- Consulting with Canadians -

Power Financial Corporation Submission in Response to Finance Canada's Large Bank Mergers in Canada:

---

Bank/Insurance Mergers Outside Canada: The Lessons For Public Policy

Study prepared by
Davis International Banking Consultants
London, England

December 2003

Related documents:

• Power Financial Corporation Submission to Minister of Finance Future Structure of The Canadian Financial Services Industry
• Concentration and competition in the Canadian Banking industry

---

Note: This study was prepared by Davis International Banking Consultants (DIBC) solely for the use of our clients. It should not be relied upon by any third party without DIBC’s prior written consent.

Executive Summary

Over the past decade Canada has evolved a unique but effective public policy for regulating the interaction between banks and insurers. Additional integration in the banking pillar has been denied since the Minister of Finance’s decision in 1998 against two proposed bank mergers. The 1999 White Paper requires the two major demutualised insurers to remain independent so as to maintain a strong insurance pillar. Tight restrictions on the sale of insurance products to a bank client base have been maintained since their inclusion in the 1992 Bank Act.

Our task in this study is to evaluate the experience of other developed markets – specifically Europe, the US, South Africa and Australia – to determine whether there are good reasons for changing this policy by permitting cross-pillar bank/insurance mergers.

This study will demonstrate that circumstances have not changed since the past review, either in Canada or in these other markets, which would argue for such a policy change. In fact, the arguments against allowing cross-pillar mergers are even stronger now than in 1999.

While each national insurance market has its own differentiating factors, our exhaustive study of the evidence in these markets points up three possible negative consequences of such mergers which might be relevant for the Canadian market.

First, a large number of cross-pillar mergers abroad have not only failed to achieve their objectives but also actually produced a threat of contagion to their partners and the national financial system. Profound cultural and structural differences have severely limited planned synergies and led to a large number of subsequent divestitures. Our analysis of the literature and events on the ground indicate that the high point of practitioner interest in these mergers was reached in 1999-2000, and the tide has ebbed since then.

Second, the growing concentration of economic and financial power among a decreasing number of large financial conglomerates has posed significant issues of systemic vulnerability, customer choice and control for many national governments. In such concentrated markets in Australia, South Africa, UK and other European Union countries, regulators have increasingly responded by banning, implicitly or explicitly, further large mergers.

Finally, the goal of a strong, independent life insurance pillar in Canada could be seriously threatened by breaking down the barriers between the life and banking sectors. Experience in strong bancassurance markets like Spain, France and the Netherlands indicates that the health of the broker and agency sales forces, as well as the viability of traditional life insurers, has been undermined by the massive shift to bank distribution. Cross pillar mergers have also diminished contestability in concentrated markets by reducing the number of insurer-owned banking affiliates.

2.0 Introduction

2.1 Background

This study by Davis International Banking Consultants (DIBC) (for profile of DIBC see appendix C) is submitted in response to the Government’s Response of June 23, 2003 requesting comments on a change in the present policy towards mergers between large banks and demutualised insurers in Canada.

In December 1998 Finance Minister Martin determined that the public interest would be the principle criterion for deciding whether to permit large bank/bank mergers. In 1999 the White Paper, Reforming Canada’s Financial Services Sector, essentially applies to the larger demutualised insurers the widely held rule applied to the large banks to prevent their take-over. To quote the White Paper (page 34),

‘As with the widely-held banks, these companies cannot be acquired. Having these large demutualized companies widely held will help ensure the maintenance of a strong insurance sector.’

The White Paper also did not adopt the recommendation of the Task Force on the Future of the Canadian Financial Services Industry to lift the restrictions on the sale of insurance products by banks, thus sustaining the 1992 Bank Act limitations which have essentially constrained such sales for banks.

The purpose of our study is to evaluate the experience of cross-pillar mergers between banks and insurers in other developed markets, in particular since the White Paper was issued, to determine whether there is a public policy argument in favour of a change in this regulatory regime.

2.2 Methodology and Preliminary Findings

After a preliminary review of the literature and study of markets outside Canada, we determined that the US, Western Europe, Australia and South Africa are the most relevant for purposes of comparability with the Canadian market.

We therefore carried out the following research:

• an extensive survey of the published literature on bank/insurance mergers outside Canada, including statements by regulators, industry groups, competition authorities and consumer groups, as well as reports by management consultants, academics, journalists and financial analysts. A bibliography of our research is attached as Appendix A. In the text below, references are made to the document from which the text is taken.
• an analysis of the available data base on such mergers and their outcome. Such data is derived from consultant, academic, financial analyst and company sources. The text below provides attribution where appropriate for specific charts and graphs.
• a series of about 30 in-depth interviews across Europe, South Africa, Australia and the US with selected bank/insurance groups as well as regulators, consultants, industry associations, analysts and other industry experts. The body of the study reflects the views from many of these expert interviewees.

For background information, we attach as Appendix B a brief summary of the history of bank/insurance linkages in the four markets as well as the regulatory response to these mergers. Key findings and conclusions from this history and regulatory action compose a major element of our study.

A central preliminary finding for this research is that each of the four markets noted above – including components of the European Union – has developed both a unique financial industry structure as well as regulatory response to the issues raised by the merger process.

Thus Canada is unique among major markets for its Insurance Business (Banks) Regulations in the Bank Act, which severely restrict both the range of insurance products sold in banks and the conditions under which they can be marketed. By the same token, Canada is one of the few markets which has a thriving, independent insurance sector. Clearly there are different means and structures to satisfy the relevant needs of public sector regulators, consumers, companies and stockholders.

What is important for our study, however, is whether any aspects of the international experience are relevant for policy makers in Canada. To the best of our ability, we have attempted to extract what we regard as significant issues from this experience which should be considered by Canadian policy makers as they assess the arguments for and against regulatory change.

2.3 Our Conclusions

We believe that there is no obvious public policy reason for changing the current regulatory structure as regards bank/demutualised insurance mergers. Our conclusions can be summarised in the three key issues discussed in the balance of this study.

3.0 Failure of Bank/Insurance Mergers to Achieve Objectives – and Possible Resulting Threats of Contagion

In the early 1990s many financial institutions and their stockholders in Europe and Australia embraced bank/insurance mergers with three strategic objectives in mind:

• offering a full range of financial products to their clients;
• increasing their capital base for strategic reasons;
• achieving revenue and cost synergies in the marketing of retail financial products.

Figure 1 provides a listing of the major cross-pillar mergers in the European market, where the enthusiasm for such bancassurance deals has arguably been strongest.

Figure 1: Leading European bank-insurance mergers and acquisitions (a)

Dominant banks	Insurance partners	Country	Size*	Subsequent sales
KBC  (Kredietbank & CERA)	ABB	Belgium	233
Dexia	DVV Insurance	Belgium	368
Rabobank	Interpolis	Netherlands	393
SNS	Reaal	Netherlands	37
SEB	Trygg-Hansa	Sweden	141	Trygg-Hansa non-life  (sold to Codan)
Handelsbanken	SPP	Sweden	144
Danske Bank	Danica	Denmark	247	Danica non-life  (to TopDanmark)
Nordea (Unidanmark) Nordea (CBK)	Tryg Baltica Vesta	Denmark Norway	67 26	Tryg Baltica & Vesta non-life  (to Tryg Baltica foundation)
Sparebanken NOR	Gjensidige	Norway	36	Gjensidige non-life not included
DnB	Vital	Norway	55
DnB	Storebrand	Norway	55	Failed acquisition attempt
Credit Suisse	Winterthur	Switzerland	689
Deutsche Bank	Deutscher Herold	Germany	795	Deutscher Herold  (to Zurich)
Lloyds Bank Lloyds TSB	Abbey Life Scottish Widows	UK UK	334‡	Abbey Life salesforce  (to Allied Dunbar)
Abbey National	Scottish Mutual Scottish Provident	UK UK	284
Halifax	Clerical Medical	UK	512‡
NatWest	Legal & General	UK	649‡	Failed acquisition attempt

Dominant insurers	Bank partners	Country	Size*	Subsequent sales
Fortis (Groupe AG) Fortis (Amev)	ASLK-CGER Generale Bank VSB	Belgium Belgium Netherlands	404 404‡
ING (Nationale- Nederlanden)	NMB Postbank	Netherlands	500
Sampo	Leonia	Finland	19†	Non-life business (sold to If)
Swiss Life	Banca del Gottardo	Switzerland	9†	currently attempting to sell Gottardo
Allianz	Dresdner Bank	Germany	434
AMB	BfG	Germany	29†	BfG (to Crédit Lyonnais)
GAN	CIC	France	–	CIC (to Crédit Mutuel)
Axa (UAP)	Banque Worms	France	5†	Banque Worms (to Deutsche Bank)
Axa	Banque Directe	France	–
Irish Life	Irish Permanent	Ireland	36
Note: former names are shown in brackets (a) includes two failed merger attempts ‡ now part of larger group whose assets are shown * total assets, US$ billion at end 2002 (source: The Banker) † bank assets only

While several of the mergers have been widely viewed as successful in achieving these objectives, a large number have not, as we discuss below.

