Bank control, takeovers and corporate governance in Germany
Julian Franks a, Colin Mayer b,*
a London Business School, London, UK
b Said Business School, University of Oxford, Radcli?e In®rmary, Woodstock Road,
Oxford OX2 6HE, England, UK
Abstract
This paper examines the three cases of hostile takeovers in Germany in the postSecond World War period. It describes the important role played by banks in 英國dissertation網a?ectingthe outcome of the bids: bank representatives were chairmen of the supervisory board inall three cases and banks voted a large number of proxies in important decisions affectingthe bids. The paper reports that low returns were earned by shareholders of twoof the target ®rms and o?ers an explanation in terms of bank control and the regulatoryregime operating in Germany. Ó 1998 Elsevier Science B.V. All rights reserved.
JEL classi®cation: G32; G34
Keywords: Takeovers; Bank control; Corporate governance
1. Introduction
The recent attempted hostile acquisition of Thyssen AG by Krupp AG
has once again brought to the fore the operation of the German corporate
governance system and the role of the banks. The banks have been seen to
Journal of Banking & Finance 22 (1998) 1385±1403
* Corresponding author. Present address: Department of Economics, Stanford University,
Stanford, CA 94305-6072, USA. Tel.: 1 650 725 7836; fax: 1 650 725 5702; e-mail: cmayer1@leland.
stanford.edu
0378-4266/98/$ ± see front matter Ó 1998 Elsevier Science B.V. All rights reserved.
PII: S 0 3 7 8 - 4 2 6 6 ( 9 8 ) 0 0 0 6 0 - 0
be instrumental in orchestrating and organising the raid. To some, most
notably German steel workers and particular politicians, the banks haveabused the social values of the German governance system. To others, theyhave let shareholders down in failing to push through the bid to the bitterend.Hostile takeovers in Germany are an important subject of study for tworeasons. Firstly, there are not many of them: only three in the whole of thepost Second World War period. Secondly, although they are the earthquakesof German corporate governance and should not therefore be regardedas examples of normal practice, they do allow us to observe theoperation of German corporate governance and the behaviour of bankswith an unusual measure of clarity. In addition, they provide an interestinglaboratory on how a virtually unregulated takeover market a?ects shareholderreturns.
In particular, we want to use the hostile bids to examine two questions:
®rstly, do banks exercise substantial control during these turbulent periodsand, secondly, if they do, in whose interests do they act. There is continuingdiscussion amongst both academics and policy makers about the power ofGerman banks. In principle, concentration of control through proxy votesshould encourage more active corporate governance by German banks than byUK and US ®nancial institutions which hold much smaller stakes and largehighly diversi®ed portfolios of shares. Since banksÕ own shareholdings are ingeneral modest, where con¯icts arise between shareholder interests and incumbentmanagement, banks may attach less signi®cance to their custodianfunctions than to the margin and fee income which they derive from commercialand investment banking. Banks may also feel they have obligations #p#分頁標題#e#http://www.400190.com/dissertation_writing/Finance/toother stakeholders such as employees and managers.We ®nd that banks do exert signi®cant in¯uence over the outcomes ofbids, their power in large part deriving from their chairmanship of supervisoryboards, and the proxy votes which they cast on behalf of individualshareholders. The latter may be especially important where there are restrictionson voting rights: these limit the votes that large blockholders cancast but not those of banks acting as custodians of small shareholders. We
®nd that returns to shareholders who do not sell their shares to acquirors arevery low and even negative in two of the bids. These low bid premia may beexplained by banksÕ concern with interests other than those of shareholders;alternatively, they may result from the virtual absence of takeover regulationin Germany.
We begin by describing the methodology and data used in the study in
Section 2. The three cases of hostile acquisitions are discussed in Sections 3±5.
Section 6 analyses the returns to shareholders of the target ®rms, the role of
takeover regulation, and the power of the banks. Finally, we derive some
conclusions in Section 7.
1386 J. Franks, C. Mayer / Journal of Banking & Finance 22 (1998) 1385±1403
2. Methodology and data
This paper reports the results of detailed studies of the three cases of hostileacquisitions which have occurred in Germany in the post Second World Warperiod. Jenkinson and Ljungqvist (1996) have correctly pointed out that theincidence of hostility may be much more pervasive than the three cases wouldsuggest. There is an active market in share blocks (see Franks and Mayer,1997) and the transfers in control to which these trades give rise are oftenopposed by the management concerned. However, what distinguishes the threecases of hostile bids is that they involve companies whose shares are widelyheld and where a change in control could not be secured by agreement betweena small number of large blockholders. As a result, banks derive a greater degreeof control in these cases from their chairmanship of the board and their custodianshipof small shareholdings.
