’ activities

(635) As described above, S&P manages and operates the CUSIP system on behalf of the ABA.

(636) IHSM uses CUSIP data as an input for various downstream products including :

(a) Pricing and Reference data. CUSIP data is used as an input into certain of IHSM’s pricing and reference datasets, namely loan pricing data, loan reference data, bond pricing data and bond reference data.

(b) Equities and regulatory reporting. IHSM's dividend forecasting service provides independent estimates of the amount and timing of dividend payments, allowing customers to understand better how companies are performing and what their projected dividends are. Dividend estimates for global securities are based on equity research, market announcements and unique quantitative insight, covering (as far as IHSM’s service is concerned) over 28,000 stocks (for which CUSIP data is required to identify relevant stocks). In addition to this research-based methodology, IHSM also applies advanced analytics and predictive modelling to predict company dividends (which, again, may make use of CUSIP data).

(c) Managed corporate actions data.

(d) Issuer solutions. IHSM offers BD Corporate, a CRM investor relationship platform. CUSIPs are a component of the securities ownership data that IHSM distributes as part of BD Corporate.

(e) Investor and administration services. IHSM offers Profile Builder, a tool for producing buy side investor profiles, proving insight into investors by incorporating global ownership data, contacts and biographical content. CUSIPs are a component of the data that IHSM makes available as part of Profile Builder.

(f) Institutional holdings/investor data. IHSM uses CUSIP data as a component of the securities ownership data that it distributes as part of its offering.

(637) As S&P’s market share in CUSIPs upstream is 100%, all of these markets are vertically affected.

(638) Table 18 below shows the market shares of the Parties’ and their main competitors in other markets downstream of CUSIPs in 2020:

(639) The Notifying Party submits that the Transaction does not raise concerns in particular because:

(a) S&P would have no ability to foreclose access to CUSIP data to downstream rivals because: (i) Under the terms of its arrangement with the ABA, [summary of contractual arrangements between CGS and the rest of the S&P business]. And (ii) S&P licenses CUSIPs on a FRAND basis and would suffer reputational and commercial damages reneging on this commitment. Moreover, the Notifying Party submits that in this case the ABA would almost certainly remove S&P’s licence as the administrator of the CUSIP system, since the ABA has an interest in making CGS’s data as broadly available, accessible and adopted in the financial industry as possible; any commercial practice that may restrict access to those identifiers would undermine their status as market standards, with a negative impact on the value of the ABA’s intellectual property.

(b) S&P would also have no incentive to foreclose access to CUSIP data as, (i) There is already a vertical link with S&P’s pre-existing business in several markets, but CGS has never restricted access to CUSIPs to support S&P’s business or to encourage the entry / expansion of its own business. There is no reason why it should attempt to do so post-Transaction. Indeed, there is no plausible basis on which S&P would wish to do so given the potentially significant ramifications on its wider CUSIP business. And (ii) it would face retaliation risk from downstream competitors, in particular from companies like [names of S&P's suppliers], all of whom supply important inputs to S&P currently across a range of product areas.

(C) The Commission’s assessment

(640) The Commission’s investigation indicates that the Transaction is unlikely to give rise to input foreclosure.

(641) First, as explained above in paragraph (630) the Commission notes that the governance of the CUSIPs and the oversight by the ABA provide a certain comfort that S&P would not be able to favour its own downstream businesses to the detriment of rivals. The ABA confirms that they focus on this aspect and have not observed such behaviour by S&P in the past and S&P would be deterred from anti-competitive behaviour in breach of its understanding with the ABA.

(642) Second, and relatedly, the Commission notes that S&P has a public FRAND commitment, which, if effective, would deprive the company from the ability to engage in input foreclosure. While, for the purposes of this decision, the Commission is unable comment on the fair or excessive nature of CUSIP prices, S&P’s FRAND commitment could deter de facto total foreclosure strategies which would be more easily detectable, as they would amount to a refusal to supply.

(643) Third, the Commission notes that past behaviour speaks against ability and incentive for input foreclosure of CUSIPs; indeed, S&P is already active downstream of CUSIPs in several different markets, including equity indices, which are larger than the markets considered here. It would not be likely that the combined entity would have the ability or incentive post-Transaction to implement input foreclosure for smaller potential gains downstream.

(644) Fourth, and last, no respondent to the market investigation raised concerns in relation to the impact of the Transaction in any of these markets due to an input foreclosure of CUSIPs.

(D) Conclusion

(645) For the above reasons, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market in relation to the vertical relationships arising between CUSIPs (upstream) and pricing and reference data, equities and regulatory reporting, issuer solutions, managed corporate actions data, institutional holdings/investor data, and investor and administration services downstream (regardless of the precise segmentation).

6.3.3.12. Credit ratings (upstream) – Indices (downstream)

(646) Indices providers select securities that qualify for inclusion in their indices using pre-determined and publicly available eligibility rules or factors that are included in the index methodology.

(647) One common factor for fixed income and CDS indices methodologies is a financial instrument’s risk profile. Financial instruments, and in particular debt securities, can be considered “investment grade”, i.e. low risk but generating a lower return, or “high yield”, i.e. offering higher returns, but with more risks as they are linked to a higher probability of default. Credit ratings are a commonly accepted way to qualify instruments as investment grade or high yield. For instance, a security is typically considered “investment grade” if it is rated BBB- or higher by S&P or Fitch, and is considered “high yield” if it is rated BB+ or lower by S&P or Fitch.

(648) As a result, index providers need to license data from credit ratings agencies to build indices that consider the credit rating of the relevant security, and by extension the risk profile of the index. The input of credit ratings is not only relevant at the stage of initially creating an index, but is also needed to maintain and calculate the index, and updating constituents (i.e. the financial instruments included in the index) based on updates to credit ratings.

(A) The Parties’ activities

(649) S&P is active as a credit rating agency, issuing and distributing credit ratings and related data. S&P is also active in the downstream market for indices, in particular equity indices, but also fixed income indices to a more limited extent.

(b) S&P would also have no incentive to foreclose access to credit ratings data as (i) it would face retaliation risk, mainly from index providers such as [names of S&P's commercial partners] who are commercial partners of the Parties across multiple markets, (ii) [S&P's commercial strategy], and (iii) S&P is already active in the supply of financial indices, including fixed income indices, and never engaged in such foreclosure strategies.

(C) The Commission’s assessment

(657) The Commission’s investigation indicates that the Transaction is unlikely to give rise to input foreclosure.

(658) First, there are some limitations to the merged entity’s ability to engage in an input foreclosure strategy.

(659) S&P’s strong market presence in credit ratings, regardless of the precise market definition thereof indicates that the company may have a certain degree of market power upstream.

