From Subsidiary to Branch Organization of International Banks; New Challenges and Opportunities for Regulators

Transcription

1 From Subsidiary to Branch Organization of International Banks; New Challenges and Opportunities for Regulators Lawrence G. Goldberg University of Miami Richard J. Sweeney Georgetown University Clas Wihlborg Copenhagen Business School Preliminary version, Febr. 20, 2005 A preliminary version of this paper was presented at conference on Integrated Cross-Border Banking: Issues in Implementation, Regulation and Supervision organized by the Center for Law, Economics and Financial Institutions at CBS (LEFIC) on Oct. 29, 2004.

2 I Introduction Bank mergers across borders have accelerated in recent years. Despite the relaxation of restrictions on mergers in Europe with the passage of the European Union Banking Directives, the number of cross-border mergers has not approached the magnitude expected by most observers. Nevertheless, important regulatory issues arise from these cross-border mergers for the bank regulatory authorities in the countries of the acquiring bank and the acquired bank. In addition, important operational questions arise for management of the combined companies. Operating across country lines leads to questions and concerns that do not appear when the bank is located only in one country and must follow the legal rules of that country only. As will be discussed below, numerous banking problems have arisen for banks with offices outside their home country. In several of these cases, the regulators had a very difficult time resolving these cases satisfactorily. 1 This paper takes as its starting point a recent cross-border exercise that has confounded regulators and managers, the case of Nordea. Unlike most cross-border consolidations that involve a dominant bank assuming control over a smaller bank in another country, the case of Nordea involves joint ownership from four closely linked and similar sized countries (Denmark, Finland, Norway, and Sweden). The original banks that merged into Nordea were all important in their countries with market shares ranging from 15 percent in Norway to 40 perceent in Finland. In addition, non-banking activities 1 See Herring (2004) for a discussion of crisis resolution for large cross-border banks. 1

3 such as insurance and portfolio management have played important roles in the original banks. An important aspect of Nordea s strategy is that it intends to serve all countries involved in an integrated organization within each product area with the purpose to achieve economies of scale within functional areas. This strategy is harder to implement in retail banking than in, for example, asset management. Nordea has changed its organizational charts many times indicating that management is uncertain about the appropriate organizational structure. The legal structure in the four countries has not and does not coincide with the internal organization across countries and functions. The current and past organizational structures have forced supervisors in the four countries to develop new approaches as well in order to adjust the supervision to the business organization rather than the legal organization of the bank. The four supervisory authorities (FSAs) have formed a joint Nordea committee with a number of subcommittees with participation of representatives from the four FSAs. There seems to be agreement between Nordea and the supervisors that the integrated bank spends more time with supervisors and tax authorities now than the four banks did in total before the mergers. A major reorganization is being planned for Nordea wants to re-organize so that the bank will be headquartered in Sweden as a European Company with branches in the other three countries. Sweden is the largest of the four countries, but it does not dominate the others within Nordea. All countries play important roles in the bank and the bank plays important roles in all the four countries. The situation differs greatly from the recent expansion by Western European banks into Eastern Europe. Western banks control a large part of the major Eastern European banks, which have been organized as 2

4 subsidiaries. The Eastern European operations do not represent large parts of the banks, however. The Nordea model might well become the model of the future where banks in smaller countries band together in order to achieve sufficient size to compete with larger banks from larger countries. On the other hand, there is no clear evidence that economies of scale and scope will successfully drive banks in this direction. The main benefit to Nordea of its proposed reorganization seems to be reduced Value-Added Tax Payments in cross-border transacations. To outside observers, the four Nordic countries look as similar to each other as almost any other combination of countries in the world. However, there is much anecdotal evidence that the differences are substantial in terms of business culture and regulatory approaches. 2 Nevertheless, if the Nordea model for cross-border integration of similar size banks in similar size countries cannot work in the case of Nordea, then perhaps it cannot work anywhere. The regulatory responses as well as the performance of Nordea are likely to be closely watched by financial institutions in Europe and elsewhere. If regulatory issues can be resolved and the bank performs well, Nordea s cross-country strategy may be imitated and start a wave of cross-country mergers. The main goal of this paper is to evaluate the regulatory and supervisory issues arising from cross-border operations, with an emphasis on the Nordea case. The next section looks at cross-border activities and reviews relevant academic literature. The third section examines regulatory problems that can arise from this activity, and the EU s approach to regulation, supervision and deposit insurance for cross border banking in 2 See Søderberg and Vaara (2003) for a description of the perceived differences among work-cultures of the four nationalities working together in Nordea. Nordea disputes many of these findings. 3