The three cases are the bid for Feldmuhle Nobel AG by the Flick Brothersin 1988 and then by Veba AG in 1989; the bid for Continental AG by PirelliAG in 1990 and 1991; and the bid for Hoesch AG by Krupp AG in 1991 and1992. Case studies of the three hostile takeovers were compiled from originalsource data. They included company accounts, press reports in both Germanand UK newspapers and Textline services. They were supplemented by interviewswith representatives of the bidder and target, supervisory board members(for example, Dr. Ulrich Weiss, Deutsche Bank board representative andchairman of ContinentalÕs supervisory board and Ellen Schneider-Lenne, DeutscheBank board representative), investment advisors (for example, Dr.Weickart, advisor to the investment group of Feldmuhle Nobel), investmentbanks (for example, Morgan Grenfell) and management consultants (for example,McKinsey, Frankfurt).#p#分頁標題#e#
Franks and Mayer (1997) report that 85% of a sample of 171 large companies
in 1991 had a single shareholder owning more than 25% of shares. None
of the three targets of hostile acquisitions had a single shareholder owning
more than 25% of shares prior to the bid. Franks and Mayer (1997) record that
banks are represented on the supervisory boards of many widely held companies,
frequently in the all-important position of chairman. Data on board
composition for the ®rms involved in the acquisitions were collected from
Hoppenstedt and Wer ist Wer. In all three target ®rms, Deutsche Bank occupied
the position of chairman of the supervisory board.
Shares in Germany are held in bearer form and banks act as custodians of
individual shareholders. In this capacity they o?er a free service to shareholders,
whereby the bank will o?er advice and vote on behalf of shareholders
in company resolutions. They are required to obtain permission annually from
shareholders to use their proxies and they must inform shareholders of impending
resolutions and how they intend to vote. Nibler (1998) reports that for
38 ®rms the top three banks controlled 29.5% of the votes at the AGMs in
J. Franks, C. Mayer / Journal of Banking & Finance 22 (1998) 1385±1403 1387
1991, and the votes for all banks totalled 62.6%. However, for a larger sample
of companies totalling 158 ®rms the average held by all banks was 24.7%. This
lower percentage is explained by greater concentration of ownership in the
larger sample.
Consistent with the observation that the targets are widely held, Gottschalk
(1988) records that in 1986 the three main banks (Deutsche Bank, Dresdner
Bank and Commerzbank) between them held 48% of Hoesch AG and 39% of
Continental AG. In the case of Feldmuhle Nobel, Deutsche Bank cast 55% of
the votes in a resolution supporting the introduction of a voting right restriction
of 5%.
Voting right restrictions limit the maximum number of votes which any one
shareholder can cast, irrespective of the number of shares which they hold.
They are commonly observed in the minority of German companies which are
widely held. Continental AG introduced a voting right restriction of 5% in
1984 and Hoesch AG introduced a restriction of 15% in 1977. As described
below, the 5% voting right restriction in Feldmuhle Nobel was introduced in
1988 in the middle of the bid by the Flick Brothers.
Voting right restrictions are virtually never observed in the large majority of
companies which have dominant shareholders, since they undermine the ability
of the dominant shareholder to exercise control. In a list of 19 companies
which had voting right restrictions in 1988, we found that 17 had no single
shareholder owning more than 25% of shares.
The performance of the acquisitions was measured by abnormal share price#p#分頁標題#e#
returns relative to the DAX index. Share price data on the bidder and target
®rms were collected from Datastream for the entire period of the bid from the
date at which the acquiror started share purchases to the date at which the bid
was completed or abandoned.
3. Bid for Feldmihle Nobel AG
3.1. Background to the bid
In 1985, ownership of the Flick Group was transferred from Friedrich Karl
Flick to Deutsche Bank. The Group included substantial share stakes in other
companies and some industrial subsidiaries (Dynamit Nobel, Buderus and
Feldhmuhle). The latter were grouped together to form a new company
Feldmuhle Nobel. The sale by Flick was made so as to realize a large sum for
the payment of taxes; he did not anticipate any family succession. Almost as
soon as it took control, Deutsche Bank sold Feldmuhle NobelÕs share stakes in
Daimler-Benz, WR Grace and Versicherungs-Holding der Deutschen Industrie
for about DM 5.9 billion. In April 1986, the shares of Feldmuhle Nobel were
o?ered for sale at a price of DM 285 per share. The issue was the largest in
1388 J. Franks, C. Mayer / Journal of Banking & Finance 22 (1998) 1385±1403
German history (DM 2 billion) and it was the ®rst time that 100% of a company
Õs share capital was o?ered via a new issue. The issue was also unusual in
that the shares were in the form of ordinary voting as against non-voting
preference shares.
Except for a stake of about 10% held by Deutsche Bank, the shares were
widely held. Some believed that the failure of a large stakeholder to emerge
re¯ected a lack of industrial logic in the structure of the new company. The
combined value of the earlier sales of share stakes, subsidiaries and the ¯otation
amounted to DM 7.9 bn.
3.2. The bid
In June 1987, Gert-Rudolf Flick and Friedrich Christian Flick, nephews of
Friedrich Karl Flick, shareholders in Feldmuhle Nobel, proposed a resolution
at the companyÕs AGM that an investigation be undertaken into the sale of
shares of WR Grace which took place in 1985. They accused the chairman of
Feldmuhle Nobel, Dr. Herbert Blaschke, of failing to obtain the highest price
possible in the sale.