(660) It is true that S&P has a public FRAND commitment, which, if effective would deprive the company from the ability to engage in input foreclosure. SPGMI’s statement on its licensing policy reads that “SPGMI licenses S&P ratings on a fair, reasonable and non-discriminatory basis”. However, this commitment results from a voluntary decision of S&P which the company could formally decide to overturn one way or another, even if it led to some reputational cost, as mentioned by the Notifying Party. In addition, there is no formal mechanism set up for customers to enforce this commitment. The extent to which the FRAND commitment is effective is uncertain, in particular in relation to the fairness of prices for credit ratings. A number of respondents to the market investigation raised concerns in relation to price increases of credit ratings. Prices for credit ratings have indeed increased in recent years. While, for the purposes of this decision, the Commission is unable to comment on the fair or excessive nature of credit rating prices, this trend of continuous price increase may indicate that the FRAND commitment is not effective. This is further confirmed by ESMA’s monitoring of credit ratings, which questions fees charged by credit rating agencies where data licences are necessary to distribute credit ratings to subscribers, and recently called for the CRA Regulation to be amended to introduce further provisions to ensure that data licenses for credit ratings be granted on FRAND terms. S&P’s FRAND commitment could deter de facto total foreclosure strategies which would be more easily detectable, as they would amount to a refusal to supply, but is likely less effective in preventing partial foreclosure strategies.

(661) In addition, most suppliers of indices typically rely on at least two credit agencies, which limits the ability of S&P to engage in a foreclosure strategy. While among these, S&P is the prevalent one, followed by Moody’s, other credit rating agencies Fitch and sometimes DBRS are also mentioned as possible alternatives. The reason underpinning the choice of specific credit rating is primarily linked to the coverage offered by the agency.

(662) Second, the market investigation indicates that the merged entity would likely not have the incentive to engage in input foreclosure. A majority of respondents to the market investigation consider that it is not realistic that S&P only makes its credit ratings available (or available under better supply conditions, such as e.g. lower fees) to the merged entity’s (fixed income) indices. This strategy would likely lead to a loss of credibility of S&P and reputational damage. One customer notes for instance that “Technically they could, but […] that could create a tsunami on the market. It is critical that ratings are available to any index provider (especially Fixed-income)”, another one states that “This is not a realistic scenario for us, because it is also business case for S&P to offer ratings to other providers / products”. Another customer of financial indices notes that “S&P as a cornerstone, global provider will have to maintain impartiality for ratings provision”. An additional customer relatedly notes that “the reliability of Credit Ratings is also linked on how they can be embedded and used within different financial products (i.e. indices). Constraining their availability will then limit significantly their reliability from a market perspective”.

(663) This is particularly relevant in light of S&P’s revenue split, which relies heavily on credit ratings. The company generated around USD 4 billion from issuance of credit ratings (and sales of related data) in 2020, which represents more than half of the company’s revenues for the year.

(664) In addition, as mentioned S&P is already active in the downstream market, in particular in fixed-income indices. However, the market share of S&P on the market for fixed-income indices has been steadily below [0-5]% (in terms of index licensing for the creation of funds) and below [0-5]% and [information on S&P market shares] (in terms of index licensing in the form of market data). The addition of IHSM is unlikely to considerably impact these incentives, as the Parties’ market shares will remain below [10-20]%, far behind market leader Bloomberg.

(665) The lack of incentives to favour its own downstream business seems supported by the way S&P has priced credit ratings for its downstream index business in 2019 and 2020. Curves of prices indicate that price increases are applied relatively homogeneously to indices providers. [S&P's commercial strategy].

(666) In addition, the Non-Horizontal Merger Guidelines stipulate that “when the adoption of a specific course of conduct by the merged entity is an essential step in foreclosure, the Commission examines both the incentives to adopt such conduct and the factors liable to reduce, or even eliminate, those incentives, including the possibility that the conduct is unlawful. Conduct may be unlawful inter alia because of competition rules or sector-specific rules at the EU or national levels” (emphasis added). While the CRA Regulation primarily applies to SPGR rather than SPGMI. ESMA regularly reviews the industry’s practices, publishes topical reports on the level of fees by credit rating agencies, relays concerns expressed by customers, and makes (non-binding) recommendations accordingly (even if credit rating agencies challenge ESMA’s supervision of their ratings distribution). Besides the specific regulatory framework applicable to credit rating agencies, and in particular the scrutiny of ESMA, any strategy aimed at foreclosing competitors in a downstream activity may give rise to additional public enforcement, in particular from competition authorities. Based on the Parties’ submission and the results of the market investigation, S&P likely has a dominant position on the markets for credit ratings (even more so if ratings of S&P would represent a standalone market), and discriminatory pricing practices would open the company to enforcement actions. The extent to which S&P is concerned about regulatory (including antitrust) intervention in relation to its credit ratings activity is further confirmed in internal documents of the company.

(667) Third, a foreclosure strategy would likely not have a material impact on competing indices providers, due to the barriers to switching index provider. Switching financial index provider is indeed not possible for customers in the short to medium term. This is largely due to end customer (i.e. the investor) requirements and/or the need for regulatory approvals, and applies to both fixed income and equity indices, according to a large majority of respondents. One customer indicates for instance that “Many fixed income indices are considered as benchmarks meaning that have strong brand recognition, reputation and high visibility on the market. As such a fixed income benchmark with specific terms (currency, credit risk, liquidity, type...) cannot be substituted with another fixed income [index]”. Another customer indicates that “Any change would be subject to investor consent and/or regulator approval”. A third customer confirms that “In many cases benchmark changes require regulatory approval which takes time. Differences in calculation methodology also present challenges”.

(668) Lastly, the extent to which any potential foreclosure theory would be merger-specific is questionable. S&P already offers financial indices including (to a small extent) fixed income indices, and IHSM’s increment is overall limited. In addition, in its

(669) 496 Non-Horizontal Merger Guidelines, paragraph 46. 497 Article 6(2) of the CRA Regulation provides that “[…] a credit rating agency shall comply with the requirements set out in Sections A and B of Annex I”, whereas Section B of Annex I (point 3c) provides that “[a] credit rating agency shall ensure that fees charged to its clients for the provision of credit rating and ancillary services are not discriminatory and are based on actual costs”.

(670) 498 ESMA Thematic report on fees charged by Credit Rating Agencies and Trade Repositories , 11 January 2018, p. 9. Also see ESMA Opinion on improving access to and use of credit ratings in the European Union, 22 September 2021, p. 5. The three largest credit rating agencies, including S&P, dispute that ESMA has the ability to supervise activities not conducted by the legal entity issuing credit ratings.

(671) 499 See for instance commercial training materials, submitted as Form CO, Chapter on Vertical Relationships, Annex D.8a.