3.1 Cultural Differences And Operational Synergies

Underpinning these failures has been a cultural incompatibility which undermines collaboration and synergies throughout a bank/insurance conglomerate. This incompatibility is best articulated in Figure 2, which was reproduced from a Deutsche Bank presentation in the early 1990s justifying the decision not to acquire an insurer at the time. In Figure 1 we have listed 18 instances in which European banks have acquired insurers; of this total, seven have subsequently involved major divestitures. In addition, we have listed 12 banks bought by insurers, of which subsequently four have been sold or otherwise disposed of by their acquirers. In Figure 3 below, we analyse the divestitures of life insurance businesses by eight banks in Europe. Figure 4 lists nine problem bank-insurance combinations in Europe, and the accompanying text analyses several transactions in more detail.

Figure 2: Cultural differences between insurers and banks

Banks	Insurers
Sale by the institution	Sale by individual
Branch offices are expensive	Intermediary earns his own money
Daily and personal management	Management by production
Reactive	Proactive
Short, frequent client contacts	Long, infrequent client contacts
Good information on clients	Little information on clients
Fixed working hours	Flexible working hours
Salaried employees	Commission-based employees
Problem solvers	Product sellers
Standardised sales approach	Individual sales approach
Positive image	Doubtful image
Source: Deutsche Bank

 

Our conversations with individual financial conglomerates in Europe, South Africa and the US have confirmed that these differences range from setting corporate policy to marketing operations at the client level. Our research indicates, for example, that such awareness might have been a factor in the Citigroup decision to divest itself of the property and casualty unit of the Travelers group.

One of the consequences of this cultural divide is the difficulty in practice of achieving the cross-selling synergies advertised as the basis for a cross-pillar merger. On the surface, the lack of overlap between the bank and insurance customer bases of such a conglomerate offers outstanding potential for cross-selling. In practice, however, even the most experienced European conglomerates such as Fortis and KBC in Belgium, and ING in the Netherlands, have struggled to blend the different cultures in their cross-selling efforts. The result has often been a de facto separation of functions: the bank staff sell simple insurance-wrapped savings and "tick-the-box" products directly linked to loans, with referrals being made to their insurance counterparts for the more complex services. When the bank merges with a traditional insurer, the standard solution is to keep the two distribution channels completely separate to avoid ‘contamination’ of the different delivery channels.

ING Group is one of the world’s largest and most international banking and insurance groups, which has been particularly successful with ING Direct, a telephone-and-internet bank now operating in eight major countries. ING was formed in 1991 when Nationale-Nederlanden, the country’s largest insurance group, acquired NMB Postbank, the third-largest bank. At the time the main purpose was seen as creating a group with sufficient resources to expand successfully internationally. There was little or no expectation that the two businesses would sell each other’s products. This has subsequently been tried, but without much success.

Moreover, the merger outraged the Dutch brokerage community, which accounted for some two thirds of Nationale-Nederlanden’s insurance sales. They threatened to stop selling its products if it tried to use its banking outlets to compete with them. The issue was resolved, as it has often subsequently in other mergers, by a compromise which tailors product features to the particular distribution channel.

While bank staff have been relatively successful in marketing simple retail life products, the reverse has not been true. Across Europe the results of insurance agents selling bank products have been disappointing. A major issue is compensation: the selling commission on banking products is simply inadequate for an insurance sales force in comparison to their traditional product line.

Hard data on the actual cost and revenue synergies from European cross-pillar mergers is quite thin. Based on projections made at the time of the mergers, cost savings have never exceeded the 5% of combined costs that was promised by Gjensidige NOR in Norway. For the much larger and more strategic acquisition of Dresdner Bank by Allianz, the dominant German insurer, we calculate projected revenue synergies – the strategic objective of the US$24 billion deal – at 2.9% of the pre-merger total, with cost synergies of only 1.0%. On balance, merger experience would indicate that such potential synergies are easily dwarfed by the direct and indirect costs of a complex merger.

3.2 Subsequent Divestitures

As indicated above, roughly one-third of the European cross-pillar acquisitions and mergers listed in Figure 1 have resulted in a subsequent divestiture, either partial or total. In some cases, particularly for non-life businesses, such a divestiture may have been contemplated in advance, given the widespread enthusiasm for buying a life insurer’s assets under management and the relative profitability of the life business.

But the number of life insurance businesses ultimately divested is a serious indication of disillusionment on the part of management with their new cross-pillar acquisition. It is highly relevant for Canada, which wishes to maintain a strong Canadian-owned insurance pillar. Finding a buyer for such a divestiture may well require a foreign buyer or disposing of what could be perceived as damaged goods.

Figure 3 provides a profile of eight major European insurance businesses divested by banks. Several, such as the three UK banks (Barclays, Alliance & Leicester and Royal Bank of Scotland) and ABN Amro, involve the sale of an in-house insurance business to an insurance partner as a result of a decision to outsource the product. BBVA in Spain sold two of its insurance affiliates to foreign buyers, Norwich Union and AXA, in favour of an in-house supplier.

Of much greater significance is the divestiture of Deutscher Herold, a major German insurer, by Deutsche Bank. Deutsche Bank established its own life insurance company, Lebensversicherungs der Deutschen Bank, towards the end of the 1980s. In 1992 it acquired a majority stake in Deutscher Herold, a leading life and non-life insurance company, and merged its own company into Deutscher Herold a few years later.

However, by the end of the 1990s, the bank no longer regarded it necessary to own an insurance company. During the merger discussions with Dresdner Bank in 2000, it announced that it would sell Deutscher Herold to Allianz, but that plan was abandoned when the merger fell through.

In 2001 Deutsche Bank sold its 75.9 percent stake in Versicherungsholding der Deutschen Bank, the holding company for Deutscher Herold, to Zurich Financial Services. The sale also included its insurance businesses in Italy, Spain and Portugal, countries where it has significant retail banking operations. The total price was €1.5 billion. Under the agreement, Deutscher Herold will continue to be the exclusive supplier of insurance products to Deutsche Bank’s retail banking clients.

Figure 3: European banks selling own life insurance operations

Bank	Details
Barclays	Barclays, one of the UK’s four largest banks, established its own in-house life insurance subsidiary during the 1980s. Unlike several of its rivals, it never acquired a life company. In 2001 it sold its life business and also its retail mutual funds operations to Legal & General. It now distributes Legal & General branded products through its branches.
Alliance & Leicester	Alliance & Leicester, a former building society, made the same decision as Barclays, selling its in-house long-term savings and investment business to Legal & General in 2001 and then its life business to L&G in 2002.
Royal Bank of Scotland	RBS established a joint venture company with Scottish Widows to manufacture life insurance products for branch distribution. When Scottish Widows was bought by Lloyds TSB, RBS acquired 100% of the joint venture. After acquiring NatWest, and thereby becoming one of the four largest UK banking groups, RBS sold 50% of both its own and NatWest’s life subsidiaries to Aviva, a leading UK independent life group.
National Australia    Bank (UK)	NAB established its own in-house life business in the 1990s to provide products to distribute through its three UK regional banks. In 2003 it closed this business and now distributes Legal & General products.
ABN Amro	ABN Amro, the largest commercial bank in the Netherlands, did not follow all its major rivals into bank-insurance mergers and acquisitions. It set up an in-house life business but in 2003 sold it to a joint venture company established with Delta Lloyd, a local subsidiary of Aviva. This will supply products for distribution through ABN Amro’s branch network.
Banco Bilbao    Vizcaya Argentaria	BBVA, one of the two dominant Spanish commercial banks, controlled traditional insurers Aurora Polar and Plus Ultra. In order to concentrate on its in-house insurer Euroseguros, the group sold Aurora Polar to Axa (in stages between 1992 and 2000) and Plus Ultra to Norwich Union.
Santander Central Hispano	After Banco Santander merged with Banco Central Hispano (BCH) to form SCH, it sold BCH’s 50% stake in insurance company Vitalicio in stages to Italian insurance company Generali. SCH is now concentrating on its own in-house insurer. Vitalicio was a joint venture formed by merging Generali’s Spanish insurance interests with those of BCH.
Deutsche Bank	After establishing its own life subsidiary, in 1992 Deutsche Bank (the largest German bank) acquired control of Deutscher Herold, a leading life and non-life company. However, in 2001 it sold Deutscher Herold and three foreign life businesses to Zurich Financial Services.
Source: DIBC research.

