---

COMPETITION TRIBUNAL

REPUBLIC OF SOUTH AFRICA

Case no.: 68/LM/Jul05

In the large merger between:

ApexHi Properties Limited

and

Prima Property Trust ________________________________________________________________

Reasons

________________________________________________________________

Introduction

1.

1. On 21 October 2005 the Competition Tribunal approved the merger between ApexHi Properties Limited and Prima Property Trust. The reasons are set out below.

The parties and the transaction

2. ApexHi Properties Limited ("ApexHi") is acquiring the entire business of Prima Property Trust ("Prima") as a going concern. The business comprises the properties, lease agreements, and other contracts of Prima and its property-owning subsidiaries.

3. ApexHi is a variable rate property loan stock company listed on the JSE in the real estate sector. Prima is a collective investment scheme, also listed on the JSE.

4. The purchase price will be paid partly in cash and partly in the linked "A" and "B" units in ApexHi.

5. ApexHi owns and rents out properties in the retail, commercial (office), and industrial categories. Prima also operates in these categories. Overlap exists between these categories in some of the geographical areas or nodes in which the parties each have properties.

Rationale for the transaction

6. ApexHi considers that the transaction will enhance and complement the quality and size of its portfolio of properties and will add diversity to and lower the risk of its business operations. Prima appears to consider that it is too small an entity, and has properties which are individually too small, for it to thrive in what the parties perceive as a highly competitive industry.

Primary acquiring firm

7. ApexHi asserts that is not directly or indirectly controlled by anybody. Its unitholders are listed as follows:

Investec funds 5%

Stanlib 7%

Redefine Income Fund¹ 10%

Marriott Property Fund² 13%

Other 65%

8. ApexHi owns a portfolio of properties valued at 30 June 2004 (the date of the latest annual report) at some R3 billion. In terms of value, these properties are broken down by category as follows:

Percentage of total portfolio

Retail 35

Commercial (offices) 47

Industrial 18

These properties are widely dispersed over the cities and towns of South Africa.

9. In terms of market capitalisation, the combined value of ApexHi's A and B units (which have different entitlement rights to the income of the company) was R4.1 billion at 5 July 2005, making it the third-largest property loan stock company listed on the JSE.

Primary target firm

10. Prima is represented for formal purposes by a trustee, Absa Bank Limited and by Prima Property Trust Managers ("PPTM") as manager. The Tribunal was not provided with information about the identity of its unitholders but they can be assumed to be a variety of institutional and private investors.

11. From the annual report for the period ending 30 April 2004 it appears that Prima's property portfolio at that date was valued at R510 million. Measured in net income, the categories in this portfolio were:

Percentage

Retail 57

Commercial 31

Industrial 10

In addition, interest contributed 2% of income, bringing the total to 100%.

12. The properties in Prima's portfolio are chiefly smaller properties, spread over seven provinces, with 47% of net income being generated in Gauteng and 12% in Mpumalanga.

13. Prima's market capitalisation at 5 July 2005 was R0.7 billion, making it the smallest of the listed property unit trusts.

Market shares in the property industry

14. From the state of the papers referred to the Tribunal in this merger, it would seem that comprehensive and reliable statistics on the size of the market and on market shares in South Africa are either not maintained or are at least difficult to come by. The Commission, faced with a bewildering presentation of ill-assorted data in the parties' merger documents, gave up the attempt to make a proper quantitative assessment, and largely approached the merger on the basis of a so-called qualitative analysis. This analysis, relying apparently on such factors as the purported canniness in negotiations of large retail and other tenants, and the alleged ease with which property lease agreements can be amended or abrogated, led the Commission to conclude that in none of the three property categories mentioned above was there a concern about possible anticompetitive effects of the merger.

15. The Commission did however essay some analysis of market shares, using statistics produced by the South African Property Owners' Association ("SAPOA") and an organisation named Independent Property Databank. In calculating post-merger market shares on the basis of the figures originally supplied by the parties, the Commission, while warning that it considered some of the figures unreliable, produced tables which included such improbable outcomes as that the merging parties' share of the convenience retail sector in the Johannesburg CBD would exceed 100%.