(672) 500 Replies to questions 9 and 25 of Questionnaire Q5.

(673) 501 It is irrelevant for these purposes whether CDS indices would form part of the wider fixed income indices market, as IHSM is the only provider of CDS indices globally.

licensing agreements with index providers, [S&P's contractual arrangements with its 502customers].

(669) The Commission’s investigation indicates that the Transaction is also unlikely to give rise to customer foreclosure. As credit ratings are an input into a multitude of downstream products (in addition to being sold standalone to end customers), and S&P also licenses credit ratings data from other credit rating agencies, such as Moody’s and Fitch, customer foreclosure concerns are unlikely to arise as a result of the Transaction.

(D) Conclusion

(670) For these reasons, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market in relation to the vertical relationships arising between credit ratings (upstream) and indices (downstream).

6.3.3.13. Credit ratings (upstream) – Other markets (downstream)

(A) The Parties’ activities

(671) S&P is active as a credit rating agency, issuing and distributing credit ratings and related data.

(672) IHSM uses credit ratings data as an input for various downstream products including:

(a) Pricing and reference data. IHSM uses credit ratings to create sector curves (i.e. a forecast view of pricing for the relevant instrument type), which form part of IHSM’s loan pricing and reference data, and in particular its pricing data products (including CDS pricing data and bond pricing data).

(b) Issuer solutions. IHSM offers BD Corporate, a CRM investor relationship platform. Credit ratings are a component of the securities ownership data that IHSM distributes as part of BD Corporate.

(c) Economic data. IHSM uses credit ratings data alongside many other inputs to inform IHSM’s economic analysis and forecasts.

(d) Investor event management solutions. IHSM offers BD Corporate, which also allows users to retrieve the profile of a specific security including characteristics such as market capitalisation, industry, location, and many other data points including potentially credit ratings.

(e) Investor and administration services. IHSM offers Profile Builder, a tool for producing buy side investor profiles, proving insight into investors by incorporating global ownership data, contacts and biographical content. Credit ratings are a component of the data that IHSM makes available as part of Profile Builder.

502 Form CO, Chapter on Vertical Relationships, Annex D.8, footnote 7.

(678) Third, and lastly, no respondent to the market investigation raised concerns in relation to the impact of the Transaction in any of these markets.

(679) The Commission’s investigation indicates that the Transaction is also unlikely to give rise to customer foreclosure. As credit ratings are an input into a multitude of downstream products (in addition to being sold standalone to end customers), and S&P also licenses credit ratings data from other credit rating agencies, such as Moody’s and Fitch, customer foreclosure concerns are unlikely to arise as a result of the Transaction.

(D) Conclusion

(680) For these reasons, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market in relation to the vertical relationships arising between credit ratings (upstream) including all plausible segments thereof based on either credit rating type or the relevant agency(ies) and (i) pricing and reference data (regardless of whether pricing and reference data are part of the same or separate markets for the different asset classes of loans, CDS and bonds), (ii) issuer solutions, (iii) economic data (regardless of the segmentation between historic or forecast economic data), (iv) investor event management solutions, (v) investor and administration services, and (vi) institutional holdings/investor data (all downstream).

6.3.3.14. Equity indices (upstream) – Portfolio valuation tools, Global securities financing data, Economic data (downstream)

(A) The Parties’ activities

(681) As mentioned above, S&P is active upstream in equity indices, with market shares varying based on the precise segment being considered (from [30-40]% in overall equity indices up to 90% and above in certain narrow regional segments such as Australia and New Zealand equity indices). Downstream IHSM is active in portfolio valuation tools with a market share of [50-60]%, global securities financing data with a market share of [30-40]% and economic data ([10-20]% in overall economic data, [10-20]% in forecast economic data and [5-10]% in time series economic data).

(B) The Notifying Party’s view

(682) The Notifying Party submits that it would not have the ability or incentive to foreclose competing downstream competitors in any of the mentioned markets. First,

512 Credit ratings represent respectively, the following share of total costs for the relevant downstream products offered by IHSM: […]% for CDS pricing data, […]% for loan pricing data, […]% for bond pricing data, […]% for (forecast) economic data, […]% for issuer solutions, […]% for investor event management solutions, […]% for administration services, , and […]% for institutional holdings and investor data. Form CO, Chapter on Vertical Relationships, Annex D.2.

513 For example, the large majority of competitors in loan, CDS and bond pricing data stated that they were not concerned with respect to access to S&P’s credit ratings as an input post-Transaction. See replies to question 42 of Questionnaire 6.

144

(A) The Notifying Party’s view

(688) As mentioned above, IHSM is active upstream in fixed income indices, with market shares varying based on the precise segment being considered (from [5-10]% in overall fixed income indices up to [90-100]% and above in certain narrow regional segments such as Asian Fixed Income). Downstream S&P is active in multi-asset indices, with a market share of [20-30]%. Different indices, including fixed income indices, can be used as an input to create multi-asset indices. [contractual arrangements and the Parties' business strategy].

(B) The Notifying Party’s view

(689) The Notifying Party submits that it would not have the ability to foreclose downstream multi-asset indices competitors, because IHSM has no market power, with strong competing suppliers, including Bloomberg and ICE. The Notifying Party further argues that IHSM cannot be an important supplier of fixed income indices as an input into multi-asset indices since [IHSM's contractual arrangements and relations with customers].

(C) The Commission’s assessment

(690) Based on the evidence available to it, the Commission finds that the Notifying Party would have no ability to foreclose downstream rivals by withholding access to IHSM’s fixed income indices. The fact that [IHSM's contractual arrangements and relations with customers], suggest that IHSM does not have market power for the upstream input. MSCI, FTSE Russell and Bloomberg are seen as stronger multi-asset providers than S&P. While multi-asset indices are typically composed of equity indices and fixed income indices, responding competitors do not consider IHSM’s fixed income indices as a must-have. Indeed, although competitors indicate that the combed entity may be in a unique position to offer strong multi-asset indices, the majority do not expect a negative impact on the market for multi-asset indices.

(691) Similarly, the Commission finds that the Notifying Party would have no ability to foreclose upstream rivals by refusing to purchase from them, as fixed income indices have multiple use cases, of which multi-asset indices comprise a very small portion (<1%), and S&P’s market share downstream, [20-30]%, would not indicate sufficient market power to foreclose.

(D) Conclusion

(692) For the above reasons, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market in relation to the vertical relationships arising between fixed income indices (upstream) and multi-asset indices (downstream).

