EC, October 20, 2006, No M.4350

Dear Sir, dear Madam,

Subject: Case No COMP/M.4350 - HEWLETT PACKARD / MERCURY INTERACTIVE

Notification of 15.09.2006 pursuant to Article 4 of Council Regulation No 139/2004^[1]

1.     On 15 September 2006, the Commission received a notification of a proposed concentration pursuant to Article 4 and following a referral pursuant to Article 4(5) of Council Regulation (EC) No 139/2004 (“the Merger Regulation”) by which the undertaking Hewlett Packard Company (“HP”, USA) acquires within the meaning of Article 3(l)(b) of the Council Regulation control of the whole of the undertaking Mercury Interactive Corporation (“Mercury”, USA) by way of purchase of shares

2.     After examination of the notification, the Commission has concluded that the notified operation falls within the scope of the Merger Regulation and does not raise serious doubts as to its compatibility with the common market and with the EEA Agreement.

I. THE PARTIES

3.     HP is a technology solution provider to consumers, businesses and institutions globally. Its offerings span IT infrastructure, personal computing and access devices, global services, and imaging and printing.

4.      Mercury provides business IT software and services. Its offerings include products in IT governance, application delivery and application management.

H. THE OPERATION

5.      The notified concentration consists of the acquisition by means of a tender offer of all the outstanding shares in Mercury by a special purpose vehicle solely controlled by HP. The offer which was launched on 17.08.2006. The proposed operation therefore consists in the acquisition of sole control over the whole of Mercury by HP.

HI. CONCENTRATION

6.      In view of the foregoing, the proposed transaction constitutes a concentration within the meaning of Article 3(l)(b) of the Merger Regulation.

IV.   COMMUNITY DIMENSION

7.      The notified concentration does not meet the turnover thresholds set out in Article 1(2) and 1(3) of the Merger Regulation.

8.      On 11.08.2006, the Commission received from the notifying party a referral request pursuant to Article 4(5) of the Merger Regulation which has been transmitted to all Member States. Since no Member State has expressed its disagreement as regards the request to refer the case to the Commission within the defined time limit, the concentration shall be deemed to have a Community dimension.

V. COMPETITIVE ASSESSMENT

A.  Relevant product market

9.      The parties’ software offerings overlap in the following product categories as defined by IDC^[2]: (i) systems infrastructure software, (ii) application software, and (hi) application development and deployment software. Since in the latter two categories, the parties’ activities combine only to a negligible degree, the analysis in the present case focuses on systems infrastructure software.

10.   Within this category, and on the basis of IDC’s software classification, the parties propose to further distinguish segments and sub-segments in order to reflect different categories of software functionalities. Horizontal overlap in the parties’ activities occurs in performance management software, which is a sub-segment of the system and network management software segment. Performance management software is used to quantify the actual performance of systems and applications once they have gone live with users, i.e. in the post-deployment phase. According to the notifying party, performance management software should be distinguished from testing software used in the pre-deployment phase, i.e. software used to analyse, design and test software being developed. In the IDC’s classification, pre-deployment testing software corresponds to automated software quality (“ASQ”)^[3].

11.   The market investigation widely confirmed that the distinction between ASQ and performance management software is relevant and warrants separate product markets. However a large majority of respondents also indicated that these two categories of product are largely complementary from both a supply and demand standpoint. From a demand perspective customers would increasingly be looking for integrated performance management solutions capable of addressing their systems and applications lifecycle needs on an “end to end” basis (i.e. pre- and post-deployment). From a supply perspective pre­deployment testing software would produce data, such as scripts, that can then be re-used in post-deployment performance management. In order to minimize costs and to ensure operational efficiency, customers would pay more and more attention to the interoperability of pre-and post-deployment software and would therefore favour vendors supplying integrated solutions.

12.   Software used to manage performance of IT systems and applications appears to be distributed among the computers in a given network based upon two different approaches: agent-based and agent-less. With the former, the necessary software is installed on each computer and is thereby able to collect comprehensive information about performance parameters and transmit them onto a monitoring console. Agent-less performance management relies entirely on the ability to get a defined subset of information via remote access to the computer. The market investigation did not provide a clear indication as to whether this distinction gives rise to separate relevant product markets. Whereas a wide majority of suppliers indicated that agent-less and agent-based performance management software provides the same functionality and should therefore not be distinguished, customers were divided on the question. However for the purpose of the present case it is not necessary to come to a definitive conclusion on this issue as the notified operation does not raise competition concerns under any alternative market definition.