16. Only on 17 October 2005 did the Commission, in response to an urgent appeal for more complete and accurate figures, receive somewhat more acceptable statistics from the parties' legal representatives. These figures suggest that in some of the areas or nodes in which there is overlap between the same-category operations of ApexHi and Prima, the post-merger market shares are in fact slight or possibly even insignificant. Specific corrections were made in respect of C-grade office space in the Johannesburg CBD (11.08%), B-grade office space in the Rivonia node (11.5%), and industrial space in the Spartan/Kempton Park node (3.2%).

17. We are unconvinced that in this case a thorough and purposeful approach to the task of assembling the essential facts of market size and market share was adopted. It is to be hoped that players in the property industry will apply their undoubted resources and talents to improving this position, and that due heed will be paid to the presentation of information in the documents filed in property mergers in future to ensure that an imbroglio of this kind is not repeated.

18. As it is, the best approach the Tribunal was able to take to the overall dimensions of the market, apart from the corrected details mentioned above, was based on the parties' exposition in their merger documents of the market capitalisation of the major participants in the property industry, to the extent that these are listed entities. From the data on this subject presented on pages 81-83 of the record, it appears that property loan stock companies collectively have a total capitalisation of R30 billion, and property unit trusts R13 billion. Moreover the market capitalisation of very large property-owning entities outside the listed property sector are Old Mutual R10 billion, Sanlam R8 billion, and Liberty Life R6 billion. This information appears to have been correct as at 8 July 2005. In this context the market capitalisation of ApexHi (R4.1 billion), and Prima (R0.7 billion), as mentioned above, demonstrate that both pre- and post-merger they are relatively small players in the overall property market, to the extent that it is listed. The unlisted portion, which has not been quantified in the papers made available to the Tribunal, adds a further and apparently significant quantum to the size of the overall market.

19. Market capitalisation is not a reliable proxy for market share, measured in terms of the usual metric of the property industry, namely the area (square metres) of competitors in like categories in geographical nodes representing relevant markets. Wrapped within market capitalisation may be a premium or discount in relation to net asset value, and even net asset value may not have a simple relationship to the area (in square metres) of underlying properties, while the issue of property categories further complicates the relationship. Valuation procedures and the timing of valuations have their own complications and inconsistencies, to add yet further complexity. Despite all this, the conclusion is inescapable that the combined ApexHi-Prima entity will not be a large player overall, and the geographical breakdowns of the ApexHi and Prima portfolios, to the extent to which they are intelligibly presented in the papers before the Tribunal, do not reveal significant nodes where there is sufficient market strength to raise competition concerns.

The issue of control

20. There is a further aspect of this case where the Tribunal considers the groundwork was not adequately done in the papers making up the record which it was required to consider. This is the issue of control of ApexHi and Prima, and for that matter certain other entities which have links to them.

21. It emerges from the papers, and specially from the organogram of the pre- merger corporate structure set out on page 17 of the record, that a company named Madison Property Fund Managers ("Madison") owns 100% of a subsidiary named Million Up Investments 158, which in turn owns 100% of PPTM, the management company of Prima, mentioned earlier. Madison is also the majority shareholder of ApexHi Manco Trust, a company which undertakes the operational management of ApexHi. Madison moreover provides the operational management for the Redefine Income Fund, a listed company owning a variety of property assets, and it has a 30% interest in the management company of a yet further listed property loan stock company, Hyprop, which has a significant interest in the Canal Walk complex outside Cape Town, which is in turn managed by a company in which Madison once again has a significant interest, in this instance 24%.

22. Madison is stated in the organogram as filed to be owned as to 50% by Standard Bank, which is known to control the Liberty assurance group -- a major owner of property in South Africa, particularly in the retail area. The other shareholders are two individuals, Messrs Cesman and Wainer. At the hearing we were told that since the date of the merger notification Standard Bank had disposed of half of its holding to a Mr Shaw-Taylor, who now, with Messrs Cesman and Wainer, each have a 25% stake in Madison.

23. Madison is thus at the centre of the operations of several listed properties and has shareholding links to others. Two of its individual shareholders, Messrs Cesman and Wainer, are also on the boards of some of these listed entities, including ApexHi. The decision taken by PPTM on behalf of Prima to conclude the merger agreement with ApexHi was taken at a board meeting of PPTH held on 7 April 2005, at which the directors present were recorded as Messrs Haasbroek (chairman), Levy, Wainer and Cesman. The relevant entry in the minutes reads:

THE APEXHI OFFER

The meeting referred to the ApexHi offer ….and accompanying draft Agreement….the contents of which were noted by the meeting.