6.3.3.16. Indices (upstream) – Desktop services, Non-real time datafeeds (downstream)

(A) The Parties’ activities

(693) Both Parties are active upstream in various segments of index licensing. IHSM is active in fixed income indices, with market shares varying (from [5-10]% in overall fixed income indices up to [90-100]% and above in certain narrow regional segments such as Asian Fixed Income), and in CDS indices with a market share of [90-100]%. Downstream S&P is active in desktop services, with a market share of [10-20]% and in NRTDs, with a market share of [5-10]%.

(694) Both Parties supply their indices both directly to customers and via third-party distributors’ desktop services and non-real time datafeeds. For instance, S&P distributes its indices through a range of third party distributors, including major competitors such as Bloomberg, LSE/Refinitiv and FactSet. IHSM also distributes its own indices via a number of third parties including Bloomberg, Refinitiv, Deutsche Bourse and MSCI.

(695) S&P also distributes third party indices through its own desktop services and NRTDs.

(B) The Notifying Party’s view

(696) The Notifying Party submits that it would not have the ability or incentive to foreclose competing downstream competitors. Regarding ability, first, the Notifying Party argues that S&P already has a significant equity indices franchise, which is much stronger and better-known than IHSM’s fixed income franchise. The threat of withholding the S&P 500 is therefore far greater than withholding IHSM’s iBoxx indices, and S&P does not engage in this pre-Transaction. Second, the Notifying Party submits that there are many alternative indices suppliers, such that the combined entity has no market power in the supply of indices. Regarding the few segments where the combined entity has higher market shares, the Notifying Party argues that it would not be able to foreclose downstream rivals on the basis of these minor markets. Third, the Notifying Party notes that SPDJI, [information on the Parties' current and future business strategy].

(697) Regarding incentives, the Notifying Party first explains that both Parties rely on third party distributors of index data today (including competing index providers) such that any foreclosure strategy would be cutting off a key route to market. S&P derives a significant proportion of its data revenue ([information on the Parties' commercial strategy and revenues]) via third party distribution channels. Second, restricting

distribution of indices could risk the status of (even) SPDJI’s “headline” indices such as the S&P 500 because the wide availability of index data across a range of distributors is important to maintain the relevance and use of SPDJI’s indices. Third, any restriction on the distribution of indices is likely to result in limited recapture by SPGMI. This is because there is significant differentiation between downstream products, particularly desktop services.

(C) The Commission’s assessment

(698) Based on the evidence available to it, the Commission finds that the Notifying Party would have no ability or incentive to foreclose downstream rivals by withholding access to IHSM’s fixed income or CDS indices. First, currently, almost half of responding index customers receive IHSM’s indices only through a direct feed, while the rest use both a direct feed and a third party data vendor. Indeed, multi-sourcing of desktop services is already prevalent today and of the responding customers, few would switch to or add a desktop or datafeed from S&P in case of an input foreclosure; one customer explains that their desktop is “used for numerous reasons other than [for these indices]”; most would not be affected due to multi-sourcing or would take no switching/adding action in case of foreclosure.

(699) Second, the fact that S&P currently supplies its equity indices to downstream competitors, and that both Parties earn significant revenues via third party distributors, indicate a lack of incentive for input foreclosure. Indeed, downstream competitors responded that it would not be realistic for this foreclosure to occur; one close competitor summarized it as “commercially disadvantageous”.

(700) Similarly, the Commission finds that the Notifying Party would have no ability or incentive to foreclose upstream rivals by refusing to purchase from them, as fixed income indices have multiple use cases and distribution channels (including direct distribution by index providers) and S&P’s market share downstream is not high enough to suggest the ability to significantly harm upstream rivals. Moreover, despite being active upstream in the supply of equity indices, S&P currently distributes third party indices on its downstream products. Lastly, no index provider competitors have expressed concerns about possible customer foreclosure of S&P’s desktop services and NRTDs.

(D) Conclusion

(701) For the above reasons, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market in relation to the

(705) The Notifying Party further submits that it would have no incentive to foreclose. First, in the counterfactual each party already supplies calculation and administration services in respect of both equity and fixed income indices, notwithstanding that S&P has a more material position in the supply of equity indices and IHSM in fixed income indices. S&P largely supplies calculation and administration services for equity indices ([S&P revenue information]% of revenues), in relation to which IHSM has virtually no presence. The Transaction will have very limited impact on the Parties’ combined position in these downstream markets. Second, entities supplying indices that also require calculation and administration services are typically those wanting to create their own proprietary indices generally for use in-house, carrying their own brand and / or using their own intellectual property. Such customers include e.g. funds, banks and other segments that also make up the Parties’ typical customer base for the supply of their indices i.e. they are not in direct competition with the Parties’ index licensing activities. S&P would therefore have little incentive to foreclose such entities (who are also customers of the downstream market).

(C) The Commission’s assessment

(706) Based on the evidence available to it, the Commission finds that the Notifying Party would have no ability or incentive to foreclose downstream rivals by withholding access to IHSM’s fixed income indices. First, the Parties would not have the ability to foreclose downstream competitors, as the main competitors downstream in each of equity indices and fixed income indices themselves provide index calculation and administration services to third parties, indicating that they have the capabilities to perform these activities and would not be significantly harmed if the Parties withheld their services. Indeed, the types of customers to whom the Parties provide such services are self-indexing funds and banks, and thus are not close competitors of the Parties downstream.

(707) Second, the fact that both Parties currently supply index calculation and administration services despite their market presences downstream, and in particular S&P mostly supplies equity index calculation and administration services, indicate a lack of incentive for input foreclosure. Indeed, responding index competitors and customers did not express concerns about a possible input foreclosure of these services.

(708) Similarly, the Commission finds that the Notifying Party would have no ability to foreclose upstream rivals by refusing to purchase from them, as neither currently purchases any index calculation and administration services from third parties because they use in-house capabilities.

(D) Conclusion

(709) For the above reasons, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market in relation to the vertical relationships arising between index calculation and administration services (upstream) and equity indices and fixed income indices (downstream).

(710) IHSM is active upstream in the supply of managed corporate actions data with a market share of [10-20]%. S&P uses this data as an input into its equity indices downstream. S&P’s market share downstream varies depending on the segment considered, from [30-40]% for overall equity indices up to [90-100]% and above in certain narrow regional segments such as Australia and New Zealand equity indices. Corporate actions are events that affect the securities of a given company, such as dividends, stock splits, M&A, disposals or spin-offs. For the purposes of this particular vertical relationship, managed corporate actions data are therefore required to inform the constituents in baskets of equities making up a given index.

(B) The Notifying Party’s view

(711) The Notifying Party submits that it would not have the ability to foreclose downstream equity indices competitors, because IHSM has no market power upstream with its low market share of [10-20]% and with strong competing suppliers, including Bloomberg and ICE. The Notifying Party further argues that IHSM already has a vertical relationship with its own (minimal) equity indices portfolio, yet has never attempted to foreclose access to competing indices suppliers.