In June 1988 it was reported that the Flick brothers were attempting to put
together a tender o?er for the company for a price of DM 350 per share with
the clear intention of replacing the management. Following the failure of the
bid to materialize, the share price dropped to DM 260. However, it was well
known that the Flick brothers and other investors were accumulating a large
stake and were being advised by specialists in control contests, a commercial
lawyer, Dr. Weickart, and Merrill Lynch.
Subsequently, the management learnt that the Flick brothers and others
associated with them had a block stake of 36.5% and attempted to forestall#p#分頁標題#e#
any hostile action by proposing a resolution at its AGM restricting the voting
rights of any one shareholder to 5% of the total shares outstanding. The
resolution was supported by Deutsche Bank which held a share stake of
about 8% on its own account and was able to vote a large proportion of the
shares through proxies and shares held in investment funds administered by
the bank. The resolution was passed with Deutsche Bank voting 55% of the
shares cast.
Some of the shares had been purchased directly in the market while others
had been acquired through the conversion of options bought from institutions
with signi®cant stakes. Sellers of the options were required not to reveal information
about the transaction for fear of stimulating a rise in the share price.
In the latter part of 1988 there were rumors that a takeover bid was again being
made for Feldmuhle Nobel. Trading in the shares was unusually high and the
share price increased to DM 310. In May 1989, Veba AG declared that it was
acquiring a 46% stake, including the stake held by the group of shareholders
led by the Flick brothers.
J. Franks, C. Mayer / Journal of Banking & Finance 22 (1998) 1385±1403 1389
Veba secured two seats on the supervisory board, one for its chairman,
Rudolph von Bennigsen-Foerder, and one other. The former became chairman
on 18 July 1989 but died on 28 October 1989. In December 1989, Veba con-
®rmed that it had increased its stake to just over 50% by buying further shares
mainly from investment funds, and expressed its intention to raise its stake to
61%.
Unknown to Veba, the Flick brothers had retained a stake in Feldmuhle
Nobel either in the hope that von Bennigsen-Foerder would successfully restructure
the company, or that Veba would eventually buy them out. However,
following the unexpected death of von Bennigsen-Foerder, the new management
was less inclined to purchase further shares because of the price that the
minority shareholders led by the Flick brothers were demanding.
Reluctant to make a full bid at a price required by the minority and unable
to take managerial control of the company in the face of opposition from the
Flick brothers, Veba put their stake up for auction. In April, 1990, Stora
Kopparbergs Bergslags, with another Swedish company, Patricia, acquired a
85% stake in Feldmuhle Nobel at a price of DM 567 and made an o?er of DM
540 to small shareholders for the outstanding 15%. Stora purchased 60.1% of
Feldmuhle Nobel and Patricia 24.9%. The sellers included Veba (51%), SCA of
Sweden(5%) and the Flick brothers (between 10% and 20%). Whereas Stora
and Feldmuhle NobelÕs paper activities were thought to be good complements,
other activities of Feldmuhle Nobel were to be sold (Dynamit Nobel and
Buderus).
In June 1990, the voting rights restrictions on Feldmuhle NobelÕs shares#p#分頁標題#e#
were lifted. The head of Veba, Klaus Piltz, vacated his position as head of the
supervisory board and was replaced by Dr. Hellmut Kruse, board chairman
of Beiersdorf. In August 1990, the takeover was approved by the Cartel
O?ce.
4. The bid for Hoesch AG
4.1. Background to the bid
Prior to the bid by Krupp for Hoesch in 1991, there had been a series of
unsuccessful attempts to arrange mergers in the steel industry. In 1981, Krupp
approached Hoesch; in 1983 Krupp approached Thyssen; failed negotiations
took place with Klockner Werke, and in 1989 with Salzgitter.
This history of failed friendly mergers convinced Krupp that a merger could
only succeed where control could be purchased in the market. Hoesch conformed
to these requirements since there was no large industrial stakeholder.
The bid by Krupp was motivated by a need to restructure the steel industry
which could not be achieved by internal means.
1390 J. Franks, C. Mayer / Journal of Banking & Finance 22 (1998) 1385±1403
4.2. The bid
In October 1991, Krupp announced the purchase of 24.9% of HoeschÕs
shares and expressed a wish to seek a ``close alliance''. Unusually, this was
announced without prior consultation with Hoesch. The shares were purchased
over a 5 month period from January to June 1991. The accumulation
of shares was facilitated by the Gulf War, which caused substantial selling,
and by the presence of foreign shareholdings ± 15±18% of the shares were
held in London. The success of this bid was crucially dependent on the ability
to acquire control covertly through the accumulation of shares before Hoesch
could amass support. Secrecy was maintained with a Swiss bank responsible
for the transactions; in September it was rumored that a Japanese ®rm was
about to bid.
None of KruppÕs major banks, including Deutsche Bank, were informed of
the share accumulation and the intention to bid. The lack of consultation re-
¯ected the need for secrecy and the changing relationships of some large
companies with their banks. Krupp did not regard any of its banks as the
Ôhouse bankÕ.