Prior to discussing the issue, Messrs Cesman and Wainer made Directors' Declarations if Interest in the proposed transaction and requested that it be noted that they are directors of ApexHi, Redefine, Hyprop and directors and shareholders of Madison Property Fund Managers (Pty) Ltd, and are to be regarded as interested parties in the proposed transaction. It was further noted that A. Levy is employed as the asset manager of Prima…. The Chairman observed that in the light of the Declarations of Interest by the other directors of PPTM, it appeared that he was the only director that did not have a potential conflict of interest in the matter."

24. At the hearing before the Tribunal, Mr Feinblum, a director of ApexHi, stated that because of the extent of the conflict of interest mentioned above an independent expert opinion had been obtained by Mr Haasbroek to confirm that the offer by ApexHi was fair and reasonable.

25. While these issues of conflict of interest appear on the face of them to have more to do with the interests of investors than with competition, there is an unexplored competition issue which should be mentioned.

26. Mr Feinblum, when asked about the extent to which Madison influences the decisions taken by the boards of directors of the entities to which Madison and its subsidiaries provide management services, was candid about the matter. He said that in some cases the directors of property companies have very little practical knowledge and experience of the property industry and therefore bring in experienced professional managers who effectively run the companies for them. This raises in a clarion manner the question whether Madison does not have effective control of some or all of the property companies which it or its subsidiaries manage, at least within the definition set out in s. 12(2)(g) of the Competition Act. S. 12(2) reads:

A person controls a firm if that person --

1. beneficially owns more than one half of the issued share capital of the firm;

2. is entitled to vote a majority of the votes that may be cast at a general meeting of the firm, or has the ability to control the voting of a majority of those votes, either directly or through a controlled entity of that person;

3. is able to appoint or to veto the appointment of a majority of the directors of the firm;

4. is a holding company, and the firm is a subsidiary of that company as contemplated in section 1(3)(a) of the Companies Act, 1973…..

5. in the case of a firm that is a trust, has the ability to control the majority of the votes of the trustees, to appoint the majority of the trustees or to appoint or change the majority of the beneficiaries of the trust;

6. in the case of a close corporation, owns the majority of members' interest or controls directly or has the right to control the majority of members' votes in the close corporation; or

7. has the ability to materially influence the policy of the firm in a manner comparable to a person who, in ordinary commercial parlance, can exercise an element of control referred to in paragraphs (a) to (f).

27. The essential features of s. 12(2) (g) are thus whether there is "material influence" over the affairs of a firm and if so the nature of that influence when compared with the other, classic, forms of control of a firm.

28. If there is control as contemplated by s.12(2)(g) by Madison over Prima and possibly also of ApexHi, then the question arises of the extent to which the merging parties in this transaction have been accurately described, and whether others, even extending to the very large retail property interests of Liberty Life (through its controlling entity, Standard Bank) are not implicated in the merger.

29. Paradoxically, however, such control by Madison might have the consequence that the transaction between ApexHi and Prima would not be a notifiable merger since both would be controlled by Madison, pre- and post-merger.

30. Both sections 4 and 12 of the Competition Act contain pitfalls for firms which co-ordinate their activities through agreements or concerted practices to the point where they enter the zone of prohibited anti-competitiveness, and these features of South African competition system should not be overlooked in their potential application to merger transactions.

31. Regrettably these issues were not explored at all in the merging parties' notifications nor in the Commission's consideration of the transaction and its ultimate recommendation to the Tribunal. The Tribunal raised certain questions with Mr Feinblum at the hearing which may warrant consideration when similar cases arise of mergers in which there are interlocking relationships between the directors of companies which nominally own assets and the companies which exercise operational management over them.

Effect on competition

32. Given the state of the information in the record which it was called upon to consider, the Tribunal has no basis for saying that the merger will have any anticompetitive effects.

Public interest

33. The transaction does not raise any public interest issues.

Conclusion

34. Accordingly, the merger must be approved.

____________ 10 November 2005

L Reyburn Date

Concurring: T Orleyn, M Mokuena

---