(C) The Commission’s assessment

(712) Based on the evidence available to it, the Commission finds that the Notifying Party would have no ability or incentive to foreclose downstream rivals by withholding access to IHSM’s managed corporate actions data. IHSM’s low market share upstream and the fact that there are several competing suppliers (two of which have higher market shares than IHSM; Bloomberg having [20-30]% and ICE [10-20]%) indicate a lack of market power for the upstream input. Moreover, S&P’s main competitors downstream did not express concerns regarding a possible input foreclosure of IHSM’s managed corporate actions data.

(713) Similarly, the Commission finds that the Notifying Party would have no ability to foreclose upstream rivals by refusing to purchase from them, as managed corporate actions data has multiple use cases, of which equity indices comprise a very small portion (1%). Moreover, S&P’s market share downstream in equity indices, [30-40]%, would not indicate sufficient market power to foreclose; its higher market shares in narrower segments would not appear relevant as all equity indices together comprise 1% of demand for managed corporate actions data, and the narrower segments themselves comprise a small portion of equity indices. Lastly, competitors in the upstream market did not raise customer foreclosure concerns regarding this vertical overlap.

(D) Conclusion

(714) For the above reasons, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market in relation to the vertical relationships arising between managed corporate actions data (upstream) and equity indices (downstream).

(715) As mentioned above, IHSM is active upstream in economic data ([10-20]% in overall economic data, [10-20]% in forecast economic data and [5-10]% in time series economic data). Downstream S&P is active in credit ratings, with a market share by revenue of [30-40]%. Credit rating agencies such as S&P may use economic data as one of many inputs to inform their ratings decisions and for general research purposes.

(B) The Notifying Party’s view

(716) The Notifying Party submits that it would not have the ability to foreclose competing downstream credit ratings competitors, because IHSM has a modest upstream share and therefore no market power, with many competing suppliers of economic data, including Moody’s, Refinitiv, Fitch, FactSet, Haver Analytics, Oxford Economics and the Economist Intelligence Unit. S&P currently multi-sources its economic data requirements from a number of suppliers (indeed, purchases of economic data from [information on S&P's supply source]).

(C) The Commission’s assessment

(717) Based on the evidence available to it, the Commission finds that the Notifying Party would have no ability to foreclose downstream rivals by withholding access to IHSM’s economic data. IHSM’s low market share and the fact that there are several competing suppliers indicate a lack of market power for the upstream input. Moreover, S&P’s main competitors downstream indicated that they are not concerned by a possible input foreclosure.

(718) Similarly, the Commission finds that the Notifying Party would have no ability to foreclose upstream rivals by refusing to purchase from them, as economic data has multiple use cases, of which credit ratings comprise a very small portion (<1%) and S&P’s market share downstream, [30-40]%, would not indicate sufficient market power to foreclose. Moreover, competitors in the upstream market did not raise customer foreclosure concerns regarding this vertical overlap.

(D) Conclusion

(719) For the above reasons, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market in relation to the vertical relationships arising between economic data (upstream) and credit ratings (downstream).

(720) As mentioned above, S&P is active upstream with the product GICS. The Notifying Party is not able to provide estimated market shares for GICS or its competitors, but it does consider that GICS is likely to be market leader (as supplied by both S&P and MSCI), with its closest competitors being FTSE Russell’s ICB, Bloomberg’s BICS, FactSet’s RBICS and Thomson Reuters TRBC (in descending order).

(721) Downstream, both Parties are active in trade analytics (S&P [5-10]%, IHSM [0-5]%, combined [10-20]%) and stock selection and strategy services (S&P [0-5]%, IHSM [0-5]%, combined [5-10]%). Only IHSM is active in economic data ([10-20]% in overall economic data, [10-20]% in forecast economic data and [5-10]% in time series economic data) and digital design for financial services ([10-20]% market share).

(B) The Notifying Party’s view

(722) The Notifying Party submits that irrespective of the relevant downstream market, if S&P attempted to foreclose access to GICS, then MSCI would step in to fill the gap created by S&P’s attempted foreclosure, since MSCI has no interest in assisting S&P to favour its own downstream operations (and it is not restricted in terms of its ability offer GICS to customers, including current customers of S&P). Thus, S&P has no market power in the supply of GICS since it always faces MSCI as an independent supplier, as well as competition from other alternative classification systems. [information on IHSM' suppliers] – any downstream competitor of S&P post-Transaction (as pre-Transaction) could do the same. In addition, the Notifying Party argues that GICS is one of many inputs, and not a particularly important one, into each of the downstream markets.

(C) The Commission’s assessment

(723) Based on the evidence available to it, the Commission finds that the Notifying Party would have no ability or incentive to foreclose downstream rivals by withholding access to GICS. First, MSCI confirmed that it does supply GICS to users, including third party data vendors who compete with S&P in various markets. MSCI does not expect the Transaction to change its relationship with the Parties, including the arrangement governing GICS. As such, if S&P were to deny or worsen access to GICS, MSCI would be able to supply GICS to the foreclosed customers, thus rendering the foreclosure strategy ineffective. Second, while GICS may be the market leader in sector classification schemes, there are several other schemes available in the market from large data vendors with wide customer bases, namely ICB of FTSE Russell, TRBC of Refinitiv, and Factset’s RBICS. Therefore, customers foreclosed from GICS would also have alternative schemes to switch to, if they did not wish to obtain GICS from MSCI due to the revenue sharing agreement. Lastly, S&P is already active (with the larger presence) in two of the downstream markets and has not restricted competitors’ access to GICS, indicating a lack of incentive for such a strategy. IHSM’s increment in those two markets, and market leading provider with a share estimated to be over [70-80]% in fixed income indices (for the creation of funds). Other competitors with a presence across asset categories include Nasdaq, MSCI, and Solactive, as confirmed by the Parties’ internal documents.

(731) Second, it is unlikely that the Parties would have an incentive to engage in such a strategy. Pre-Transaction, S&P already has a share exceeding 30% in equity indices, and a presence in fixed income indices, however S&P did not leverage its position in equity indices to increase its presence in fixed income. Similarly, IHSM holds a 100% market share in relation to CDS indices and have not leveraged this position to increase its holding in equity indices (or in fixed income indices), where its share remained stable and limited.

(732) Third and lastly, it is unlikely that such strategy would have an impact on prices and choice. Only few respondent to the market investigation specifically raised concerns in relation to a possible foreclosure strategy in the markets for indices as a result of tying or bundling of different indices by the combined entity, but these respondents do not indicate how the Transaction would specifically impact pre-existing ability and incentives of the Parties. However, in particular due to the barriers to switching referred to above, it is unlikely that this theory of harm would materialise.