13.   According to the parties, there are two different environments in which performance management software may be used: the mainframe environment (also called the non- distributed environment) and the distributed environment (i.e. systems using platforms such as Linux, Unix and Windows). However, the notifying party contends that the performance management products offered in each environment are functionally identical. Both HP and Mercury’s products target only the distributed environments, as they offer no solution for the mainframe environment. According to HP, entry by a mainframe-oriented supplier into the distributed area would be relatively easy, as the functions of the software would be identical. As a matter of fact, the main suppliers in the mainframe environment, like IBM, BMC and Computer Associates, also offer distributed performance management software. Hence, competition in the distributed environment would also come from vendors with products designed for the non-distributed environment.

14.   The market investigation did not provide a clear indication as to whether the distinction between distributed and non distribution warrants separate relevant product markets. Whereas a wide majority of customers indicated that performance management software used in the two environments are complementary rather than interchangeable, suppliers were divided on the question. However for the purpose of the present case the exact definition of the relevant product market can be left open as the notified operation does not raise competition concerns under any alternative market definition.

B.  Relevant geographic market

15.   With respect to the relevant geographic market, the notifying party, following the view expressed by the Commission in previous decisions⁴, submit that the geographic scope for IT software in general (and hence also the geographic scope of the narrower performance management software sub-segment) is at least EEA-wide and probably world-wide. According to the parties, prices are largely harmonised within the EEA. Baseline pricing is for the US/English language version and a supplement is added if the software is made available in a different language. However, its functionality remains identical in all language versions. The market investigation confirmed that the geographic scope of the relevant markets is at least EEA-wide.

16.   For the purpose of the case at hand it is not necessary to decide whether the relevant geographic market is world-wide or EEA-wide as the notified operation does not give rise to competition concern under any alternative definition.

C.  Analysis

Horizontal effects

17.   Table 1 below, based on IDC data, sets out the world-wide and EEA shares of the main vendors in the performance management software market in 2005.

4      See for example Commission decisions in Cases COMP/M.3062 - IBM/Rational and COMP/M.3978 - Oracle/Siebel.

5      IDC’s data collection for EMEA includes Europe, the Middle East and Africa. The EEA represents around 90% of sales in IDC’s EMEA region. According to the parties, it represents fairly the market share situation in the EEA both for the parties and their competitors. Moreover, such approach was followed for the purposes of market share discussion in previous cases dealing with the IT software sector (see for example Commission decision in Case COMP/M.3978, op.cit).

18.   At the world-wide level, HP and Mercury’ combined share of the performance management software market would be slightly below 15% (circa. 14%). At the EEA level, the parties would hold a slightly higher market share (circa. 16%). Strong competitors like IBM (with a market share of circa. 20%) and BMC (market share of circa. 14%) are present in the EEA market, followed by a number of other competitors such as Computer Associates, Compuware and Microsoft. The remaining 24.9% (world-wide: 29.9%) of the market is held by a large number of smaller players. It therefore appears that the new entity would continue to face effective competition both from established players but also from the many mid-size and smaller players who seem able to expand in this market.

19.   Table 2 below sets out the world-wide and EEA shares of the main competitors in the (hypothetical) distributed performance management software market in 2005. According to the parties, IDC does not collect data for Europe hence only data on a world-wide level were submitted in the notification. However, at the Commission’s request, the notifying party further provide EEA shares calculated from raw IDC data.

20.   In the distributed performance management software sub-segment, the new entity would face the same competitors as in the overall performance management software market (distributed + non-distributed). At the world-wide level, HP and Mercury’ combined share of the distributed performance management software market would be circa. 22.6%. At the EEA level, the parties would hold a higher market share (circa. 27.1%). HP’s market share is higher in the distributed environment than in the overall market (i.e. distributed + non- distributed). This reflects the fact that HP has traditionally focussed on the distributed segment of the market. Competitors at both world-wide and EEA levels include BMC, IBM, Computer Associates, and mid-size players, such as Quest Software, Microsoft, NetlQ and Compuware. The remainder of the market (27% world-wide and 19.3% in the EEA) is held by a large number of smaller players. It therefore appears that the new entity would continue to face effective competition both from established players but also from the many mid-size and smaller players who seem able to expand in this market.