In the US, the most notable divestiture has been the flotation by Citigroup of its non-life business, Travelers Property Casualty, in 2002 following the landmark merger in 1998. Recently the chief executive of Travelers Life & Annuity has suggested that Citigroup might sell his part of the business also.

As this reversal of priorities shows, bank buyers can change their minds about the value of owning an insurance product provider as often as twice in a decade. For public policy in Canada, as we discuss under paragraph 5.0, this in our view is a significant factor in the effort to maintain a sound Canadian life insurance sector.

Other divestitures in Figure 1 reversed the original acquisition of retail banks by European life insurers as part of an effort to ensure distribution. In several cases, the choice of banking vehicle has been unfortunate, to say the least. For example, Aachener & Muenchner, a pioneer in German bank/insurance, encountered asset quality as well as marketing problems by acquiring Bank fuer Gemeinwirtschaft (BfG), finally selling it several years ago to SEB in Sweden. Similarly, the decision to acquire Dresdner Bank proved to be disastrous, at least in the short term, for Allianz – see section 3.3.1 below.

3.3 Major Losses From Bank/Insurance Mergers Threaten Contagion

In the European market, massive losses suffered by a number of bank/insurance conglomerates have threatened both the viability of the individual institution as well as the national financial system through the contagion process. Some of these threats occurred as a result of the collapse of equity values during the 2001-2003 period, which devastated the balance sheets of those life companies which had invested a major portion of their assets in the equity markets. Others were driven by losses in the property and corporate lending markets suffered by both the banking and insurance arms of the group. Still others have involved mis-selling scandals as banking groups have been hurt by the over-aggressive tactics of insurance companies that they have acquired.

The bottom line of many of these substantial losses has been to force some of Europe’s previously well-capitalised financial conglomerates to raise new equity capital on unattractive terms, sell valuable subsidiaries and/or drastically shrink their core businesses to offset operating losses. In markets like the UK and Germany, regulatory authorities were obliged in early 2003 to suspend their solvency rules for the worst-affected insurance companies or take other action to preserve public confidence in the sector.

Thus in the UK the traditional insurance regulations tying equity market values to solvency ratios were suspended following the 2003 market collapse. In Germany, press reports indicated that two thirds of the German life insurers failed to meet stress tests in 2003. Subsequently, Mannheimer Lebensversicherung became the first German life insurer in many years to fail, and tax regulations were eased to permit insurers to adjust their statutory reporting.

Figure 4 lists a number of the recent problem bank/insurance combinations in Europe.

Figure 4: Problem bank-insurance combinations in Europe

Acquirer	Target	Comments
Credit Suisse	Winterthur	Insurer weakened by falls in equity market
Lloyds TSB	Abbey Life	Heavy losses from mis-selling
Lloyds TSB	Scottish Widows	Insurer weakened by falls in equity market
Abbey National	Scottish Mutual	Insurer weakened by falls in equity market
Allianz	Dresdner Bank	Bank weakened by heavy loan losses
AMB	BfG	Poor performance by bank – subsequently sold
GAN	CIC	Poor performance by bank – subsequently sold
UAP	Banque Worms	Bank weakened by heavy loan losses
Swiss Life	Banca del Gottardo	Unsuccessful diversification; substantial decline in value
Source: DIBC Research

The most spectacular case of contagion – in terms of both speed and size – was probably Allianz’s acquisition of Dresdner Bank. However, Allianz, as Europe’s largest insurance company, had sufficient resources to absorb the setback. Banks suffering particularly badly from insurance acquisitions include Credit Suisse, Commonwealth Bank of Australia and Lloyds TSB. In the case of Credit Suisse and Lloyds TSB, they have been forced to sell some of their most valuable foreign subsidiaries in order to recapitalise the group.

3.3.1 Allianz and Dresdner

The acquisition of Dresdner Bank by Allianz in 2001 as a distribution vehicle for retail products has encountered massive asset quality problems which have not only forced out both the chief executives involved in the original transaction but also threatened the viability of Germany’s leading insurance group with roughly 20% of the market.

A year earlier Dresdner – one of Germany’s four largest retail banks – had announced a merger with market leader Deutsche Bank. The proposed deal, which fell through, would have given Allianz about one third of the merged group’s branch-based retail banking business and Allianz would also have acquired its valuable fund management operations. Following the collapse of the Deutsche deal, Dresdner subsequently held discussions with Commerzbank, another of the country’s big four, but these were also abortive.

The acquisition of Dresdner by Allianz was expected to enhance the partners’ strength in asset management and expand the sales of Allianz’s pension and insurance policies through the bank’s network of about 1,200 branches. While this seems to have been reasonably successful, the benefits have been overwhelmed by the reduction in value of Dresdner itself. Moreover, Allianz already sold its policies through the bank’s branches, thanks to its existing 21.7% stake in Dresdner.

In 2001/2 Dresdner plunged into losses thanks to domestic bad debts and a collapse in its investment banking business. It claimed that, "For the USA and Europe, 2001 was the banking industry’s worst year since the Second World War."

3.3.2 Credit Suisse and Winterthur

One of the two dominant banks in Switzerland, Credit Suisse bought Winterthur, a major insurance company, in 1997. The bank already distributed Winterthur’s insurance policies through its domestic branch network but it made the acquisition to protect its position when it appeared that another group might acquire Winterthur.

In 2002 Winterthur plunged into losses and Credit Suisse was forced to provide it with two capital injections totalling CHF2.6 billion (US$2.02billion).

In addition, during 2003 the group has been forced to sell many of Winterthur’s valuable international businesses. These include Churchill in the UK (for £1.1 billion), its operations in Italy (for €1.465 billion) and Republic in the US (for US$0.127 billion).

3.3.3 CBA and Colonial

Commonwealth Bank of Australia (CBA), one of the four dominant banks in Australia, acquired Colonial in 2000 for about A$10 billion (US$7.35billion). Colonial was a financial group with life insurance, fund management and banking interests. As a result of the acquisition, the contribution of life and fund management to CBA’s revenues increased from 3 percent to 36 percent.

CBA is now the third-largest life insurance group in Australia and the largest manager of retail pension and annuity funds. However, in 2003 it was forced to write down the value of its life and fund management operations by A$426 million (US$312.9million), following losses in the life business. Analysts suggested that the write-down should have been closer to A$2 billion (US$1.47 billion).

3.3.4 Lloyds Bank and Abbey Life

Lloyds Bank, one of the UK’s four largest banks, acquired a majority of Abbey Life in 1988, buying out the minority a few years later.

Abbey Life was a successful start-up, using a large salesforce to sell pensions and life insurance through the system of "cold calling" potential customers. Lloyds wished to capture its expertise and sales culture, to boost its own Black Horse life insurance business.

Initially the acquisition appeared to be achieving its objectives, but in the mid-1990s Lloyds made the first of a series of provisions to compensate customers for mis-selling pensions. Over-enthusiastic salesmen had encouraged customers to switch to Abbey Life pensions, generating commissions, although this was not in the customers’ interests. Eventually, the group made provisions of £1,012 million for pensions mis-selling.