(D) Conclusion

(733) For these reasons, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market in relation to the conglomerate relationships arising between different categories of indices.

6.3.4.2. Conglomerate effects – issuance platforms and desktop services

(A) The Parties’ activities

(734) As mentioned above, S&P is active in desktop services with its product CapIQ. Its market share in 2019 is estimated at [10-20]%. IHSM is active in issuance platforms with its product Ipreo. Its market share is estimated at [50-60]% for corporate bond issuance platforms, [50-60]% for municipal bond issuance platforms and [30-40]% for equities issuance platforms. The Commission investigated possible conglomerate effects arising from the combination of CapIQ with Ipreo.

(B) The Notifying Party’s view

(735) First, the Notifying Party argues that it is not plausible that leveraging could exclude rivals in desktop services. There are examples of desktops that are currently competing successfully at small scale. In desktops, Moody’s and ICE both compete successfully with shares of around [0-5]% globally. The Parties consider that there are a large number of other desktop suppliers competing at even smaller scales, including SIX Financial, EMIS, CEIC, Dow Jones, Quick Financial, Alphasense, CoStar, RCA, Reonomy, Verisk Analytics and BamSEC. Moreover, the larger rivals could also retaliate by withdrawing their own data and information products from the S&P platforms, and use their own products to match any tying or bundling strategies by the Parties.

(736) Second, according to the Parties, IHSM’s customer base for issuance workflow solutions in particular consists to a large degree of large, sophisticated customers. For each type of issuance workflow solutions, the top ten customers are nearly always comprised of the largest investment banks, including: [Banking customers]. These customers are therefore likely to be capable of resisting any leveraging strategies by the Parties, and also capable of retaliating against the Parties in this or other markets in response to any leveraging strategies including sponsoring entry.

(737) Third, the Notifying Party claims that the theory of harm is inconsistent with pre-merger outcomes as today IHSM does not engage in pure bundling, mixed bundling, contractual tying or technical tying of issuance workflow solutions with other products, despite already having some potentially related products outside of issuance workflow solutions, and S&P currently distributes a large range of its own data and information products through its desktops. For example, S&P’s credit ratings products are distributed through several third parties including major aggregated desktop suppliers such as Bloomberg, FactSet, Refinitiv and ICE. This is inconsistent with the view that S&P is using its current portfolio of data and information products to leverage into distribution platforms. Furthermore, in some cases S&P pays a fee to third parties to distribute certain S&P products. [S&P's commercial strategy]. This makes business sense for SPGMI because SPGMI expects to derive revenues from sales of product licenses that are likely to result from these arrangements. There is no reason to believe that this will change post-Transaction.

(C) The Commission’s assessment

(738) Based on the evidence available to it, the Commission finds that the Notifying Party would have no ability or incentive to foreclose desktop services rivals by leveraging the issuance platforms to benefit desktop services, i.e. bundling or tying CapIQ with Ipreo.

(739) First, the Commission notes that multi-sourcing of desktops is already prevalent, confirming some differentiation among different suppliers; indeed, all responding customers multi-source desktops and those using S&P often use alongside it at least two other suppliers. In addition, no customer said that they would switch away from their current desktop provider to S&P’s CapIQ in case of a bundling of the latter with Ipreo. This indicates that the combined entity would not be able to foreclose rivals in this market via such a strategy, since no customers would stop using rivals’ products as a result.

(740) Second, the Commission considers that the target customer group, i.e. the large investment banks using both desktop services and issuance platforms, would have a degree of buyer power to resist anticompetitive bundling, and the ability to sponsor entry/expansion in the relevant markets. For instance, Directbooks is a new entrant into the issuance platforms space, created in 2019 by a consortium of nine investment banks ([…]) “to optimize the communications process and workflow for the primary issuance of securities” and “to alleviate the pain points and inefficiencies in the primary market process felt by both underwriters and investors”. The founders expect it to grow over time, “While DirectBooks will initially focus on Investment Grade transactions, its functionality should make it a platform of choice for other asset classes in the future.”

(741) Given the importance of buy-in by investment banks and the investors they serve in the usage of an issuance platform, it does not seem plausible that the combined entity would have the incentive to risk negative reactions from this customer group. Indeed, competitors indicate that customers would react negatively to forced bundling or tying of these two products; one competitor explains, “Customers would be weary of discounting tactics early on post merger. Customers would fear of short-term discounts with longer-term significant price increases/rightsizing of costs… Customers want options and the ability to purchase the best products in the market.”

(742) Third, while responding competitors and customers consider that the Transaction will have a negative impact on the market for desktop services globally, the concerns are often not related to the possibility of bundling with issuance platforms, and are not borne out by the current and expected customer behaviour described above.

(D) Conclusion

(743) For the above reasons, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market in relation to conglomerate effects arising between issuance platforms and desktop solutions.

549 Replies to question 6 of Questionnaire 9.

550 Replies to questions 25-26 of Questionnaire 9.

551 https://www.directbooks.com/history

552 https://www.directbooks.com/history

553 Replies to question 30 of Questionnaire 8.

554 Replies to questions 33 of Questionnaire 8 and 16 of Questionnaire 9.

ratings by preventing rival loan administration services from populating credit ratings data for customers with the requisite licence. The Notifying Party submits that it would make no commercial sense for S&P to undermine the commercial FRAND policy post-Transaction, because doing so would risk damaging both S&P’s wider reputation (both as a supplier of credit ratings data and more generally) and the value of its credit ratings business.

(749) Third, the Notifying Party claims that the use of ratings data in WSO is fundamentally contingent on S&P (and WSO’s) common customers having the requisite licence for the use of that data. In any hypothetical attempt to foreclose WSO’s competitors, S&P would also be denying these customers – many of whom will be large and sophisticated – the ability to use S&P’s data in the way(s) that they wish to do. In particular, those significant customers are likely to be active across the loan lifecycle and the significant harm to the customer relationship and S&P reputation is likely to result in loss of credibility for S&P Global Ratings and a loss of credit ratings issuance revenue.

(C) The Commission’s assessment

(750) Based on the evidence available to it, the Commission finds that the Notifying Party would have no ability or incentive to foreclose rivals by leveraging WSO to benefit S&P’s credit ratings, or vice versa by bundling the products or technically tying them.

(751) First, regarding leveraging credit ratings data to favour WSO, the Commission notes that for the reasons explained in Section 6.3.3.12, S&P would be discouraged from favouring WSO in the provision of credit ratings data or would invite regulatory scrutiny in case it engaged in such behaviour. Loan administration solution competitors are divided on whether credit ratings are a key feature of their product, or not. Some do, saying it would be “very problematic” for customers if S&P no longer made its credit ratings data available to them in a seamless/automatic way post-Transaction, but they do not consider this realistic behaviour on the part of the combined entity. Others either do not carry credit ratings or do not consider them important for their products.