Hoesch called the Krupp purchase unfriendly. Neukirchen, who had become
CEO of Hoesch three months before at Deutsche BankÕs instigation, strongly
opposed the bid. Neukirchen addressed workersÕ meetings and attacked the
merger with Krupp, as did the WorkersÕ Council. As a response, the Hoesch
supervisory board considered a programme of restructuring called ``Hoesch
2000''. In addition, talks were initiated with British Steel about taking a
blocking minority stake.
In November, Krupp announced that, with its own stake of 24.9% and with
30.4% controlled by banks and institutional investors favourable to the merger,
it could exercise e?ective control. It also purchased shares in HoeschÕs convertible,#p#分頁標題#e#
which did not count in any stake that formed the basis of the 25%
disclosure rule, prior to conversion. The purchase was necessary to prevent
other parties from buying and then converting the shares. Krupp was able to
exert e?ective control with only a slender majority, not the 75% that is frequently
cited (see, for example, Franks and Mayer, 1997).
Although there were voting restrictions in Hoesch, the limitation was 15%
compared with only 5% in Feldmuhle Nobel. As a result, a minority shareholder
would have required a large stake of 15% to match the size of KruppÕs
votes.
The attitude of the banks was mixed. When informed, Deutsche Bank did
not oppose the bid possibly recognizing that mergers in the steel industry were
necessary and had proved di?cult to arrange in the past. WestLB, the one
major shareholder in Hoesch, which held about 17% and was also represented
on the board of Krupp, did not give public support to the bid and their attitude
to the bid was uncertain. This uncertainty may have re¯ected the fact that the
J. Franks, C. Mayer / Journal of Banking & Finance 22 (1998) 1385±1403 1391
Land was a major shareholder in the bank and wished to maintain a low public
stance in a merger that potentially involved extensive restructurings and a loss
of jobs.
Late in November, Krupp and Hoesch agreed to merger talks. A share for
share exchange was proposed, with the terms to be decided after the businesses
had been valued. It was also decided that Neukirchen would leave the management
board in the event of a successful merger being negotiated. The success
of the bid with only a small majority shareholding therefore came in the
face of strenuous opposition from the management board and workers of the
target.
In December, Krupp increased its share stake to 51%. Subsequently, it was
increased to 62% so as to provide the Krupp Foundation, which had a 75%
stake in Krupp, with a majority of 54% in the merged company. Other
shareholders in Krupp were Iran (25%) and WestLB (10%) and dispersed
shareholdings amounted to 10±13%. The acquisition of a majority of the shares
ensured that Krupp had control even in the event of the merger failing to take
place.
In February 1992, trade union o?cials listed their demands to Krupp. They
included the maintenance of production sites and jobs. Krupp announced that
additional job losses resulting directly from the merger, over and above those
contained in the ``Hoesch 2000'' plan, would be limited to 1800 from a total of
110,000. However, while this meant that large plants would not be closed,
smaller plants were vulnerable.
In March, Hilmar Kopper, Deutsche BankÕs chief executive announced his
support for the bid because it made ``good industrial sense''. Deutsche BankÕs#p#分頁標題#e#
nominee, Herbert Zapp, was chairman of HoeschÕs supervisory board and
Kopper stated that Zapp had no obligation to defend Hoesch against the bid.
It was unusual for a bank to support a merger which had been so vehemently
opposed by the head of the management board.
In March 1992, Krupp announced that it was to be transformed into an AG.
In June, Hoesch accepted KruppÕs proposal to merge at its AGM and voted to
abolish the voting right restriction. 99.7% of shareholders agreed to the merger.
Gerhard Cromme of Krupp and 3 others were voted onto the supervisory
board, and it was agreed that Neukirchen would leave the company with a
payment of DM 6 million in November 1992.
In July, Hoesch announced it was to become part of Fried, Krupp AG
Hoesch-Krupp. However, by November 1992 the merger had not taken place
because of complaints by three small shareholders to the courts. The court
action prevented the immediate registration of the merger. One of the complainants,
a former employee of Hoesch, objected to the name of the new
company. The other shareholders objected to the treatment of minorities.
Subsequently, objections to the merger were dropped and the merger became
e?ective on 8 December.
1392 J. Franks, C. Mayer / Journal of Banking & Finance 22 (1998) 1385±1403
5. The bid for Continental AG
5.1. Background to the bid
In 1989, Continental and Pirelli were the fourth and ®fth largest tyre producers,
respectively, each with approximately 8% of world market share;
however, together their turnover fell short of that of either Goodyear or Michelin.
At the end of the 1980s serious overcapacity emerged in the tyre industry
which necessitated consolidation. Continental purchased General Tyre
of the US in 1987 and, in 1988, Pirelli competed unsuccessfully with Bridgestone
for ownership of Firestone Tyre of the US. In September 1990 Pirelli
announced a proposal to merge with Continental.
5.2. The bid
Pirelli proposed a Ôreversed acquisitionÕ. Continental was to acquire the tyre
business of Pirelli through Pirelli Tyre Holding of Amsterdam for between DM
1.8 billion and DM 2.2 billion. Half of this was to be ®nanced by a rights issue
amounting to an increase in ContinentalÕs capital of between 50% and 60%; the
remainder was to be raised through borrowings. Pirelli was to use the payments
to increase its share stake in Continental to nearly 28%. Combined with the
23% shareholding held by investors who supported the bid, Pirelli would then
have a controlling holding.