In 2002 Lloyds TSB was fined £1 million and forced to set aside £165 million to compensate about 50,000 customers for similarly mis-selling endowment mortgages (where life insurance policies are supposed to pay off the customers' mortgages on maturity, but have increasingly failed to do so). Nearly one third of Abbey Life’s endowment sales will require compensation payments.

In 2003 Lloyds TSB was fined again, £2 million for mis-selling "precipice bonds" (an investment product based on stock market performance) and further heavy compensation payments of £98 million are expected.

Much of the problems at the group can be attributed to the culture of aggressive selling of products to bank customers by commission-driven salesmen. Eventually, the group abandoned this model, and in 1999 paid £7.3 billion for Scottish Widows, a more traditional life company which uses independent financial advisers to distribute its policies. In 2000 Lloyds TSB closed Abbey Life to new business and transferred its salesforce to Allied Dunbar, a subsidiary of Zurich Financial.

Scottish Widows has also proved to be an unfortunate investment, hard-hit by the fall in stock market values which necessitated additional capital from its parent. Sales of its life and pension products through the bank’s branches have been disappointing and recently Scottish Widows’ long-serving chief executive resigned. Reports suggest that it could be sold if it does not start to perform better.

3.4 Impact On Investors Of Bank/Insurance Mergers

While it is not the province of this study to evaluate bank/insurance mergers from a stockholder value standpoint, the wealth of studies by academics and consultants on the subject does provide a perspective which can assist in the formation of public policy.

A plethora of academic and other studies have examined the value added by mergers in general and those in the financial services business in particular. A typical recent effort is that by Boston Consulting Group, which concludes that 61% of the over 300 major mergers in a wide range of industries between 1995 and 2001 actually destroyed stockholder value. While experts can differ on the methodology used to arrive at such conclusions, it is reasonably clear from the research that factors such as overpaying for acquisitions and integration problems have probably destroyed value for a large number of large mergers in recent years.

In the case of bank/insurance combinations, one of the problems in conducting such research from a statistical standpoint is the limited number of actual deals in recent years, compared for example to bank/bank mergers.

However, a recent study by McKinsey & Co investigated 237 acquisitions by US and European insurance companies of other financial institutions between 1990 and 2001. These deals produced an average price decline of 3.0% in the year following the announcement for bank acquisitions against a gain of 8.0% for the purchase of more focused divisions of other insurance companies. Figure 5 summarises the researchers’ findings.

Figure 5: Stockholder value creation by U.S. and European insurers – 1990-2001

 

1. Component of returns not due to marketwide influences; calculated relative to insurance market index.

2. For 237 insurance deals in Europe and the US from 1990-2001:short-term returns include those returns from 5 days before announcement (to capture any information leakage) to 5 days after announcement; medium-term returns include those returns from 5 days before the announcement to 1 year after.

Source: McKinsey & Co.

While the actual number of such bank/insurance mergers is limited for the purposes of statistical analysis, the McKinsey findings are corroborated by a study in 2003 by Beitel et al which uses an event study approach to examine the abnormal (above market averages) return features of ninety-eight large M & A transactions in European banking between 1985 and 2000. The study includes eleven bancassurance deals and other bank/non-bank merger combinations. As in the case of the McKinsey study, they find that ‘non-diversifying transactions create significantly more value than diversifying transactions’, ie bank to bank compared to bank to insurer (page 13).

This academic evidence in favour of non-diversifying mergers is reflected in an apparent investor decline in enthusiasm for such cross-pillar transactions in recent years. The enthusiasm described in Appendix B in Europe in the early 1990s for bank/insurance mergers has waned on the part of both practitioners and the investment community. Furthermore, the trend in markets like the UK, as indicated above, is for strategic alliances rather than mergers to provide the necessary insurance product for a bank retail client base. To use the words of the former CEO of Credit Suisse, ‘we prefer to buy the milk rather than the cow’ – although Credit Suisse later reversed its position and subsequently acquired the Winterthur Group.

Perhaps the most graphic illustration of investor disillusionment with full bank/insurance mergers in recent years was the proposed acquisition in the UK in 1999 of Legal & General by National Westminster Bank. The announcement of the planned merger caused such a steep drop in NatWest’s price that it became the target for two hostile bids and subsequently was acquired and lost its independence.

National Westminster (one of the big four UK banks) agreed to acquire Legal & General, a major life insurance group that had moved successfully into retail fund management. However, investors disliked the terms of the deal and did not trust NatWest to make a success of it. The bank had a long record of making acquisitions that destroyed value for its shareholders, often selling out at a substantial loss.

The collapse in NatWest’s share price allowed Bank of Scotland to make an opportunistic take-over bid for the bank, but it was Royal Bank of Scotland (RBS) that won the ensuing bid battle. The acquisition of NatWest by RBS in 2000 is regarded as one of the most successful financial services deals ever seen anywhere in the world. There is no doubt that it has proved to be much more rewarding for NatWest’s shareholders than buying Legal & General would have been. In contrast to many other UK banks, RBS has chosen not to acquire a life insurance company and it subsequently sold half of NatWest’s home-grown life subsidiary to Aviva, a major independent UK insurer.

The almost complete absence of major bank/insurance acquisitions in the US since the landmark Citibank/Travelers merger indicates that this preference by bankers for distribution alliances rather than ownership of the product is a global trend. Thus a leading retail bank like Wells Fargo has preferred to acquire brokers and agents to become, through its subsidiary Acordia, the largest US bank-affiliated insurance broker.

Another indication of waning interest is the flurry of reports commissioned in 1999-2000 by international bodies such as the Group of Ten and OECD to study the public policy implications of bank/insurance combinations. These studies, listed in the bibliography in Appendix A, typically describe the variety of structural approaches to bancassurance and the

difficulty of quantifying the result of these combinations. Several then project the bancassurance model as the wave of the future, with increasing integration of operating and management functions.

Thus an OECD study in 1999, Convergence in the Financial Services Industry, concludes that:

‘pure cross-selling of banking, insurance and investment products through the same outlet is only a first step, and perhaps even a minor step ... financial institutions will have to examine the whole marketing cycle of their financial products (product development, market development, and to some extent the operations management.’

As indicated above, the actual trend in recent years has been in the opposite direction.

A more measured view is provided by another study from 1999 sponsored by the OECD. Among other things, Financial Services Integration World-wide: Promises and Pitfalls notes that early forecasts projected bank sales of life insurance in the UK at one third of the total, a far cry from the 10-12% actually achieved in recent years. Interestingly, this study points to a possible increase in systemic risk from financial services integration:

‘if financial firms are larger because of integration (a likely result), their now-less-frequent failure might cause greater harm to the financial system’ (page 47)

4.0 Nearing The Limits Of Acceptable Concentration

The growing concentration of economic and financial power among a decreasing number of large financial conglomerates has posed significant issues of systemic vulnerability, customer choice and control for many national governments.

In Australia, South Africa, the UK and other European Union markets, regulators have increasingly responded by banning, implicitly or explicitly, further large mergers. We examine the position in each of these markets below. The US, a far less concentrated market, has imposed the much tougher constraint that no bank should control more than 10% of domestic deposits.

Measuring the actual trend and level of concentration is, however, not a simple task. The fusion of banks and insurers into conglomerates renders the traditional database on bank and insurer market shares redundant. The comprehensive OECD study in 1999 mentioned above thus concludes that:

‘We have been collecting data for more than 15 years now and, although our database has grown over the years, the quality of the statistics is rather poor .. there is a great lack of reliable market information, and the different sources of statistics use different definitions and measure different samples so that comparison is hard to make.’

(page 2)

In the following analysis, however, we attempt to provide some sense of the trend and level of financial concentration in different markets.

The most authoritative study of concentration in banking and insurance in the world’s largest economies was the Report on Consolidation in the Financial Sector, published by the Group of Ten in 2001.

Its data – see Figure 6 – show a very substantial increase in banking concentration in 13 advanced countries during the 1990s, as measured by the market share of the top five banks in each country. Concentration increased in every country except Japan. Canada was the third most concentrated market at the end of the period, one of five countries where the top five banks controlled over 70% of the market.