5 comparison with the approach taken by regulators and supervisors in the USA. The fourth section deals with the special problems arising in the Nordea case, especially the conversion from a subsidiary arrangement to a branching arrangement. In Section V alternative approaches to supervision of international branch banking are discussed. We argue that the Nordea case offers an opportunity to develop a supervisory structure that enhances market discipline for large international banks. II Cross-Border Activity Enthusiasm for cross-border activities by banks has varied greatly over the years. Banks see profit opportunities in expanding across borders but also see difficulties in coordinating operations across country lines. There have been a significant number of instances where lack of control of foreign operations has caused financial institutions serious problems and some of these cases are discussed briefly below. The European Banking Directives have formally made it much easier to expand across European borders, but still we do not see as many cross-border combinations as we might suspect. In recent years with the launch of the Euro and increased market integration in the European Union, there has been an increase in cross-border activity in Europe. Whereas domestic mergers and acquisitions have traditionally been friendly, cross-border acquisitions give rise to nationalistic feelings and regulatory resistance. In fact national regulators and governments in the EU have stopped cross-border mergers on several occasions. One example is the case in Portugal several years ago where a Spanish bank (Banco Santander) was not allowed to acquire a Portuguese bank. Banco Santander was ultimately allowed to buy a smaller part of the Portuguese bank. On Febr , the EU Commissioner for the internal market wrote a letter to the Italian Government asking it to 4

6 correct the impression that it is trying to stop foreign acquisitions of Italian banks. 3 We have seen European banks acquiring US financial institutions, and US banks expanding internationally. In the first case several European banks have acquired capital market capability in New York as well as in London. In the second case American banks have been able to build up investment banking in Europe during a time when the Glass- Steagall act prohibited them from integrating commercial and investment banking in the USA. American commercial banking activity abroad has generally been marginal in the host countries and oriented towards a limited clientele. Only as a result of market reforms in Eastern and Central Europe have Western banks acquired dominant positions in foreign countries. In most of these cases, the banks have acquired a local bank with the objective of running it as a local bank while transferring operational expertise to the subsidiary. There exists a large literature examining the factors affecting the growth of banks across country lines. Conventional wisdom says that banks initially expand abroad in order to follow their home country customers when they become multinational firms. Even if this explanation may be valid for the very first step outside the home country, it is clear that there is substantial host-country oriented financial activity by foreign banks in many countries. After learning about a market by means of representative offices, many banks draw on their expertise to expand into new fields. International expansion has also been induced by regulation. The Eurodollar markets developed in the 60s partly as a result of American interest rate ceilings (Reg. Q), which could be circumvented by American banks placing dollar deposits in branches in London. US commercial banks have also been able to run investment banking 3 The Economist, Febr. 19,

7 operations in Europe when the Glass-Steagall Act prevented them from doing this in the USA. Although foreign banks have expanded their services in host countries after the initial establishment, they have rarely become full-fledged local financial institutions as noted above. Even in some developing countries, where foreign banks long have had retail operations and a substantial presence, they have focused on the large corporations with international activities. The factors hindering international expansion across the range of banking services are a combination of discriminatory regulation and practices in host countries, and costs of acquiring necessary local knowledge in retail banking in particular. The expansion of Western banks into Central and Eastern Europe is the response to liberal policies with respect to FDI in banking. These policies have enabled Western banks to obtain the necessary local expertise by acquiring whole financial institutions that often remain under local management. This type of expansion is motivated by benefits from knowledge transfer in credit risk management, information systems, and capital markets 4. The Nordea expansion can also by EU standards be viewed as a response to relatively non-discriminatory policies in the Nordic countries in recent years with respect to foreign banking acquisitions. There is controversy, however, about the appropriate approach to cross-border expansion in these countries. While Nordea s approach aims at integrating functions across the countries, the approach taken by other banks has been to acquire a foreign bank and to let it run as a separate business unit in a subsidiary. 4 See, for example, Kowalski et al (2003) 6