21.   In addition, the presence of vendors with offerings for both the non-distributed and the distributed environments and which are large IT vendors, such as IBM and BMC, makes it unlikely that the proposed transaction would give rise to competition concerns even in a narrow EEA market for distributed performance management software. An additional important element pointing to that conclusion is that customers of such products are typically large and sophisticated buyers (telecommunications operators, public authorities, financial companies, etc.) with significant bargaining power. Even smaller customers can rely on consultants and system integrators, like Accenture, Cap Gemini Emst&Young and Deloitte, to exert pressure on suppliers.

22.   Finally, should distinct product markets be found for agent-based and agent-less performance management software, the notifying party indicates that the overlap between the parties’ activities in each of these markets would be lower than in the overall performance management category. HP explains that while HP’s strength lies mainly in agent-based software, Mercury’s performance management products are generally agent­less. The market investigation confirmed that Mercury’s focus is on agent-less solutions whereas HP’s focus is on agent-based.

Conglomerate effects

23.   Mercury mostly produces pre-deployment testing software which accounts for roughly two- thirds of its turnover and where it is the market leader with an estimated market share of 45.2% world-wide (45.9% in the EEA) in 2005. According to IDC, HP does not supply pre­deployment testing software but only post-deployment performance management software, where Mercury is also active. As noted above, testing products and performance management software are often seen by customers as complementary products as they address similar needs at different phases of the systems and applications lifecycle (i.e. pre- and post-deployment). The proposed transaction will therefore expand HP’s software offering into a neighbouring market. The Commission has therefore assessed whether this increase in the scope of HP’s software offering could raise conglomerate concerns.

24.   In this respect, it should be noted that there are several software vendors, including IBM, Compuware, Microsoft, SAS, and Allen Systems Group, who supply products in both the pre- and post-deployment phases of the systems and applications lifecycle. Therefore, any attempt by the merged entity to try to leverage its position from one of these markets into the other through bundled offerings could be replicated by several competing vendors. Amongst them, IBM and Compuware hold significant market shares (world-wide and in the EEA) in both testing^[7] and performance software markets. In addition, the market investigation also revealed that different software vendors can team-up in order to a maximize interoperability between testing and performance management software in order to offer attractive solutions to customers. It seems therefore unlikely that the addition of Mercury’s testing products and HP’s performance management software will significantly impede effective competition in any of these markets.

25.   A respondent in the market investigation also raised concerns that the proposed transaction would change the incentives of HP to grant timely access to key interoperability information on its OpenView system management platform to “independent” vendors of performance management software (i.e. software vendors who do not have a system management software offering). This respondent explained that should HP implement such a strategy, independent vendors might be foreclosed from customers using HP’s OpenView product because performance management software needs to interact with other system management software. Without it being necessary to examine whether this third party’s allegations are technically founded, it is sufficient to recall that HP was already active in the performance management software market pre-merger and that the notified operation will result only in a relatively modest increase of HP’s market share (see above). It is therefore unlikely that HP’s incentives to grant timely access to interoperability information on its OpenView platform to independent suppliers of performance management software would change post-transaction. In any case, should HP’s incentive in this respect change post­merger, this would not entail a significant negative impact on competition due to HP’s limited share of the market(s) for system management software (9.4% world-wide and 13.1% in the EEA based on IDC data^[8]).

VI.   CONCLUSION

26.   For the above reasons, the Commission has decided not to oppose the notified operation and to declare it compatible with the common market and with the EEA Agreement. This decision is adopted in application of Article 6(l)(b) of Council Regulation (EC) No 139/2004.

[1]     OIL 24, 29.1.2004 p. 1.

[2]     IDC is an industry analyst which provides market data, research and analysis on the information technology (IT) sector. The IT industry widely recognises and relies upon IDC’s classification.

[3]     In IDC’s classification, ASQ is a sub-segment of the application development and deployment category.

[4]     Including Wily Technology, which was acquired by Computer Associates in March 2006.

[5]     IDC’s data collection for EMEA includes Europe, the Middle East and Africa. The EEA represents around 90% of sales in IDC’s EMEA region. According to the parties, it represents fairly the market share situation in the EEA both for the parties and their competitors. Moreover, such approach was followed for the purposes of market share discussion in previous cases dealing with the IT software sector (see for example Commission decision in Case COMP/M.3978, op.cit).

[6]     Including Wily Technology, which was acquired by Computer Associates in March 2006.

[7]     IBM’s share of the testing software market was 14.1% world-wide and 17.2% in the EEA. Compuware’s share of the testing software market was 7.8% world-wide and 7.1% EEA (these figures are based on IDC data for 2005).

[8]     Mercury is only active in the system management software market(s) via its performance management software offering.