Figure 6: Combined market shares of top five banks (%)

First date: Italy 1992
Last date: Spain & Switzerland 1997; Germany, UK & Sweden 1998
Source: Group of Ten

The G10 data for life insurance shows a rather different picture, with several countries becoming less concentrated during the 1990s. All three markets where there was already over 60% concentration in 1990 – Australia, the Netherlands and Japan – had experienced a significant reduction by 1998. Although G10 did not publish data for Canada’s position in 1990, it was by far the most concentrated country in 1998. In 2003, with mergers concluded between Sunlife and Clarica and Great West Life and Canada Life, Canada remains the leader in concentration.

Figure 7: Combined market shares of top five life insurance companies (%)

Source

: Group of Ten

4.1 European Concentration

As indicated in Appendix B and Figure 1, a wave of cross-pillar mergers as well as bank/bank deals beginning in the late 1980s and early 1990s has created a limited number of financial conglomerates which dominate many European markets.

Concentration has been particularly marked in the smaller European countries such as the Nordic region, Ireland and the Benelux area, where large local banks and insurers turned to cross-pillar transactions as a strategy to build market share and market capitalisation for both offensive and defensive reasons. Thus in Benelux insurers like the former Nationale Nederlanden (now ING Group) and AG/Amev (now Fortis Group) became much larger financial conglomerates by buying into the banking sector. In the Nordic region, on the other hand, the process was reversed by larger banks acquiring insurers.

Such bank/insurance mergers were justified by the perceived synergies of selling a broader retail product range of bank and insurance products to the combined customer bases of the merging institutions. Less often mentioned were the advantages of building a market capitalisation which enabled them to acquire other financial institutions – in particular outside their limited home markets. A defensive motive was the take-over protection afforded by creating a larger institution with fewer predators.

The result has been the effective merging of the European bank and insurance sectors in competitive terms by the formation of financial conglomerates selling a wide range of retail and corporate/investment banking, life and non-life insurance, fund management and other financial services. Thus in Ireland, Belgium, the Netherlands and Nordic markets, between two and four such conglomerates dominate the financial landscape. In contrast, in the larger markets such as the UK, Germany, Italy and Spain, the presence of savings and co-operative banks as well as other domestic and foreign-owned banks and insurers produces a lower level of concentration.

In the bank/insurance realm, the only reliable data time series we were able to identify is that provided for the Nordic region by Fox-Pitt Kelton, a leading speciality investment bank focusing on the bank and insurance sectors. Figure 8 thus shows the remarkable increase over the period 1990-2001 in the proportion of insurance assets owned ultimately by bank-dominated financial conglomerates. It reflects the acceleration of acquisitions in recent years of insurers by banks in the region to create the current situation in which four conglomerates in Sweden, two in Denmark, and three each in Finland and Norway dominate the national financial sector.

Figure 8: Proportion of insurance assets owned by banks in Nordic countries: 1990-2001

The regulatory response to this rising concentration has been limited to date. In many European Union countries, the view is that the presence of a single currency and single European financial market – at least in theory – justifies higher levels of national concentration than otherwise would be the case.

This reasoning was questioned, however, in 2001 when the European Commission in a landmark decision posed substantive challenges to a proposed merger in Sweden between two of the four financial conglomerates (SEB and Swedbank) which ultimately led to the merger being abandoned. The proposed deal had been approved by the Swedish regulators.

Our understanding from conversations with European Commission regulators is that each merger proposal is judged on its own merits in the context of the role of the participants in the overall European market. No specific guidelines have thus been issued. Articles in the financial press recently, however, indicated that the merger authorities have resisted mergers in any sector which create a ‘dominant position’, which is usually taken to mean a market share of 40% or above. This was the level which could have been breached by the proposed Swedish SEB/Swedbank merger in 2001. In addition, a supplementary test introduced in late 2003 would forbid mergers which would ‘significantly impede effective competition’ - defined as companies with 20-40% of a market increasing their share without necessarily crossing the 40% mark.

This theme of cross-pillar mergers creating yet more conglomerates which are ‘too big to fail’ was raised as a key concern in our conversations with regulators and the European Commission in Brussels. They estimate that roughly 20% of the European Union’s life insurance premia and one third of its retail deposits are in the hands of such bank/insurance conglomerates, and a new directive on such conglomerates at the end of 2002 has been introduced to improve the regulation process. Thus both in the Brussels Commission and national regulatory agencies like the FSA in the UK, attention is being focused on an integrated analysis and regulatory oversight of large financial conglomerates to identify possible risks that might escape individual bank or insurance regulators.

Only in the UK and Norway have local regulators in Europe taken a proactive public stance against concentration in the financial sector. In the early 1990s, the UK’s Office of Fair Trading effectively prevented Lloyds Bank acquiring Midland Bank by referring the proposed deal to the Monopolies and Mergers Commission, allowing HSBC to acquire Midland instead.

In 2001, the UK Competition Commission vetoed the proposed acquisition by Lloyds TSB of Abbey National, both bank/insurance conglomerates, on the grounds that the resulting market share of Lloyds TSB in the key product of personal current accounts would increase from 22% to 27% and the share of the big four banks would also rise by 5% to 77%. Moreover, it believed that Abbey had potential to develop as a significant competitor in the SME market, where the big four banks already held an 85% market share.

More recently, in 2003, the Norwegian Competition Authority objected to a proposed merger between DnB and Gjensidige NOR, which would give them a dominant share of the country’s banking, life insurance and fund management businesses. However, the Government has subsequently given its approval on condition that the merged group sells some companies and ownership interests.

4.2 Concentration In South Africa

As described in more detail in Appendix B, the high level of concentration in South Africa is a product of the apartheid years when foreign banks disposed of their holdings to local insurers and banks, while several problem banks were bailed out by these same local institutions. Thus we estimate that the four largest conglomerates currently command perhaps 80% of the banking market plus a dominant share of the insurance and mutual fund sectors.

In 2001, a bid by one of the four (Nedcor) for another (Stanbic) prompted a national debate on the concentration issue, culminating in the decision by the Ministry of Finance to deny the application on a variety of grounds:

• an increase in concentration (measured by the Herfindahl index) to the level of 0.21 (or 2100 as measured in the US), ‘which is well outside accepted international thresholds. Levels of concentration have been unacceptably high from an international perspective’. In comparison, we calculate that this index for the six major banks in Canada in 2002 would have been 0.14 (or 1400). In the US, the Justice Department considers a ratio of 0.18 (or 1800) to be ‘highly concentrated’.
• potential damage to an insurance competitor. Liberty, the insurance arm of the target group, might be deprived of a key distribution channel as a result of the take-over of Stanbic and the substitution of Nedcor insurance products.
• job losses from the merger of 10,000-15,000 employees.

We understand that this remains the position of the Ministry and that no other merger among the top four conglomerates will be permitted.

4.3 Concentration In Australia

As described in detail in Appendix B, the process of banking consolidation in Australia was terminated in the early 1990s by a proposed acquisition of a troubled insurer, National Mutual (NatMut), by a major bank (ANZ). The Australian Treasurer declined to approve the merger, establishing at the same time a ‘pillar’ policy similar to that which currently is in place in Canada. He determined that the banking pillar would comprise the existing four banks, with a separate insurance pillar consisting of NatMut and AMP. No mergers would be allowed among these six institutions so as to preserve what was termed at the time ‘the national interest’.

Subsequently the troubled NatMut was acquired by a foreign insurer, AXA. AMP is also currently passing through a difficult period, and one of the four domestic banks, NAB, has expressed interest in acquiring its Australian operation. Although such an acquisition will violate the original pillar policy, market sources indicate that it might be approved.

The Treasurer has been silent to date on its posture in the event of such an application.

4.4 Merging To Create A National Champion

One of the arguments for financial service concentration is to create a ‘national champion’ which can not only ensure a powerful locally-owned competitor but also take advantage of expansion opportunities abroad to compete on an international scale. This argument is particularly attractive in the smaller European markets but has also been made in France, Canada and South Africa.

While some major national financial institutions have successfully expanded abroad, the overall record of international expansion is not impressive. Experience seems to show that leading national institutions frequently underestimate the level of local competition and unique nature of the overseas market when they contemplate their international strategy. In recent years the trend has been the opposite: local financial institutions are buying up across Europe, the US and Asia the beachheads established by foreign peers which once perceived themselves as regional or global competitors.