8 Early empirical papers dealt with U. S. banks expanding abroad or foreign banks expanding into the United States. For example, Goldberg and Saunders (1980) examined U. S. bank growth in Great Britain. Goldberg and Saunders (1981a) looked at the determinants of foreign banking activity in the U. S. Goldberg and Saunders (1981b) analyzed the growth of foreign banks in the U. S. by organizational form. Grosse and Goldberg (1991) examined the growth of foreign banks in the U. S. by country of origin. Many other studies followed these and frequently used different country combinations, but many still involved the U. S. The types of factors that were related to foreign bank growth included GNP, trade between the two countries, changes in regulations, foreign direct investment, and differences in country risk. The motivations behind the recent cross-border mergers might well also be similar to the motivations that were found in the previous empirical literature. III Regulatory Issues and Approaches to Regulation and Supervision of International Banks When banks operate across country lines regulatory questions naturally arise. Who is responsible for regulating different aspects of the bank s operation, the home country regulator or the host country regulator? Who is responsible for supervision? Are there procedures for crisis management? A related issue is the assignment of responsibility for deposit insurance schemes. Berger, et al. (2000) reviews the problems that can arise from cross-border mergers, which have been discussed in the literature. Cross-border mergers can expand the coverage of the safety net and deposit insurance. More protection to larger institutions may lead to too big to fail. This, of course, increases moral hazard and 7

9 gives banks an incentive to undertake riskier activities. The expansion of the safety net also may increase government protection for non-bank financial activities. Coverage may in fact be extended to non-bank financial institutions as well as the parts of banks engaging in these activities. Consolidation may expose the government safety net to losses from operations in other countries. Depending on the relative size of the home and foreign operation, this exposure may not be acceptable. Finally, consolidation may increase the cost of coordinating regulatory response across countries if a large organization fails. In the EU the national central banks are the lenders of last resort and different countries may have different incentives to bail out different institutions. 5 Goodhart (2004) and Krimmingen (2004) emphasize the problems that arise when an international bank faces distress. Approaches to resolution may differ in law and practice, and conflicts of interest may arise. We return to these issues. Herring (2004) has documented the types of problems that can arise in the case of cross-border financial institutions. He argues that large financial conglomerates appear too complex to fail. Management has difficulty in coordination of activity across country lines and supervisors have difficulty supervising these cross-border institutions. When a problem arises, supervisors have had significant problems in resolving the situations. He analyzes several cases where financial conglomerates have combined banking with another type of financial activity. The cases involve Bankhaus Herstatt, Drexel Burnham Lambert, the Bank for Credit and Commerce International (BCCI), Barings PLC, and Long Term Capital Management. Even cross-border mergers that only involve banking can cause management and regulatory problems. The EU. 5 See Dermine (1999) for a discussion of this issue. 8

10 The EU has formally dealt with the division of labor among supervisors and regulators of cross-border banking in the EU. The principle of home country control and mutual recognition has been established, implying that a bank can work across the EU on a single banking license under home country supervision. Working under a single banking license requires that host country activities are conducted in branches. Under the single banking license the deposit insurance coverage is obtained in the home country. Thereby, it is possible to have competing banks in one country with different deposit insurance coverage and subject to different rules by home country law and regulation. Such a situation is of course perfectly consistent with the principle of mutual recognition, and it is expected to lead to competition between different institutional structures. These principles do not seem to have great support among EU governments in the area of banking judging from the active discouragement of cross-border acquisitions. As noted we have not seen important cross-border activities organized in branches. When host country activities are performed within subsidiaries as is common, supervisory responsibility is shared between the home and the host country. Deposit insurance coverage becomes a host country issue. The home country is responsible for supervision of the consolidated entity, while the host country supervises the local bank. The supervisors must agree on information sharing in a Memorandum of Understanding. The principle is that the host country supervisor must keep the home country informed about risk-taking in the subsidiary but the home country must only reveal relevant information about the whole entity for the host country. Thus if there are problems in the bank outside the host country, its supervisor need not be informed. This situation has led to fears in Eastern Europe that a home country supervisor may collude with a problem 9

11 bank to shift losses to the host country s subsidiary before failure occurs. Thereby bailout-or resolution costs would be shifted to the host country. Facing this risk, supervisors in Eastern and Central Europe often require subsidiaries to hold more capital than what the Basel accord would specify. 6 The risk of collusion between the home country supervisor and the parent bank at the possible expense of subsidiaries host countries is not limited to Eastern Europe although it is particularly acute there since such a large share of the banking sector is foreign controlled. A solution to the problem of Eastern and Central Europe would seem to be to create a branch structure, since supervisory responsibility then would rest entirely with the home country. Furthermore, deposit insurance for all depositors in all branches would be the responsibility of the home country. From a bank s point of view the branch solution may be inconsistent with the bank s strategy relative to the host country, however. If the strategy is to operate the host country bank as a truly local bank under a local name with local management and internal culture, then a subsidiary organization is more appropriate. In the case when the bank operates in host countries in branches, a different type of supervisory conflict may arise. From the host country s point of view it is not certain that the home supervisor will consider the interests of the host country when the bank approaches distress. For example, to cut losses the home supervisor may force the bank to shut down some operations quickly. If these operations are abroad and sizeable the host country may be subject to financial disruption beyond its control. Essentially, the host country loses control over the bail-out policy for the banks operating within its jurisdiction. We return to this issue below. 6 See European Shadow Financial Regulatory Committee (2004a) 10