Pirelli viewed its approach to Continental as a friendly merging of interests
in di?cult market conditions. However, Continental rejected the o?er describing
it as a ``hostile takeover with several serious defects''. In particular, it#p#分頁標題#e#
argued that the price proposed by Pirelli for its Tyre HoldingÕs assets was well
above ContinentalÕs estimate of its value of DM 800 million. Continental also
argued that the borrowings would take PirelliÕs gearing to unacceptably high
levels, and that Pirelli had overestimated the gains to the merger at DM 400
million over four years.
While rejecting the bid, Continental left open the possibility of negotiations
occurring at some future date. However, as a precondition it required Pirelli to
give an undertaking that no trade in Continental shares occur for two to three
years and no con®dential or sensitive ®nancial information be disclosed while
talks were underway. Pirelli refused to give such an undertaking thereby
con®rming ContinentalÕs view that the bid was hostile.
In December 1990, a private shareholder, Mr. Alberto Vicari, led a small
group of Continental shareholders controlling about 5% of Continental shares
in calling for an extraordinary general meeting to resolve uncertainty surrounding
the merger and to give shareholders the opportunity to express their
views about the merger. Two sets of motions were proposed. The ®rst favoured
continued independence: they involved raising the majority required to
J. Franks, C. Mayer / Journal of Banking & Finance 22 (1998) 1385±1403 1393
overturn the 5% voting right restriction from 50% to 75%, raising the majority
required to dismiss supervisory board members from 50% to 75% and requiring
a 75% majority to sell important parts of the business. The second set of
proposals opened the way for a merger: the voting right restriction would be
eliminated and the management required to prepare a merger with Pirelli for
the next annual meeting. At the 1989 AGM a proposal to remove the voting
restriction had been rejected by a very slender margin of 2%.
The Continental supervisory board supported the management board in
opposing the bid. They recommended that shareholders vote against the resolutions
with the exception of the one requiring a 75% majority to eliminate the
5% voting right restriction. They recommended that shareholders vote against
the resolutions requiring 75% majorities for dismissal of members of the supervisory
board and disposal of large assets on the grounds that a minority
shareholding would then potentially be able to prevent desirable changes in the
board and the companyÕs corporate strategy.
However, the supervisory board, in particular its chairman, Ulrich Weiss
from Deutsche Bank, did not oppose the concept of a merger. Weiss had close
associations with several Italian ®rms: he was Deutsche BankÕs board member
at Fiat, the Italian car maker closely linked to Pirelli, Chairman of Banca
dÕAmerica e dÕItalia and a member of the board of the Fiat automotive group.#p#分頁標題#e#
Through one of these associations he met Leopoldo Pirelli, the Chairman of
Pirelli, and early in 1990 Leopoldo Pirelli proposed cooperation between the
two companies.
There were large share transactions prior to the EGM in March 1991. These
created two groups of shareholders: supporters of Pirelli and Continental. As
at March 1991, disclosed share stakes were as follows.
The suspicion arose that PirelliÕs supporters were acting in concert and
thereby violating the 5% voting right restriction. Just before the meeting,
Continental wrote to its principal shareholders asking them to declare the
amount held, whether they were acting as bene®ciaries for other parties and
whether they had entered into indemnities to reimburse any losses or expenses
of third parties. Pirelli strenuously denied any such schemes.
Pirelli supporters Continental supporters
Allianz Group 5% BMW 5%
Italmobiliare 3% Daimler-Benz 5%
Mediobanca 5% Deutsche Bank 5%
Nord. Landesbank 3% Volkswagen 5%
Pirelli Verwaltungs-Gesellschaft 5%
Sopaf (Italy) 5%
Source: Financial Times 29/7/90.
1394 J. Franks, C. Mayer / Journal of Banking & Finance 22 (1998) 1385±1403
At the shareholdersÕ meeting on 13 March, 66% of shareholders voted in
favour of rescinding the voting right restriction. The rule requiring Continental
to commence negotiations with Pirelli was defeated. However, elimination of
the voting right restriction could not be implemented. In April, two lawsuits
opposing the elimination of the voting right restriction were lodged ± one of
these was a class action on behalf of minority shareholders. The concern seemed
to arise that elimination of voting right restrictions left minority shareholders
exposed to the accumulation of share stakes and of discriminatory pricing.
On 3 May, in exchange for Pirelli relinquishing their demand for two places
on the supervisory board, to which their shareholding entitled them, the supervisory
board of Continental voted unanimously (with one abstention) to
request the resignation of the Chief Executive O?cer, Horst Urban. The supervisory
board wanted negotiations on cooperation with Pirelli to proceed in
a constructive way and UrbanÕs presence was viewed as an impediment to this.
Wilhelm Winterstein, the longest serving member of the Continental board,
was elected as a temporary replacement for Urban and in July he was succeeded
by Hubertus von Grunberg. In June 1991, Pirelli responded by replacing
the managing director of Pirelli, Gianbattista De Giorgi; he was felt to
be taking too aggressive a stance on the merger to promote constructive negotiations.