(752) In addition, there is evidence that multi-sourcing is common in loan administration solutions. Suppliers comment; “Yes, oftentimes clients will use multiple products and services to cover their data, reporting, and portfolio management needs. This can be broken out by front and middle office software or by specific business function. Clients do purchase from more than one supplier in this market”, “Customers can choose to buy software from several suppliers and services from several as well” and “Our view would be that customers do typically multi-source this product/service”. Indeed, no customer indicated that they would stop using WSO competitors’ products if S&P ratings were not available with those products.

(753) Second, regarding leveraging WSO to favour S&P’s credit ratings, the Commission first notes that the market shares (<30%) submitted by the Notifying Party do not appear to fully reflect WSO’s market position. Indeed, in internal documents, WSO is referred to by IHSM as the market leader, with an estimated market share of [50-60]% in each of traded loan software and traded loan solutions. Two competitors are of the same view; one states “[IHSM is the ]Top loan data and loan portfolio management software. Largest number of customers” and another adds, “in the broader asset management market IHSM's WSO solution is generally considered to be the market leader”.

(754) Nevertheless, there is evidence that customers often purchase credit ratings data (and other relevant loan data) separately and independently from loan administration solutions, and thus bundling would not be likely to have a significant effect; “With respect to data, the customer would likely utilize multiple data vendors including multiple rating agencies, one or more loan pricing provider, and, potentially, market/asset data provider.” Moreover, given the prevalence of multi-sourcing, the Commission would expect there to be no significant impact on competing credit ratings suppliers, as there would be other loan administration solutions, which could integrate their ratings in a similar fashion to WSO and S&P’s credit ratings.

(755) Third, responding competitors consider that the Transaction will have a neutral, or positive (“assuming greater innovation and data quality”), impact on the market for loan administration solutions. Similarly, responding competitors do not expect the Transaction to have a negative impact on the market for credit ratings due to this overlap. There were similarly no substantiated concerns from customers regarding this overlap.

(D) Conclusion

(756) For the above reasons, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market in relation to conglomerate effects arising between credit ratings and loan administration solutions.

6.3.4.4. Conglomerate effects – credit ratings and issuance platforms

(A) The Parties’ activities

(757) As mentioned above, S&P is active in credit ratings issuance and distribution. Its market share in 2019 is estimated at [30-40]% at the global level and [40-50]% at the EU level.

(758) IHSM is active in issuance platforms with its product formerly called Ipreo. Credit ratings can only be used with corporate bond issuance platform, where IHSM’s market share is estimated at [50-60]%, or with municipal bond issuance platforms where IHSM’s market share is estimated at [30-40]% .

560 IHSM internal document No. ASH000200, slide 8.

561 Reply to “Request for information on Traded loan software/services” email on 3 September 2021.

562 Replies to “Request for information on Traded loan software/services” email on 3 September 2021.

563 Minutes of calls with competitors on 9 June 2021, 1 July 2021, 19 August 2021.

564 Replies to questions 45, 47 of Questionnaire 9.

(759) Various data relevant to the issuer and issuance are included in these platforms by the relevant banks working on the issuance. These inputs include credit ratings data. However, IHSM has no involvement in procuring these inputs. The Commission therefore investigated possible conglomerate effects arising from the combination of S&P’s credit ratings with Ipreo.

(B) The Notifying Party’s view

(760) First, the Notifying Party submits that post-Transaction no leveraging from credit ratings into IHSM products can plausibly arise for a number of reasons. Firstly, S&P does not have the ability to leverage from credit ratings issuance or distribution as it faces several credible rivals; the Transaction will not increase the Parties’ market power; such theory of harm is inconsistent with pre-merger outcomes; credit ratings and issuance platforms are not in practice purchased together; and customers are strategic and well-informed. Secondly, the Parties would have no incentive to foreclose rivals through leveraging from credit ratings as any attempt could prompt regulatory scrutiny; and because undermining S&P’s FRAND principles would damage S&P’s reputation.

(761) Second, the Notifying Party submits that post-Transaction no leveraging from IHSM’s issuance platforms into S&P products can possibly arise for a number of reasons. Firstly, IHSM does not have the ability to leverage from issuance platforms as the Transaction will not increase the Parties’ market power; such theory of harm is inconsistent with pre-merger outcomes; credit ratings and issuance platforms are not in practice purchased together or in fixed proportions; it is not possible that leveraging could exclude rivals in credit ratings given that there are other uses of credit ratings; such attempt could prompt regulatory scrutiny; and customers are strategic and well-informed. Secondly, the Parties would have no incentive to foreclose rivals through leveraging from issuance platforms because foreclosure would likely be ineffective, which is demonstrated by the lack of current tying or bundling practices.

(C) The Commission’s assessment

(762) Based on the evidence available to it, the Commission finds that the Notifying Party would have no ability or incentive to foreclose rivals by leveraging credit ratings to benefit IHSM’s Ipreo, or vice versa by bundling the products or technically tying them.

(763) First, regarding leveraging credit ratings data to favour Ipreo, the Commission considers that the merger entity would not have the ability to foreclose rivals in issuance platforms by leveraging its position in credit ratings.

(764) The Commission notes that for the reasons explained in Section 6.3.3.12, S&P would be discouraged from favouring Ipreo in the provision of credit ratings data or would invite regulatory scrutiny in case it engaged in such behaviour.

565 Form CO, Chapter on Conglomerate effects, paragraphs 4.46 ff.

566 Form CO, Chapter on Conglomerate effects, paragraphs 4.126 ff.

567 Replies to question 23 of Questionnaire 9.

568 Replies to question 23 of Questionnaire 9.

569 Replies to question 19 of Questionnaire 9.

570 Minutes of a call with a competitor on 7 June 2021.

571 Reply to question 28 of Questionnaire 9.

572 Replies to question 37 of Questionnaire 9.

(D) Conclusion

(773) For the above reasons, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market in relation to conglomerate effects arising between credit ratings and issuance platforms.

6.4. Commitments

6.4.1. Framework for the assessment of the Commitments

(774) Where a concentration raises serious doubts as regards its compatibility with the internal market, the parties may undertake to modify the concentration to remove the grounds for the serious doubts identified by the Commission. Pursuant to Article 6(2) of the Merger Regulation, where the Commission finds that, following modification by the undertakings concerned, a notified concentration no longer raises serious doubts, it shall declare the concentration compatible with the internal market pursuant to Article 6(1)(b) of the Merger Regulation.