12 The most important factor behind the lack of important cross-border branch networks is probably the lack of acceptance of banks operating under different rules than domestic banks within the same jurisdiction. In other words, European governments are not willing to accept the principle of mutual recognition fully. Furthermore, the home country governments and supervisors may be unwilling to take responsibility for deposit insurance of both domestic and foreign bank operations. A possible consequence of this is that the home country supervisor s and government s incentive to bail out the bank when a large part of the operations are abroad is not as strong as when the bank is entirely domestic. Thus, the fear of the host country supervisor that local branches will be shut down quickly when problems arise in the bank may be well founded. The other side of the coin is that there are incentives for home country authorities to charge deposit insurance premiums of sufficient height for tax-payers not to be affected even in crisis. Thus the subsidy component in the banking system could decline. The incentives to bail out an international bank with foreign branches are also weaker than the incentives to bail out a domestic bank. If the foreign branch operations are sufficiently large the bank may become TOO BIG TO SAVE, while a large domestic bank tends to be TOO BIG TO FAIL. We return to this issue below. In June 2004 the EU Commission presented a proposal for a new Capital Requirement Directive (CRD 3) based on Basel II. The new proposal includes a shift in the balance of responsibilities of home and host country supervisors of banks with foreign subsidiaries. Specifically, when it comes to validation of risk evaluation models to be used to determine capital requirements under the Internal Ratings based approach, the home country supervisor is given the final say if the two supervisors cannot agree 11

13 within six months. At the same time the deposit insurance responsibility for the subsidiary remains with the host country supervisor. If this proposal is implemented, the reasons for the host country supervisor to distrust the home supervisor may be strengthened further. By the same token the impetus from the host country supervisor to favor branches over subsidiaries may be strengthened. 7 The USA. It is useful to examine the approaches to foreign bank regulation taken in the US relative to the EU. The fractured nature of the American regulatory structure for financial services is well known. The partly overlapping responsibilities of national and state authorities, and of the Fed, the FDIC and the SEC are well known. The FDIC has emerged as the primary regulator, however, in particular for banks operating across state borders. An important difference relative to the EU is that a strong interstate (national) regulator and supervisor exist in the USA. Another difference is that the FDIC has established rules Prompt corrective action for resolving a banking crisis. The American approach to foreign banks operating in the USA can be characterized as national treatment as opposed to mutual recognition. Subsidiaries of foreign banks operate under US rules and they participate in the US deposit insurance system. Branches of foreign banks exist but they are discouraged. They are subject to US rules for deposit insurance. Their assets are fenced in to cover claims in the US. Thus, the US does not recognize the home country s responsibility for a bank. An important aspect of the American regulatory structure is structured early intervention or prompt corrective action. These procedures refer to the management of banks in or approaching distress. The FDIC requires actions by a bank losing capital and 7 See European Shadow Financial Regulatory Committee (2004b). 12

14 these actions become more severe as the bank s capital shrinks. The system is designed to protect the deposit insurance fund and tax-payers, and to provide incentives for banks management to avoid distress situations. The deposit insurance has relatively high coverage and it guarantees rapid payments to depositors in case a bank fails. Taken together the regulatory structure should offer credible guarantees of insured deposits as well as relatively high credibility that non-insured creditors will not be bailed out. Few countries have pre-determined rules for dealing with banks in distress. The common ad hoc approach implies that there is a strong likelihood of bail-outs when problems must be resolved quickly. Furthermore, there is scope for discrimination between domestic and foreign creditors 8. IV Nordea and issues of supervision Nordea presents a special problem for management and for regulators. Nordea has grown from the combination of four banks in each of four countries. None of the banks is much larger than the others and none of the countries is so large that it will dominate the other countries. Currently the organization is built around bank subsidiaries in each country. This gives primary regulatory responsibility to each of the four home countries for the bank subsidiary located in that country. However, there can be carryover effects from problems with the bank in a country on the other subsidiaries. Furthermore, there is considerable non-banking activity done in most of the countries and this can raise regulatory and management issues as discussed above. The strategy of Nordea is to serve all four countries in various product areas within a functional organization. This strategy has not yet been fully implemented in 8 Lastra (2005) discusses laws in various countries for dealing with cross-border bank insolvencies. 13