Thereafter negotiations over cooperation between Pirelli and Continental
proceeded steadily. There were discussions about joint purchasing, warehousing#p#分頁標題#e#
and R&D activities. Talks were organised at two levels: strategy was
formulated by top management and detailed implementation discussed by
heads of divisions. Impetus was given to the negotiations by the worsening
®nancial conditions of the two companies.
As late as the end of October 1991 reports appeared suggesting that cooperation
pacts were about to be signed before the end of the year but on 1
December the talks collapsed in disarray. PirelliÕs shares plunged 25% and it
emerged that Pirelli had o?ered its Italian allies indemnities for losses sustained
contingent on no agreement being reached on a merger by 30 November.
PirelliÕs allies had been acting in concert and thereby violating the 5% rule. The
indemnities cost Pirelli L 350 billion and considerable loss of face. On 17
February 1992, Leopoldo Pirelli resigned and management control was vested
in a seven-member executive committee.
6. Analysis of the bids
6.1. Bid premia and takeover regulation
Table 1 records abnormal share price returns measured relative to the
DAX of the three targets from nine months before to nine months after the
J. Franks, C. Mayer / Journal of Banking & Finance 22 (1998) 1385±1403 1395
Table 1
Cumulative abnormal returns around the announcement of the three hostile takeover bids
Target 9 months
before
3 months
before
1 month
before
Week of the announcement
1 month
after
3 months
after
9 months
after
Hoesch AG 16.5% 3.7% )12.8% )10.8% 0.1% )12.2% )2.3%
Continental
AG
)1.2% 2.1% )9.5% )7.7% 4.2% )13.0% )39.6%
Feldmuhle
Nobel AG
1st bid 19.9% 12.0% 8.3% )0.3% 0.4% )15.7% )20.4%
2nd bid 15.0% 12.6% 7.4% 2.6% )2.5% )8.2% 9.7%
1396 J. Franks, C. Mayer / Journal of Banking & Finance 22 (1998) 1385±1403
bids. Table 2 reports dividends per share of the three companies. In two out
of three of the bids (Feldmuhle Nobel and Hoesch) there is no evidence of
poor ®nancial performance prior to the bids. There were positive share price
movements over nine months prior to the announcement of bids and dividends
per share increased or remained unchanged. In the case of Continental,
there was little overall movement in its share price over the nine months prior
to the bid but there was a reduction in dividends per share in the year of the
bid, 1990.
There were negative share price reactions to the announcement of the bids
for Continental and Hoesch both in the week of the announcement and in the
month leading up to the announcement of the bids. Neither bid was opposed
and, at least in some respects, was supported by Deutsche Bank. In contrast,
the one bid which was resolutely opposed by Deutsche Bank, the attempted
acquisition of Feldmuhle Nobel by the Flick brothers, displayed positive share#p#分頁標題#e#
price movements in the month prior to the bid although little movement in the
week of the announcement. In no case were bid premia commensurate with
those in the UK or the US, between 20% and 30%.
The share price movements suggest marked di?erences between prices paid
for blocks of shares by the Flick brothers and Krupp and those accruing to
other minority shareholders. For example, shareholders in Hoesch who sold
out prior to the announcement of the takeover received a bid premium of up to
16.5% whereas those who waited for the announcement received a negative bid
premium of up to )12.8%. Such price discrimination would not be possible in
the UK where the Takeover Code requires bidders to o?er a price at least as
high as that paid to target shareholders at any time in the previous prior 12
months. There was no equivalent rule in Germany during this period. There are
also marked di?erences between Germany and the UK concerning two tier
o?ers. In the ®nal bid for Feldmuhle Nobel by Stora, the bidder o?ered to
purchase an 85% stake at a price of DM 567 and DM 540 to small shareholders
for the outstanding 15%. In the UK, two tier o?ers are not permitted.
The low bid premiums o?ered may have been a?ected by the ability of the
bidder to acquire substantial stakes in the market surreptitiously. In Germany,
Table 2
Dividends per share (DM)
Target 1985 1986 1987 1988 1989 1990 1991 1992
Hoesch AG 5 5 5 8 10 10 n.a. n.a.
Continental
AG
5 6 7 8 8 4 0 0
Feldmuhle
Nobel AG
10 10 10 10 10 12 30 n.a.
Source: Annual reports.
J. Franks, C. Mayer / Journal of Banking & Finance 22 (1998) 1385±1403 1397
a company is required to disclose a stake only when it exceeds 25% of the
outstanding equity capital. Furthermore, it is unclear whether the purchase of
convertibles or options to purchase other partiesÕ stakes are included in the
threshold. Krupp acquired 24.9% of HoeschÕs equity before announcing their
stake. They also purchased shares in the convertible and secured the support of
other shareholders before the bid so that when the bid was ®rst announced they
had control over a majority of the shares. In contrast, in the UK, company
legislation prescribes a disclosure threshold of 3% of a companyÕs equity and
the Takeover Code requires a company to include convertibles and options in
calculating this threshold.