Interestingly, the merits of such a national champion to be formed from the existing players is being actively discussed in France, which was the home base of one of the earliest proponents of the concept, Crédit Lyonnais (CL). As a large state-owned commercial bank, nationalised in 1982, CL was encouraged by the French government to build up a pan-European network through acquisitions.

With no independent shareholders to act as a restraining influence, the bank embarked on an aggressive acquisition strategy from 1989. It bought a series of relatively small banks across Europe, including Germany, Spain and Italy, at prices that subsequently proved to be far too high. It also opened a significant number of branches in the UK and other countries.

The purchases included BfG Bank in Germany, Chase Banque de Commerce in Belgium, Banco Jover and Banco Commercial Español in Spain and Credito Bergamasco and Banco San Marco in Italy.

The banks acquired by Crédit Lyonnais generally lacked sufficient scale to prosper and many were loss-making. Their cost undermined the bank’s capital position and so, when it was hit by poor performance in its industrial holdings and heavy loan losses at home and abroad, CL was forced into effective bankruptcy.

As a condition for allowing the French government to rescue CL, the European Union insisted that almost all the foreign acquisitions should be sold. These disposals generated further heavy losses, and demonstrated the reckless prices that CL had been prepared to pay. The bank was privatised in 1999 and acquired by Crédit Agricole in early 2003.

Recently, CL has been accused of illegally acquiring the US insurance company Executive Life in 1993. A tentative settlement was reached with the US Justice Department in which CL would pay US$100 million and the French state US$475 million. At the time of writing this report, however, the French Government has declined to accept this settlement.

Other supposed national champions have suffered similar fates. In the insurance world, the leading Australian insurer AMP is currently liquidating its massive overseas empire in the UK and elsewhere, where its losses are threatening the independence of the parent organisation.

Skandia, the leading Swedish insurer and formerly the second largest Swedish company in terms of market capitalisation, also expanded into the US, UK and other markets on the back of a variable annuity strategy which made it a role model for success during the bull market. In its aftermath, however, substantial losses have been incurred, and the entire top management team has been removed, the US operation sold and home country operations severely cut back. The company is also in the headlines currently for corporate governance infractions.

4.5 Summary

As indicated by Figures 6 and 7, Canada has one of the most concentrated financial sectors of any of the major developed nations. In countries like South Africa and Australia with a similar concern for the abuse of economic power, sustaining customer choice and ensuring local control of the financial sector, the regulatory authorities have also taken a firm view against further concentration. In Europe, the UK authorities have already declared further mergers among the top clearing banks as off limits. In addition, we believe based on our recent investigations for this study that additional mergers of major financial conglomerates in countries like Ireland, the Benelux area and the Nordic region would be strongly resisted – if not by the local authorities, then those in Brussels.

Sustaining the current pillar structure in Canada is thus totally consistent with global trends in the regulation of financial institutions.

5.0 Sustaining A Strong, Independent Insurance Sector

Maintenance of a strong and independent insurance pillar has been a basic public policy objective of the Government of Canada. Such a pillar has virtually disappeared in markets like Australia and South Africa for a variety of reasons – the impact of apartheid, management mistakes and more successful strategies of banking competitors. The Australian experience in particular confirms that there is no assurance that a pillar policy can save an insurance sector from self-inflicted mistakes.

It is in the European Union, however, where useful lessons for the future can be drawn when evaluating the liberalisation of cross-pillar relationships.

5.1 Lessons Learned From Europe

We have discussed above the consolidation of the European financial services sector during the 1990s and its impact on concentration and the viability of individual merger strategies. We now examine in more detail what happened on the ground in key markets to individual insurers and their traditional distribution channels – employees, brokers and agents.

In markets like France, Italy and Spain, banks won in a matter of a few years a dominant share of the retail distribution of life products. A combination of good brand reputation, the similarity of life products to standard bank time deposits, and a nation-wide branch network enabled the banks to make deep inroads into the 70-80% of retail life distribution typically held by insurance agents and brokers in the 1980s. Figure 9 tracks the rise in bank distribution since 1995 in three countries – Sweden, Belgium and Italy. In others, such as France and Spain, massive bank share gains were made in earlier years.

Figure 9: Change in Life Insurance Distribution Channels in
In Selected European Countries: 1995 – 2001  

A) 2001 data not available
B) Includes post offices
Source: CEA

In such markets, retail life insurance essentially became known as a ‘bank product’ – simple products which could generally be sold across a bank counter by staff with minimum training or advice from insurance specialists. Such low cost distribution by traditional bank staff can naturally undercut the selling margins inherent in the life economic model, and at least in the early ‘bancassurance’ years, banks not only won market share but also garnered sizeable distribution fees by sheltering under the traditional life margins.

In practice, the impact in these markets on the viability of the traditional insurers and distribution channels was transforming. In Spain, where 18% of the insurance sector was already owned by banks, the banks chose to manufacture the life product internally and to sell the traditional insurer. Thus Banco Bilbao Vizcaya (BBV) divested itself of two insurers, Aurora Polar and Plus Ultra, to foreign insurers interested in penetrating the Spanish market, AXA and Norwich Union, while SCH similarly sold its stake in Vitalicio to Generali. There remains in Spain only one major independent insurer, the demutualised Mapfre, which uses a savings bank network as a distribution channel.

In the Netherlands, within two years of the dismantling of a two pillar policy at the end of the 1980s, all of the leading banks had become part of a financial conglomerate and sold only their own-brand life products. The insurers ING and Fortis had bought banks, while Rabobank had bought an insurer, and ABN Amro had set up its own internal manufacturing capability. Of the major insurers, only Aegon was left without a banking partner.

In France, the largest insurer, UAP, was unable to implement a banking alliance with BNP, suffered major asset losses, and eventually was acquired by AXA. Other insurers were sold to foreign insurers like Generali. Today the dominant life companies are the bank Credit Agricole and government-owned insurer CNP, both of which benefit from exclusive distribution through related bank channels. One of the consequences was a decimation of the agency channel in France; the Financial Integration study cited above notes that the number of French agents fell by 50% during the 1990s.

Finally, as noted above in section 3.0, the change in distribution strategy of banks in the UK and Germany during the 1990s also generated a number of sales of insurance affiliates, usually to other insurers. In the UK most insurance companies have abandoned their own directly-employed salesforces, and it is estimated that for every nineteen salesmen in 1990 there is now only one left.

Recent research by Datamonitor shows that banks already account for more than half of life insurance distribution in three of the largest five European countries – see Figure 10. Moreover, it predicted significant gains in banks’ share of distribution in all five markets between 2002 and 2007. At the same time insurance companies’ own distribution channels – telephone and internet, salesforces and other company employees – are consistently losing market share.

Figure 10: Projected life insurance distribution in major European countries, % by channel – 2002 and 2007E

	Banks & bancassurers		Tied & multi-tied agents		Independent advisers & brokers		Insurers’ own staff & salesforces
	2002	2007E	2002	2007E	2002	2007E	2002	2007E
Spain	77	81	–	–	20	18	3	1
France	61	67	8	5	9	9	22	19
Italy	56	66	34	30	1	1	9	3
Germany	19	24	51	54	21	16	9	6
UK	18	28	17	13	56	52	9	7
Source: Datamonitor

This trend potentially threatens the future of the independent insurance sector in a number of ways. Most notably, banks will choose the insurance supplier which offers the best rates of commission, undermining domestic insurance companies and the independent distribution channels they have built up.

In Spain, for example, several of Europe’s largest insurers have exclusive distribution agreements with local banks. Germany’s Allianz uses Banco Popular, the third largest commercial bank, and has cemented the arrangement with a 9.9% shareholding. The UK’s Aviva has deals with five regional savings banks, giving it access to 9 million customers through about 3,500 branches. The largest saving bank, La Caixa, has a successful joint venture with Fortis, covering both life and non-life policies.

As banks capture greater share, often coupled with exclusive distribution deals, they are in a position to dictate financial terms and take the lion’s share of the profits. This will weaken the insurance companies and could undermine their ability to develop new products.

Domination by banks may also affect the type of products that are developed, concentrating attention on simple products that can be sold easily to banking customers. This could hurt customers and restrict their choice of products. Relatively sophisticated products may become more expensive and more difficult to obtain.