15 retail banking in particular, while asset management and corporate banking services are highly integrated. The legal organization differs from the functional organization, because taxation systems and regulations are separate and different in the four countries. These differences have created serious challenges for both Nordea and the supervisors. Nordea must adjust its organization and accounting systems relative to tax systems in particular, while the supervisors must adjust their activities relative to Nordea s functional organization. The value-added tax systems in particular had created a tax burden when services are offered cross-border. Nordea has apparently found it difficult to achieve a legal structure that is satisfactory both from the point of view of achieving strategic objectives and from the point of view of national tax and regulatory authorities. The bank has changed its organizational charts frequently, and sometimes more than once in a year. The latest proposal is to convert the organization to one where the bank is headquartered in Sweden, the largest of the countries, and the other countries would have branch offices. This raises new regulatory issues. What if branches in one country have difficulties? How are they bailed out if at all? What if the bank as a whole faces distress or insolvency? What role is played by different supervisors and deposit insurance systems? Before discussing the regulatory issues with respect to Nordea, we first describe the history of the organization. The Nordea Group was established in 2000 but derives its origin from banks and insurance companies from the Nordic region as far back as the early 19 th century. The first cross-border merger took place in 1997, when the Finnish Merita Bank and the Swedish Nordbanken formed MeritaNordbanken. The merger of MeritaNordbanken and 14

16 Danish Unidanmark in 2000 was followed by the acquisition of the Norwegian bank Christiania Bank og Kreditkasse to create the largest financial institution in the Nordic area. During 2001 Nordea acquired Swedish Postgirot Bank. Since December 3, 2001 the entire Group is operating under the Nordea brand. Details of the transactions are as follows. On October Merita and Nordbanken merged to form a leading bank group serving the Nordic countries and Baltic Region. The Swedish exchange-listed company (Nordbanken Holding) would own shares corresponding to 60 percent of the capital and 50 percent of the votes in MeritaNordbanken. The Finnish exchange-listed company would own shares corresponding to 40 percent of the capital and 50 percent of the votes in MeritaNordbanken. On November 26, 1997 Merita Abp accepted the public offer from Nordbanken Holding, and the companies merged. Initial announcement of Nordbanken to acquire all the Merita shares and the Merita Convertible bonds was made on September 20, Exchange ratios were based on and reflect the 40/60 economic relationship agreed in the 1997 Co-operation and Merger Agreements providing for the creation of the MeritaNordbanken Group. On March 6, 2000 the Boards of Directors of Nordic Baltic Holding and Unidanmark announced their intention to merge the MeritaNordbanken and Unidanmark Groups to form the leading financial institution in the Nordic and Baltic region. The share exchange gave Unidanmark shareholders 30 per cent and MeritaNordbanken 9 For each Merita Share 1.02 Nordbanken Holding Shares was received. For a Merita Convertible Bond with a Nordbanken Holding Convertible Bond with nominal value of FIM 10,000 nominal value of EUR 1, was received. 15

17 shareholders 70 per cent of the New Group 10. A special dividend of DKK 10 per Unidanmark share was subject to the completion of the exchange offer. The exact number of shares in Unidanmark for which the share exchange offer was accepted is 68,464,774, corresponding to 97.1% of the total number of shares in Unidanmark, excluding the holding of own shares, the cancellation of which was resolved by the Annual General meeting on 21 March On the exchange of Unidanmark acceptance shares, Nordic Baltic Holding issued 869,776,488 new shares of SEK at 3.50 each. On March 6, 2000 Nordic Baltic Holding (NBH) AB (publ) and Unidanmark A/S announced the merger of the two groups. An initial public offering to buy all the shares of Christiania Bank og Kreditkasse was made on September 24, On October 16, 2000 Christiania Bank og Kreditkasse became part of the Nordic Baltic Holding Group. MeritaNordbanken Plc acquired all the shares of Christiania Bank og Kreditkasse ASA at the price of NOK 49 per share. From October 16, 2000 interest compensation for all acceptances was 7.5 percent per annum. The due diligence condition was waived. The Norwegian Government Bank Investment Fund sold its shares in Christiania Bank og Kreditkasse to MeritaNordbanken for NOK 49 per share. It sold its 191,000,000 shares, constituting percent of the shares in Christiania Bank og Kreditkasse, to MeritaNordbanken. The price of NOK 49 per share valued Christiania Bank og Kreditkasse at NOK 27 billion. Table 1 from a Bank of Norway publication shows the traditional perceptions of a host country s responsibility towards foreign branches and subsidiaries. This division of 10 The shareholders of Unidanmark were given new shares in Nordic Baltic Holding for each Class A and B share in Unidanmark. 16

