ApexHi Properties Limited and Prima Property Trust (68/LM/Jul05) [2005] ZACT 80 (10 November 2005)

60 Reportability
Competition Law

Brief Summary

Competition Law — Merger Approval — Approval of merger between ApexHi Properties Limited and Prima Property Trust — ApexHi acquiring Prima's business as a going concern — Transaction aimed at enhancing portfolio quality and reducing operational risk — Commission's qualitative analysis concluded no significant anticompetitive effects — Tribunal found post-merger market shares in overlapping categories to be slight or insignificant — Merger approved as it does not substantially lessen competition in the property market.

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 property­owning 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 Fund 1          10%
Marriott Property Fund 2          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   third­largest  
property loan stock company listed on the JSE.
1  Another property company listed on the JSE.
2  Also a JSE­listed property company.
2

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 ill­assorted 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 so­called 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  
3

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 post­merger 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   same­category   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 C­grade office space in the  
Johannesburg CBD (11.08%), B­grade 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

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   81­83   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 property­owning 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),  
4

as mentioned above, demonstrate that both pre­ and post­merger 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 ApexHi­Prima 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

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  
5

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 Shaw­Taylor,  
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

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. 
6

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 ­­
a) beneficially owns more than one half of the issued share capital of the  
firm;
b) 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;
c) is   able   to   appoint   or   to   veto   the   appointment   of   a   majority   of   the  
directors of the firm;
d) 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…..
e) 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;
f) 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
g) has the ability to materially influence the policy of the firm in a manner

g) 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.
7

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 post­merger.
30. Both sections 4 and  12  of the Competition Act  contain pitfalls for  firms  
which   co­ordinate   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
8

Concurring:  T Orleyn, M Mokuena
9