In sum, liberalising the ownership and sale of insurance products in Europe has been a major unsettling influence on ownership structures as well as distribution channels. One can thus argue that the result is not consistent with the Canadian policy objective of a strong, independent and locally-owned insurance sector. If Canada were to introduce the same liberalisation of insurance ownership and distribution has been seen elsewhere, it could be expected to follow the pattern already experienced across Europe. Banks would become important distributors of life (and to a lesser extent non-life) insurance.

Banks would tend to gravitate to a small number of insurance companies that would become specialists serving their needs. These companies might become overly dependent on banks for their future prosperity and at considerable risk when contracts were due for renewal or if bank ownership should change. Their products would increasingly be developed with banks in mind, possibly ignoring the needs of the wider market. Other insurance companies, in turn, might be weakened by the loss of business to those that specialised in serving banks.

Some – perhaps all – big banks might aspire to take full control of the insurance products they distribute through their branches. They could achieve this either by acquiring an insurance company or developing their own in-house operation. In either case, they would benefit from success in cross-selling through bank branches that is a feature of most markets outside Canada. This success would inevitably weaken the remaining independent insurance companies, undermining their scope to innovate and compete, and thus reducing customer choice and service.

---

Appendix A Bibliography

Amel, Dean et al: Consolidation and Efficiency in the Financial Sector: A Review of the International Evidence (2002)

Beitel, Patrick et al: Explaining the M & A Success in European Bank Mergers and Acquisitions (2003)

Bo, Jacob et al (McKinsey Quarterly): Assessing Insurance Deals (2003)

Comite Europeen des Assurances: European Insurance in Figures, (2003)

Davis, Steven et al (Nomura International): South African Banks: Strategy Games (1999)

Davis, Steven I (presentation to EBR Forum): Operational Synergies in Bank/insurance mergers: Does cross selling work (2002)

Davis, Steven I et al (Salomon Brothers): European Insurance Distribution: a Battle for the Customer – more losers than winners (1992)

Davis, Steven I et al (Salomon Brothers): Dutch Insurance Distribution – a case study at the cutting edge of bancassurance (1993)

Davis, Steven I.: A Report Card for Bancassurance in Europe – and lessons for the future (2002)

De Nicolo and Kwast (IMF Working Paper): Systemic Risk and Financial Consolidation (2002)

Department of Finance, Canada: Response of the Government to Commons and Senate Committee Reports (2003)

Dombey, Daniel (Financial Times): Doughty pioneer of a cross selling future (2001)

Dorval, Bernard: Development of Bancasssurance in Canada (2002)

Economist: Why banking and insurance work well together in the south – (2003)

European Banking Report: European Banking and Insurance Monthly Briefing (2003)

Finaccord: Partnership opportunities with European banks, mutual funds, life insurance and retirement savings (2003)

Fox-Pitt Kelton: The Meaning of Life- a bank investor’s view of life insurance (2002)

Fox-Pitt Kelton: The European savings market (2002)

Freedman, Charles et al: The Financial Services Sector: An Update on Recent Developments (2002)

Gentay, Nadege and Molyneux, Philip: Bancassurance (1998)

Group of Ten: Consolidation in the Financial Sector (2001)

Imeson, Michael: Bancassurance in Europe (article in the Banker) (2002)

Insurance Information Institute (US) : Financial Services and Insurance: the Long March (2001)

Life Insurance International (Lafferty Publications): Different countries, different models (2003)

Life Insurance International (Lafferty Publications): Pursuing the Bancassurance route (2002)

Norwegian Competition Authority: Notice of possible merger intervention into DnB/Gjensidige merger (2003)

Office of the Superintendent of Financial Institutions (Canada):Report to the Ministry of Finance, Proposed banking mergers, (1998)

Pons, Jean-Francois: Competition in the Financial Services Sector in Europe Today (2002)

Skipper, Harold D, Jr: Financial Services Integration World-wide: Promises and Pitfalls (1999)

Task Force on the Future of the Canadian Financial Services Sector (1998).

Thomson. Maria et al: UK Lessons for bancassurance (2002)

Totaro, Leonardo (presentation to EBR Forum): Threats and opportunities in European Bancassurance (2002)

Van den Berghe et al – Convergence in the Financial Services Sector (1999) - report commissioned by the OECD

Appendix B

Summary History Of Evolution Of Bank/Insurance Mergers And The Regulatory Posture In Key Markets

1) South Africa

The unique structure of South Africa’s financial services business was shaped in the 1970s-80s by the apartheid era. The major foreign banks sold their local units for political reasons. Thus Barclays was converted to the current First National Bank and Standard Chartered’s local office became the present Stanbic, while Nedcor, which encountered problems, ended up in the hands of the insurer Old Mutual.

As a result, two of the four leading banks (Nedcor and ABSA) are effectively controlled by insurers (Old Mutual and Sanlam). Stanbic actually acquired the holding of the Liberty insurance group in itself, thus reversing the roles. The mining group Anglo American effectively divested its banking holdings by merging First National into First Rand. The result is an extremely concentrated financial sector with four banks now commanding an estimated 80% of total banking assets.

The two dominant insurers – Old Mutual and Sanlam – then demutualized in the late 1990s. A considerable debate took place over whether their banking interests should be spun off, but the decision in each case was to retain the holding but not to integrate the banking and insurance businesses. Thus ABSA owns the country’s largest insurance brokerage sales force and is not an exclusive provider of its parent’s products. While there is no prohibition of insurance products being sold through a bank branch network, such cross-selling is not a high strategic priority for most of these conglomerates.

No significant change took place in this structure (except Nedcor buying the fifth largest bank) until 2001 when Nedcor made a bid for Stanbic. The Ministry of Finance concurred with the recommendations of the Register of Banks and Competition Commission and turned down the merger. The grounds for this decision were the high level of existing banking concentration, job losses at the target bank, and possible damage to the viability of Stanbic’s insurance subsidiary Liberty which might have been replaced as supplier to Stanbic’s banking customers.

Our conversations with bankers in South Africa confirm that the effective block on mergers among the leading banks is still in place.

2) European Union

A significant reshaping of the European financial services sector began in the late 1980s – early 1990s driven by the prospect of a single currency and financial market, the perceived attractiveness of bank/insurance mergers, and the decision to permit European financial institutions to offer a full range of financial services. A landmark event was the Second Banking Directive which permitted banks and other financial institutions to offer any financial service. Particularly in the smaller markets, banks and insurers turned to each other to create financial conglomerates in lieu of mergers with their peers.

Thus the typical retail financial institution in Europe offers insurance as well as banking and fund management products.

Regulators in the national markets in the 1990s generally took a liberal posture towards of in-market mergers in their financial sector. Many took the view that such mergers should be regarded in the context of the single financial market which was being created, rather than the national market involved.

In 2001, however, two events occurred. First, in the UK, which has a pro-active Competition Commission, a proposed merger between Lloyds TSB, one of the top five banks, and Abbey National, a smaller institution, was vetoed by the authorities because it would have increased Lloyds TSB’s share of the key personal current account market from 22%% to 27%. Since then, the Norwegian Competition Authority in 2003 has acted to block a merger between two bank/insurance groups, DnB and Gjensidige NOR, which would create a single conglomerate with perhaps 50% of the financial sector. It is expected at time of writing this report that the merger will proceed subject to agreed divestitures.

The second event was the decision in 2001 of the European Commission in Brussels to offer substantive objections to the proposed merger in Sweden between two of the four dominant financial conglomerates, Swedbank and SEB. Rather than negotiate a compromise with Brussels, the two banks decided to abort the merger, although it had already been approved by the Swedish authorities. Similar objections were made to the acquisition of Dresdner Bank in Germany by the insurer Allianz, but changes in the transaction satisfied the Brussels authorities and the transaction proceeded.

3) Australia

The report of the Campbell Committee in the early 1980s was the first major review of the Australian finance sector since before the Second World War, and its scope set the pattern of Australian economic development for the next two decades.

Almost everyone agreed with the Government’s decision to implement the across-the-board recommendations to open up the Australian economy to international forces; floating the currency, freeing up financial flows, removing banking regulation and allowing foreign competition in the banking sector.