18 supervisory responsibility in the EU was discussed above. Table 2 from the same source also illustrates a very important point. Regulators are more concerned with supporting a bank in a financial crisis if the parent bank and the host country subsidiary are important banks in their respective countries.(too Big to Fail) Even in its current structure the regulators could potentially face a difficult problem if a subsidiary bank in one country encounters trouble. Nordea itself likely would try to prop up that bank by providing funding from the other banks. However, at what point does Nordea give up and let the one bank fail? The regulator from the country of the failing bank would have the responsibility of saving the failed bank or helping it in transition. Does the regulator call for help from the other banks in the organization and does it ask the other regulators to step in and help? Regulators have devised schemes for these scenarios, but frequently something arises that was totally unanticipated. We noted above that the parent bank and the home country supervisor may collude to shift bad loans to the failing subsidiary and thereafter to abandon it. With the branching model different complications can arise. If loans in one country default widely and depositors withdraw their funds, what does the home office in Sweden do to protect the whole organization? Does management close up in one country and forget about that country in the future? However, in the case of Nordea power and ownership is divided into four relatively equal shares by country. Thus, abandoning one country is probably not an option within the framework of the bank. Primary regulatory oversight will be in Sweden. How much will the Swedish authorities care about what happens in one of the other countries? Is Nordea becoming Too Big To Save? Will the four national regulators be able to coordinate policies? Is the 17

19 supervisory regime envisaged by the EU appropriate for banks of Nordea a type? These are important issues that the regulators must consider as they decide whether to approve the reorganization of Nordea again. The regulatory regime facing Nordea in its current subsidiary organization is one of cooperative supervision, while deposit insurance systems are country specific. Clearly this organization of supervision does not correspond to the functional organization of Nordea, although it does correspond to the legal entities created by Nordea. To supervise effectively the four Nordic supervisors have organized a Nordea committee with involvement from all countries. This committee has then created a number of joint subcommittees. It is hard to tell how well this structure works but there are reasons to be somewhat skeptical. One reason is that the supervisory approaches of the four countries are as different as the business cultures in the four countries. Sweden s FSA is by others considered relatively lax, because it avoids very detailed supervision of the balance sheet in favor of supervision of procedures and incentives in the organization. Finland and Denmark on the other hand have the reputation of being more detail oriented and thereby more interfering in decisions affecting risk. Nordea spokes-persons argue that the total time spent with supervisors is greater now than the time spent by the four individual banks in total before the mergers However, the problems of dealing with four tax systems seems to be greater than the problems of dealing with supervisors. Another problem with the current structure is that Nordea is large enough in each of the countries to be considered Too Big To Fail. In Finland, its market share is as high as 40 percent. The smallest share is in Norway with 15 percent. Thus the 18

20 involvement of all four countries in the regulation and the supervision implies that a relatively broad bail out is the only feasible consensus solution in case of distress. On the face of it the branch structure, if implemented, simplifies both Nordea s organization and the supervisory responsibility. Clearly, greater correspondence between legal and functional organizations is an advantage from an internal efficiency point of view. VI. Towards Enhanced Market Discipline with a Branch Organization The Nordic regulators face a choice among different models for supervision of a reorganized Nordea as a bank with headquarters in Sweden and branches in Denmark, Finland and Norway: (i) They can continue with the current inter-supervisory committee approach, while depositors are covered by host country deposit insurance systems. (ii) The inter-supervisory committee approach continues while depositors in all host country branches become covered by the Swedish deposit insurance system. (iii) Supervisory responsibility becomes entirely Swedish while depositors in all host country branches become covered by the Swedish deposit insurance system. (iv) A fourth model would place supervisory responsibility in Sweden and deposit insurance in the host countries. (v) Establish a Nordic supervisory authority and deposit insurance system. When deciding among these alternatives supervisors must worry about information availability and costs of supervision, crises management procedures, and incentives of supervisors created by their responsibility relative to insurance systems and tax payers. In 19