Rules regarding voting restrictions and concert parties could also have affected
bid premia. The incumbent management of Feldmuhle Nobel thwarted a
probable tender o?er by the Flick brothers by subjecting their stake of 36.5% to
a 5% voting restriction. Pirelli attempted to circumvent ContinentalÕs 5%
voting right restriction by encouraging allies to purchase blocks of 5% each and#p#分頁標題#e#
guaranteeing them against loss. Such a concert party is not illegal but is subject
to the voting restriction determined by the German courts. As a consequence,
when the concert party was revealed, it was deemed to have violated the rules
under which the extraordinary General Meeting voting was conducted. In the
UK the former chairman of Guinness was sentenced to a jail term for giving
guarantees against losses sustained by individuals who promised to vote in
GuinnessÕ favour.
The ability to discriminate between shareholders means that changes in
control can be engineered in Germany at lower gains to shareholders than in
the UK; this makes takeovers in Germany cheaper than in the UK and means
that restructurings can be organized at lower cost. The failure of the recent bid
by Krupp for Thyssen in which it was stated that Krupp would o?er all target
shareholders the same price may re¯ect the recent growth of shareholder activism
and changes to takeover rules in Germany.
6.2. The power exerted by the banks
Table 3 summarizes the methods, motives and outcomes of the three bids.
Table 4 summarizes the bid tactics and defences which the companies employed.
Table 5 records the in¯uence exerted by the banks and their attitude to
the bids. Deutsche Bank exerted considerable in¯uence on all three bids. In
Feldmuhle Nobel, Deutsche Bank played a key role in introducing a voting
right restriction which prevented the Flick brothers from seizing control. The
voting right restriction not only prevented the Flick brothers from gaining
control, it also prevented control by Veba AG despite the fact that they acquired
a majority holding and were favourably inclined to the management of
Feldmuhle Nobel. Our interpretation of the BankÕs conduct was that it did not
wish to see the break-up of a company which it had so recently brought to the
1398 J. Franks, C. Mayer / Journal of Banking & Finance 22 (1998) 1385±1403
Table 3
Summary of three hostile bids
Target Date Bidder Industry Method Motive for bid Outcome
Feldmuhle
Nobel
June
1988
Group of investors
headed
by Flick
brothers
Paper, chemicals
and explosives
(1) Accumulation
of sharestakes
(36.5%). (2) Purchase
of stake by Veba.
(3) Sale to Stora
Changing management
Flick brothers sell
stake to Veba who
is unable to exercise
control (September
1990)
Hoesch October
1991
Krupp Steel and
other steel
based activities
(1) Accumulated
stake of 24.9% by
Krupp. (2) Merger
negotiations
Industrial re-structuring,
previous
failed attempts at
friendly mergers
Krupp gains managerial
control.
Merger held up by#p#分頁標題#e#
courts (July 1992)
Continental September
1990
Pirelli Tyres (1) Merger proposal.
(2) Bidder stake.
(3) Concert party.
(4) Merger negotiation
Industrial restructuring
Bid fails (December
1991)
J. Franks, C. Mayer / Journal of Banking & Finance 22 (1998) 1385±1403 1399
market and it did not believe that there was a need for the bid on industrial
restructuring grounds.
The bids for Hoesch and Continental illustrate the limitations on the control
conferred by proxy votes. In Hoesch, Deutsche Bank installed the head of the
management board 3 months prior to the bid by Krupp. However, it was
Table 4
Bid strategies and bid defences
Target Bid strategy Bid defence
Feldmuhle
Nobel
Ownership ± Covert purchase of
38.5% stake by Flick brothers et
al. Veba purchases Flick stake and
acquires 51%
Ownership ± Use of proxy votes by
``housebank'' to introduce voting rights
restriction
Voting Restrictions ± Limits ability
of Veba to take management
control
Voting Restrictions ± 5% voting rights
restriction introduced after stake purchased
by Flick brothers
Supervisory Board Control ±
Veba takes 2 seats on supervisory
board including Chairman
Supervisory Board Control ± No action
Hoesch Ownership ± Covert purchase of
24.4% ± Control over additional
30.4% ± Stake increased to 62% to
give Krupp majority ownership of
merged company
Ownership ± Large shareholder ± West
LB does not support bid ± Discussions
with British Steel on blocking minority
Voting Restrictions ± No action Voting Restrictions ± Restriction of
15%
Supervisory Board Control ±
Krupp takes 2 seats on supervisory
board including Chairman
Supervisory Board Control ± Attempt
to extend co-determination to Krupp's
steel interests
Continental
Ownership ± Pirelli acquires 5% ±
Allies acquire 22% ± Pirelli indemni
®es allies for potential losses
± Purchases dilute proxies held by
banks
Ownership ± Allies (BMW, Daimler-
Benz, Volkswagen) purchased shares
Voting Restrictions ± Pirelli attempts
to remove restriction
Voting Restrictions ± 5% restriction
(1990) Shareholder proposed increasing
majority required to overturn voting
restriction to 75% ± Accused Pirelli and
associates of acting as a ``concert party''
Supervisory Board Control ± Fails
to gain any seats on supervisory
board
Supervisory Board Control ± No action
1400 J. Franks, C. Mayer / Journal of Banking & Finance 22 (1998) 1385±1403#p#分頁標題#e#
unable to prevent the bid from taking place because, by covertly acquiring
shares in the market, Krupp was able to diminish the proxy votes which the
banks could cast. In Continental, the Chairman of the supervisory board engaged
in preliminary discussions about the bid and played an important role in
replacing the chairman of the management board to facilitate the acquisition.