The first response in the banking sector was consolidation. The National Bank of Australasia merged with the Commercial Banking Company of Sydney to form the National Australia Bank while the Bank of New South Wales acquired the Commercial Bank of Australia to form Westpac Banking Corporation. The six ‘major’ banks were reduced to the four major banks now in existence. Each acquisition needed the approval of the Treasurer, but as the consolidation was seen as a natural response to the opening up of the market there was no difficulty achieving the approval.

Even then the prevailing attitude among the banks was one of fear of being unable to compete with larger international banks, which were presumed to have greater experience as well as larger scale. Australia, it was felt, needed banks of greater size, which was taken to also mean international status.

Westpac led the charge overseas in the 1980s expanding into the US, UK and Asia. Almost all these initiatives turned out to be value diluting for shareholders. When Westpac faced large write offs from domestic property loans early in the 1990s, these expansions were almost all sold off. ANZ similarly expanded into Asia with costly write-offs in India through regulatory problems with its Grindlay’s exposure. NAB made substantial acquisitions in US and UK retail banking. These proved to be very successful, but the subsequent purchase of HomeSide, a US mortgage specialist, was a disaster. The Commonwealth Bank, previously government owned, has remained more modest, limiting its overseas expansion to a successful acquisition in the New Zealand market. The recent sale by Lloyds TSB of its New Zealand operation to ANZ now means the New Zealand banking sector is entirely owned by the four major Australian banks.

During the 1990s, the two major life companies, AMP and National Mutual (NatMut), engaged in a head-on fight to acquire each other’s agency teams by offering generous loans and payments to change sides. It turned out that NatMut did not have the deep pockets many expected, and late in the 1990s it turned to one of the Melbourne based banks, ANZ, for a (rescue) merger. The merger was opposed by the other banks, Westpac in particular, on the grounds of incompatible taxation regimes, which would have given the ANZ an unfair advantage over the other banks. The Treasurer agreed and declined to approve the merger, enshrining the need for the four major banks and the two major life companies to remain separate from one another in the ‘six pillars’ policy. The basic test for approval by the Treasurer was, and still is, a very vague assessment of the ‘national interest’. The policy did not preclude acquisitions by overseas companies, and the now weakened NatMut was acquired by AXA in 1996. At this point, the ‘six’ pillars policy was reduced to the current ‘four’ pillars policy and applied only to the four major banks.

For a period in the early 1990s the AMP reigned supreme as the dominant life company and one of Australia’s oldest companies. In the early 1990s it attempted an alliance with Westpac but this failed. In the mid 1990s, AMP also chose demutualization and with cash in hand, pursued its well-established acquisitions policy in the UK. Along with a disastrous acquisition in its domestic market, AMP’s UK business has been badly hurt by recent stock market falls. AMP has recently announced a complex demerger process to float off its UK entities and return to a domestic, and much smaller institution. Public speculation is that AMP, once it has stabilised, will be acquired by either NAB or by Westpac. There is no suggestion at this stage of the Treasurer not approving such an acquisition.

At the turn of the millennium the four major banks now dominate the finance sector. Over the previous 20 years they have moved in to every major area of the finance sector, particularly the newly emerging superannuation and funds management sector spurred on by the government’s compulsory contribution to one’s own retirement.

Twenty years ago, at the dawn of deregulation, banks were regulated to their own banking sector. Despite the billions of dollars of shareholders funds lost in the forays overseas, following deregulation they have consolidated to the position where, despite not being able to merge or acquire one another, they collectively control 80-85% of the entire sector. The previously significant two large life companies have largely, by their own hand, domestically and overseas, reduced their presence to such an extent they no longer receive policy attention at this level.

4) USA

The history of bancassurance in the US is one of progressive deregulation, which has its origin in the passage of the Glass Steagall Act of 1933, to separate other industries, especially securities, from commercial banking. The failure of nearly 10,000 banks following the stock market crash in 1929 was widely attributed to bank exposure to the securities industry.

The second major regulatory milestone was the Bank Holding Company ("BHC") Act of 1956, enacted to limit the expansion of banking institutions into nonbanking activities. It prohibited banks or any company that owns or controls two or more banks from conducting any other business, except businesses that the Federal Reserve Board considered as "closely related to banking". An amendment in 1970 extended this prohibition to one-bank holding companies.

Throughout the 1980s and 1990s, bank distribution of insurance products grew by specific-case court and regulatory decisions, some of them incorporated into a few pieces of legislation. In 1982, bank holding companies were explicitly barred from distribution or underwriting insurance, with a few relatively small exceptions (e.g., credit insurance). In 1986, the Office of the Controller of the Currency (OCC) ruled that national banks could sell insurance from offices located in towns with populations of 5,000 or less, a loophole traceable to an obscure 1916 amendment to the National Banking Act. In 1995, the US Supreme Court ruled that banks could sell annuities, which were previously considered insurance and so not eligible to be sold by banks.

From the resulting patchwork of permitted activities, banks fashioned a huge expansion in sales of annuities, but they sold relatively little other life or property-casualty insurance. In 2000, banks sold 29% of all fixed annuities in the US (up from 21% in 1995), and 10% of all variable annuities (from 7% in 1995). But bank sales of other lines were relatively small: in 2001, bank market shares were estimated at 2.4% of individual life and accident & health insurance and 4% of non-life. The latter market shares were boosted by Wells Fargo's 2001 acquisition of Acordia, a network of insurance agencies formed in 1989. In aggregate, Acordia agencies placed over US$6 billion insurance premiums in 2002 – making them fifth among US insurance brokerage/agency organizations. The combination of Acordia and Wells' other insurance distribution operation contributed about 4% of Wells' total revenue of US$20.4 billion in the first 9 months of 2003.

The major event that precipitated the most significant legislation affecting bancassurance was Citicorp's announcement in April 1998 of its intent to merge with Travelers Group, the insurance and investment banking firm, to form Citigroup. To permit completion of the Citicorp/Travelers merger (in October 1998), the Federal Reserve Bank granted a two-year trial period, pending new legislation. Citigroup did not have to wait long for that legislation.

The Financial Services Modernization Act of 1999, also known by the names of its principal sponsors in Congress, as the Gramm Leach Blily Act ("GLBA"), repealed certain sections of Glass Steagall and the BHC Act of 1956, thereby permitting banks to enter new financial services though newly-designated financial holding companies (FHCs) or – for national banks – financial subsidiaries. Permitted activities for both kinds of entities include acting as (or owning) an insurance company and managing insurance company investments.

Since President Clinton signed the GLBA in November 1999, many insurers have formed their own banks, but very few major banks have acquired insurance underwriters, and Citigroup even spun off the non-life side of Travelers in 2002. Banks have overwhelmingly favored the approach of buying insurance agencies, which continue to sell insurance from their own offices, supplemented by joint marketing programs with the parent bank.

Two exceptions to this strategy, however, stand out:

As part of a series of acquisitions in the US, Royal Bank of Canada acquired Liberty Life and its associated services company in 2000 and Business Men's Insurance Company in 2003, spending an aggregate of US$800 million to acquire the two organizations. RBC has said that it plans to sell the products of these companies through RBC Centura, a southeast regional commercial bank that it acquired in 2001.

In 2003, Bank One acquired certain life and annuity components of Zurich Life from Zurich Financial Services, for $500 million. In the Bank One announcement, Jamie Dimon, Bank One Chairman and CEO, gave the strategic rationale: to enhance Bank One's life insurance and annuity products, distribution capabilities, and systems.

Appendix C –About DIBC

Davis International Banking Consultants (DIBC) is a bank strategy consulting firm based in London, England. Founded in 1980 by Steven I. Davis, a former career banker, DIBC has carried out several hundred consulting assignments for financial institutions, government agencies, suppliers to the financial services sector and others. A particular specialty is best practice analysis on issues across the financial sector. In 1993, DIBC was engaged by the Norwegian Ministry of Finance to make a series of recommendations on the restructuring of Norwegian banking during the Scandinavian banking crisis. In the bancassurance sector, the company has prepared a number of studies for individual clients as well as for distribution to the clients of investment banks.

In addition, Mr. Davis has written eight books on the management of financial institutions, the latest of which is Excellence in Banking Revisited. DIBC's website (www.dibc.co.uk) contains further details on the company and its clients.