573 (775) As set out in the Commission’s Remedies Notice, the commitments have to eliminate the competition concerns entirely, and have to be comprehensive and effective from all points of view.

(776) In assessing whether commitments will maintain effective competition, the Commission considers all relevant factors, including the type, scale and scope of the proposed commitments, with reference to the structure and particular characteristics of the market in which the transaction is likely to significantly impede effective competition, including the position of the parties and other participants on the market.

(777) In order for the commitments to comply with those principles, they must be capable of being implemented effectively within a short period of time. Concerning the form of acceptable commitments, the Merger Regulation gives discretion to the Commission as long as the commitments meet the requisite standard. Structural commitments will meet the conditions set out above only in so far as the Commission is able to conclude with the requisite degree of certainty, at the time of its decision, that it will be possible to implement them, and that it will be likely that the new commercial structures resulting from them will be sufficiently workable and lasting to ensure that the serious doubts are removed. Divestiture commitments are normally the best way to eliminate competition concerns resulting both from horizontal and non-horizontal overlaps.

6.4.2. Proposed Commitments

(778) In order to render the concentration compatible with the internal market, the undertakings concerned have modified the notified concentration by entering into the following commitments in relation to the concerns identified with respect to

573 Commission Notice on remedies acceptable under Council Regulation (EC) No 139/2004 and under Commission Regulation (EC) No 802/2004, the “Remedies Notice” (OJ C 267, 22.10.2008, p. 1-27).

574 Remedies Notice, paragraph 9.

575 Remedies Notice, paragraph 12.

576 Remedies Notice, paragraph 10.

6.4.3. The CUSIP Commitments

(A) Initial Commitments

(779) In order to eliminate the Commission’s concerns relating to the horizontal overlap between LXIDs and loan CUSIPs in the global loan identifier market, the Notifying Party offered to divest the Loan CUSIP Business (“the Loan CUSIP Commitments”). Considering the nature of this option, which consisted in a carve-out, and the limited size of the Loan CUSIP divestment business, the Notifying Party submitted alternative Commitments in case the Loan CUSIP Commitments remedy was not acceptable following a market test. These alternative Commitments consist in the divestment of the whole CUSIP Business (“the Initial CUSIP Commitments”).

(780) The Loan CUSIP Commitments include a transfer of the rights to issue, disseminate and be compensated for loan CUSIPs from S&P to a purchaser, enacted by way of: (i) a modification of S&P’s existing agreement with the ABA to reflect the fact that S&P will no longer have any rights or licences to issue loan CUSIPs or disseminate the database of loan CUSIPs; and (ii) a new agreement between the purchaser and the ABA, under which the ABA will appoint the purchaser as its agent to operate the Loan CUSIP Divestment Business, on behalf of the ABA, which consists of (a) issuing or assigning loan CUSIP identifiers to eligible financial instruments and their issuers, and (b) disseminating the commercial database of existing, historical and new loan CUSIP identifiers and related descriptive information that is maintained and updated on a real time basis.

(781) The Loan CUSIP Commitments also include, among others, the transfer of the following tangible and intangible assets to a suitable purchaser:

The benefit of intellectual property rights (belonging to the ABA) related to loan CUSIPs;

An assignment of the existing agreement between CGS and the Loan Syndications and Trading Association (“LSTA”) regarding the issuance of loan CUSIPs;

All other supplier and customer agreements and relationships (or, in the case of shared contracts, the portion of such contracts which relates to the Loan CUSIP Business) which contribute to the current operation or are necessary to ensure the viability and competitiveness of the Loan CUSIP Business;

In respect of any contracts used by the Loan CUSIP Business which are shared with the wider S&P group, the portion of those contracts which relates to the Loan CUSIP Business on terms and conditions equivalent to those at present afforded to the Loan CUSIP Business.

Customer, and other records of the Loan CUSIP Business;

The personnel necessary to maintain the viability and competitiveness of the Loan CUSIP Business;

Benefit of the technology (e.g., data, databases and software) that is used in and necessary for the operation of the Loan CUSIP Business as of closing through one of the following mechanisms, at the option of the Purchaser:

(a) Migrating the technology to the Purchaser;

(b) Creating a logically separated, standalone and mirrored version of the technology and migrating it to the Purchaser; or

(c) Where this is not possible, or otherwise at the option of the Purchaser, S&P shall offer to enter into a transitional arrangement.

All licences, permits and authorisations issued by any governmental organisation for the benefit of the Loan CUSIP Business;

All electronic books, records and files that are related to the Loan CUSIP Business;

At the option of the Purchaser, the benefit of all transitional service arrangements, for up to 12 months, which are necessary to ensure the viability and competitiveness of the Loan CUSIP Business for a transitional period after divestiture, such as IT, HR and finance/payroll services; and

Any other tangible and intangible assets that are primarily related to the Loan CUSIP Business, which contribute to the current operation or which are necessary to ensure the viability and competitiveness of the Loan CUSIP Business.

(782) The Notifying Party argued that the Loan CUSIP Commitments eliminate entirely any competition concerns relating to loan identifiers by removing the entire overlap between the Parties’ activities in this market, as S&P would no longer be authorised to issue CUSIPs in the loan space, with the ABA instead appointing the Purchaser as the new provider to take over the operation and management of the loan CUSIP business.

(783) Nevertheless, the Parties have offered as an alternative to the Loan CUSIP Commitments and solely in the event that the Commission concludes that said commitments are insufficient following the results of the market test, to divest and transfer the entirety of S&P’s CUSIP issuance and data licensing business, as carried out currently by CUSIP Global Services; these are the Initial CUSIP Commitments.

(784) The Initial CUSIP Commitments include:

(a) All tangible and intangible assets owned by S&P that are primarily related to the CUSIP Business which contribute to the current operation or which are necessary to ensure the viability and competitiveness of the CUSIP Business, including all customer, credit and other records, and all required software;

(b) A transfer of the rights to issue, disseminate and be compensated for CUSIP identifiers from S&P to the purchaser, enacted by way of:

– a modification of S&P’s existing agreement with the ABA to reflect the fact that S&P will no longer have any rights or licences to issue CUSIP identifiers or disseminate the database of CUSIP identifiers; and

– a new agreement between the purchaser and the ABA, under which the ABA will appoint the purchaser to act as, on behalf of the ABA, (i) the sole issuer or assigner of CUSIP identifiers to eligible financial instruments and their issuers, and (ii) the sole disseminator of the commercial database of existing, historic and new CUSIP identifiers and related descriptive information that is maintained and updated on a real time basis;

– an assignment of the agreement between CGS and LSTA dated 13 June 2007 (the “LSTA Agreement”);