But here too, the signi®cance of the banksÕ proxies in Continental was diminished
by the acquisition of share blocks by supporters of both Continental
and Pirelli.
There is an explanation other than self-interest or impotence for the apparent
failure of banks to protect the interests of minority shareholders.
Support for the bids by Pirelli for Continental and by Krupp for Hoesch was
justi®ed by the required rationalizations of the European tyre and the German
steel industries, respectively. 1 These rationalizations imposed considerable
costs on workers in the two companies and these would have been still higher if
1 The recent support of the banks for the attempted acquisition of Thyssen by Krupp was also
justi®ed by a continuing need for rationalization of the German steel industry.
Table 5
Position and in¯uence of banks
Bank's in¯uence
with company
Attitude to bid Action
Feldmuhle Nobel Deutsche Bank
chaired supervisory
board; held stake
of 8% Voted large
number of proxies
Opposed Flick
brothers
Supported voting
right restriction
Hoesch Deutsche Bank
chaired supervisory
board of Hoesch;
had 3 months previously
installed
new head of management
board
Deutsche Bank
supported bid even
though opposed by
head of management
board
None
Continental Deutsche Bank
chaired supervisory
board; voted large
number of proxy
votes; closely involved
in shaping
attitude to the bid
Opposed bid but
supported merger
talks
Opposed removal
of voting right restriction;
dismissed
head of management
board when
he opposed merger
talks
J. Franks, C. Mayer / Journal of Banking & Finance 22 (1998) 1385±1403 1401
large bid premia had been paid. Banks may therefore have been balancing the
interests of di?erent parties in moderating the gains to shareholders.
7. Conclusion
This paper has examined the role of German banks in hostile takeovers in
Germany. It has considered the degree of control exercised by banks and
whether they acted in shareholdersÕ, their own or some other partyÕs interests.
It has also examined the role of regulation in a?ecting returns to shareholders.
The paper has found that banksÕ in¯uence derives from their chairmanship#p#分頁標題#e#
of supervisory boards and the proxy votes which they can cast on behalf of
individual shareholders. As the Feldmuhle Nobel case illustrates, proxy votes
provide banks with the means to protect minority shareholders. When combined
with voting right restrictions, they can require acquirors to purchase
nearly all of the shares of a target in the market before gaining control. The
fact that the banks did not succeed in securing a bid premium in the bids for
Continental and Hoesch suggests that they did not place much weight on
shareholder interests. Whether they were in fact frustrated in their attempts to
protect minority shareholders by market purchases of proxy votes by the
predators or whether their actions re¯ected self-interest or those of other
stakeholders is di?cult to determine.
An alternative explanation is that the low bid premia resulted from ine?ective
regulation. The ability to price discriminate over time and between di?erent
shareholder groups and to accumulate substantial share blocks without having
to disclose them all reduce the cost of acquisition. However, they also mean that
minority shareholders stand to earn low or negative returns from takeovers.
Acknowledgements
This paper is part of an ESRC funded project on ``Capital Markets, Corporate
Governance and the Market for Corporate Control'', no. W102251103.
It has bene®ted from interviews with numerous individuals in Germany and
UK. We are particularly grateful to Gerhard Hablizer of Commerzbank, Ellen
Schneider-Lenne and Ulrich Weiss of Deutsche Bank, Charles Lowe of
Deutsche Bank UK, Dr. Klaus Christian Hubner and Berndt Jonas of Fried.
Krupp AG, Dr. Peter Krailic of McKinsey, Peter von Elten and Andreas
Buddenbrock of JP Morgan, Greg Morgan of Munger, Tolles & Olson,
Malcolm Thwaites of Morgan Grenfell, Dr. Kiran Bhojani and Dr. Wilhelm
Heilmann of Veba AG, Michael Treichl of Warburg, Nicholas Weickart of
Weickart, Simon, & Westpfahl, Hans Peter Peters and Andrew Neumann of
WestLB.
1402 J. Franks, C. Mayer / Journal of Banking & Finance 22 (1998) 1385±1403
References
Franks, J., Mayer, C., 1997. Ownership, control and the performance of German corporations.
Mimeo.
Gottschalk, A., 1988. Der Stimmrechtsein¯uss der Banken in der Aktionarversammlungen der
Grossunternehmen, 5 WSI-Mitteilungen, pp. 294±298.
http://www.400190.com/dissertation_writing/Finance/Jenkinson, T., Ljungqvist, A., 1996. Hostile takeovers and the role of banks in German corporate
governance. Mimeo.
Nibler, M., 1998. Bank control and corporate performance in Germany: The evidence. Ph.D.
thesis. University of Cambridge.
J. Franks, C. Mayer / Journal of Banking & Finance 22 (1998) 1385±1403 1403#p#分頁標題#e#