21 addition, the choice should consider dynamic developments towards efficiency of the financial system. We can rule out the idea of a Nordic supervisor to begin with as completely infeasible. If anything, an EU or EMU supervisor could be envisioned in the future, but such an institution is far off. It can also be noted that Norway is not a member of the EU, while only Finland is a member of the EMU. We have to consider national supervisory authorities as the realistic institutional structure for the foreseeable future. The lender of last resort (LOLR) function is similarly a national responsibility of the central banks 11. Model (iv) can also be ruled out on the grounds that the assignment of supervisory responsibility does not coincide with risk-taking of deposit insurance systems and taxpayers. In the US, the branches of foreign banks working with the US deposit insurance system face restrictions with the implication that the advantages of branches are lost. The Swedish FSA seems to favor a structure between (i) and (iv). It is willing to take primary supervisory responsibility with substantial involvement of the other FSAs, but unwilling to have all depositors covered by the Swedish deposit insurance system. The risk facing Swedish tax payers is potentially very large. However, the EU deposit insurance directive places responsibility for deposit insurance in the home country. Denmark, Finland and Norway are likely to accept either model (i) or model (ii). In both cases, the inter-supervisory committee responsibility continues, but the responsibility for deposit insurance differs. Under EU rules for deposit insurance, it would seem that model (i) where deposit insurance is a Swedish responsibility is the most realistic alternative. 11 In the EMU national central banks are formally LOLRs, but the ECB must become involved if liquidity is to be increased. 20

22 Focusing the responsibility to one supervisor in the country responsible for the insurance of all the depositors (model iii) can create what one may call a more incentive compatible supervisory structure than the one existing now. The organization of regulation can also become more transparent and simple with clear assignment of responsibility. Market discipline on the bank s behavior may also be enhanced. These advantages do not come automatically, however. Substantial conflicts of interest among regulators can arise. To begin with the LOLR role within the bank safety net, this role could be centralized to the home country whether liquidity problems originate in the home country or one of the host countries. If the host country central banks retain LOLR responsibility, they can hardly be expect to refrain from involvement in supervision either directly or through the host country FSA. Since Sweden is outside the EMU it can serve as a LOLR without involvement from the European Central Bank. A liquidity infusion in Sweden for purposes of supporting the bank as a whole may lead to fears of a Swedish depreciation. Such fears would be un-founded, however. The liquidity claim on the Swedish central bank would be transferred by the bank s head office to the branches in the country where liquidity is needed. To obtain the liquidity infusion in domestic currency the branches would go to the local central bank. In the end the local central bank would hold a claim on the Swedish central bank, while the latter would hold bank assets as collateral for the liquidity infusion Thus, the home country LOLR-central bank would take the risk of lending to the bank, while host country central banks would only face the risk of holding a claim on the home country central bank. Any potential inflationary consequences of the LOLR operation would appear in the country where liquidity is added. 21

23 The LOLR operation of the Swedish central bank would have monetary consequences in the host country of the branches. This implies that LOLR operations require central bank coordination whether the LOLR responsibility is placed in the home or the host country. One concern with model (iii) for the supervision of the branch structure is that the FSAs in the host countries (Denmark, Finland and Norway in our case) will not be willing to live by the principle of mutual recognition, and thereby leave the responsibility with the home country FSA (Sweden) in spite of the fact that risk is taken by the home country. The mirror image of this concern is the Swedish FSA is not willing to take the responsibility for depositors and financial systems in all four countries as the EU directives state. The LOLR risk and the risk faced by the deposit insurance system may be considered too large. At the same time the other supervisors may not be willing to simply give the Swedish FSA a green light for supervision of large parts of their banking systems. In spite of these concerns we argue that there are great advantages to a regulatory regime as in model (iii) wherein the Swedish FSA accepts supervisory responsibility for the whole bank and the Swedish deposit insurance scheme covers depositors in all countries, if at the same time the FSAs in the other Nordic countries accept the role of the Swedish FSA. The main advantage of this system is that it could lead to enhanced market discipline and a more efficient financial system. Statements to the effect that depositors and other creditors are not protected beyond the explicit, partial insurance scheme become credible. The reason is that the Swedish supervisor and government do not want to bail out banking operations outside Sweden. The bank is Too Large to Save in this 22

24 sense. If statements about the limits of the protection of the bank are credible, market discipline is likely to have a strong effect on the bank s behavior with respect to risk taking and capital structure. Thereby, so called moral hazard problems in the bank s risk taking are reduced substantially. The greater efficiency of the financial system under model (iii) would not be realized without reforms of supervisory organizations and crisis management procedures. The favorable outcome requires that the non-swedish FSAs can rely on the Swedish FSA to treat all branches fairly in a crisis situation, and that they have faith in the Swedish FSA as supervisor and head crisis manager, They need to accept the role of market discipline as well. If this faith and acceptance does not exist, the host country FSAs may intervene in a crisis to take over and bail out the branches in their countries. If markets expect this to happen, then the market discipline is going to be weak and the banks incentives become biased towards excessive risk taking. In order to prevent such an outcome the following aspects of the supervisory and crisis management framework need strengthening: 1. The Swedish FSA s rules for resolution of a crisis in the bank need to be clear and credible ex ante. These rules need to include binding measures commonly called prompt corrective action specifying the measures that will be taken when the capital ratio of the bank falls below certain triggers The rules for prompt corrective action must assure all countries involved that the intervention will be fair in relation to all branches and creditors independent of country. 12 See also Mayes (2004) The Role of the Safety Net in Resolving Large Cross-border Financial Institutions, Bank of Finland Working Paper. 23

25 3. For the home country supervisor to obtain credibility as the supervisor of branches in all countries, the FSAs in the host countries need to be informed about all supervisory activities and the results of these activities. The responsible supervisor must be able to obtain local expertise from the other supervisors upon request. Responsibility must not thereby be shifted towards the host countries, however. One solution for the home country supervisor is to set up local branches with local employees. 4. The FSAs in the host countries must contribute to the credibility of the regulatory regime by making it clear that they take no regulatory, supervisory, or crisis resolution responsibility, but they accept the ex ante determined rules for structured intervention and partial deposit insurance. 24

26 References Berger, A. et al (2000). Borchgrevink, H. and T.G. Moe (2004), Håndtering av finansielle kriser i grensekryssende banker (Management of Financial Crises in Cross-border banks), Bank of Norway. Dermine, J. (1999). The Economist (2005), February 19. European Shadow Financial Regulatory Committee, (ESFRC) (2004a), Challenges to Financial Regulators in the Accession Countries, May (2004b), Towards New Bank Capital Requirements in Europe Oct 24. Goldberg, L. and A. Saunders (1980) (1981a) (1981b). Goodhart, C.A.E. Multiple Regulators and Resolutions. Mimeo, Financial Markets Group, LSE. Grosse and L. Goldberg (1991). Herring, R. (2004), International Financial Conglomerates; Implications for Bank Insolvency Regimes, Working Paper, The Wharton School of the University of Pennsylvania. Kimminger, M.H. (2004), Deposit Insurance and Bank Insolvency in a Changing World: Synergies and Challenges. IMF Conference, May 28, Washington, D.C. Kowalski, T., E. Kraft, A. Mullineux, V. Vensel and C. Wihlborg (2003). Lastra, R. (2003), Cross-border Bank Insolvency: Legal Implications in the Case of Banks Operating in Different Jurisdictions in Latin America. Journal of International Economic Law 6(1). Mayes, D. G. (2004), The Role of the Safety Net in Resolving Large Cross-border Financial Institutions, Bank of Finland.

27 Søderberg, A.M. and E. Vaara (eds) (2003), Mergers Across Borders; People, Culture and Politcs, CBS Press, Copenhagen. 1

28 Tables and boxes from Håndtering av finansielle kriser Iigrensekryssende banker (Management of Financial Crises in Cross-border banks) by Henrik Borchgrevink and Thorvald Grung Moe, Bank of Norway. Table 1. Traditional perceptions of host country s responsibility towards foreign branches and subsidiaries. Authorities in the host country are responsible for: Solvency Assessment Liquidity Support (Central Bank) Net capital support (Political authorities/department of Finance) Investment Protection (Protection Fund) Subsidiaries Branches X X X X X (X) 1 1 Within the European Free Trade Association (EFTA) area, branches of credit institutions from another EFTA state are entitled to buy supplementary coverage in the host country s investment guarantee fund if the host country s investment guarantee scheme is better than the scheme in the home country the branch is a member of. Table 2. Home country s and host country s viewpoints on support in a financial crisis in a cross-border bank. The parent bank is a major bank in the home country The parent bank is a minor bank in the home country The bank is a major bank in the host country 1. Common interest in support 3. Different viewpoints on support The bank is a minor bank in the host country 2. Different viewpoints on support 4. No support