1. Research
  2. Products & Topics
  3. Periodicals
  4. Germany Monitor
November 20, 2012
Region:
The political dynamics in Europe have shifted against universal banks in recent months. This is a dangerous development that threatens the key role such banks play in modern economies and risks putting European banks at a competitive disadvantage to their peers in the US and Asia. [more]
Universal banks: Optimal for clients and financial stability Current Issues Global financial markets The political dynamics in Europe have shifted against universal banks in recent months. This is a dangerous development that threatens the key role such banks play in modern economies and risks putting European banks at a competitive disadvantage to their peers in the US and Asia. There are three key advantages of the universal banking model: — Broad range of services for customers: Universal banks have evolved over the past 200 years in parallel with the industrialisation of developed countries. They provide customers with comprehensive financial solutions in a “one-stop shop” that can offer tailor-made services, higher volumes of credit and lower funding costs than narrower “specialist banks”. This is particularly attractive to large, internationally active companies such as those that have powered Germany’s export-driven economic success. — Lower costs for customers and the real economy: Universal banks are able to leverage revenue and cost synergies through economies of scale and scope. These benefits are passed on to universal banks’ customers and investors. Ultimately these benefits lower the costs of finance for society as a whole. — Greater financial stability: Universal banks tend to be more stable thanks to the diversification of their operations. For example, since 2000 and throughout the cycle, post-tax returns on equity at large universal banks have been more stable than at specialist banks. Given their broader coverage, universal banks are also better positioned to monitor the financial health of specific clients as well as to spot unsustainable risk accumulation across financial markets. The debate over the business model, while politically popular, may distract policy- makers from much more relevant issues such as the implementation of Basel III, effective restructuring and resolution regimes for banks, and effective macro- prudential supervision. Such measures would improve the overall stability of the financial sector, while also ensuring that European universal banks can continue to provide the full range of services required by companies based in the region. Author Jan Schildbach +49 69 910-31717 jan.schildbach@db.com Editor Bernhard Speyer Deutsche Bank AG DB Research Frankfurt am Main Germany E-mail: marketing.dbr@db.com Fax: +49 69 910-31877 www.dbresearch.com DB Research Mana gement Ralf Hoffmann | Bernhard Speyer November 20, 2012 Universal banks: Optimal for clients and financial stab ility Why it would be wrong to split them up -15 -10 -5 0 5 10 15 20 25 30 2000 01 02 03 04 05 06 07 08 09 10 11 H1 12 Investment banks Universal banks Commercial banks Universal banks' ROEs are more stable DX Average post-tax ROE in %, selection of large European & US banks Sources: Company reports, DB Research Universal banks: Optimal for clients and financial stability 2 | November 20, 2012 Current Issues Universal banks: Optimal for clients and financial stability 3 | November 20, 2012 Current Issues 1. Introduction The financial crisis has clearly shown the need for comprehensive reforms in global financial markets, particularly in banking. Reforms were needed in a vast number of areas, from regulation of capital and liquidity levels and improve- ments in market infrastructures to banks’ business practices and system-wide (so-called macro-prudential) supervision, to name only a few. These regulatory and bank-driven initiatives targeting virtually every aspect of banks’ activities and the financial system as a whole have since made substantial progress. This progress notwithstanding, the fundamental adjustment of a system as large and complex as the banking sector takes time, of course. In recent months, however, a new tone has entered into the reform debate especially in Europe, focusing on the question as to whether, in addition to the numerous measures already implemented or at least undergoing negotiation, a radical “big bang” is needed with regard to the traditional structure of the continental European universal banking model. More specifically, these initiatives call for a separation of (loosely defined) investment banking activities from deposit taking; either in the form of a complete separation or in the form of a requirement to establish holding companies with separate capital, funding and governance structures for commercial and investment banking operations. It is a common characteristic of many of these initiatives that they do not present a balanced analysis of the universal banking model but instead concentrate one- sidedly on the purported risks stemming from this model. This paper aims at closing part of that gap by exploring the benefits a universal banking system offers to all its stakeholders: a) private household clients b) business and institutional clients c) the banks’ owners d) the financial system as a whole e) governments and taxpayers It starts with a discussion of the concrete advantages integrated universal banks can offer their clients, followed by an analysis of the beneficial effects on financial stability and, as a consequence, for citizens and taxpayers. Some preliminary remarks In the following, a “universal bank” is defined as a financial institution which may offer – and often does offer – a broad range of banking services, including most, though not necessarily all of the following: retail, business and corporate banking, services for (financial) institutional clients, asset and wealth manage- ment, payment services and investment banking. By contrast, the terms “specialist banks” in a “split banking system” refer to banks which cover only few of these services. It is important to note that it is a hallmark of universal banking systems that banks are allowed to offer a broad variety of services under one roof, but are not compelled to do so. As a consequence, universal banks, in reality, come in many shapes and forms. While some activities, such as payment services, are a regular feature of almost all universal banks, others, such as asset and wealth management, are less frequent. Similarly, most universal banks limit themselves to parts of the full spectrum of investment banking. This great diversity of universal banks underlines that a one-size-fits-all approach to the analysis of specific banks or groups of banks and specific supervisory and regulatory actions is unlikely to be conducive. Universal banking in the US 1 Historically, in most people’s minds a split banking system has been most forcefully associated with the US. There, split banking was introduced in the wake of the Great Depression as the political investigations into the causes of the stock market crash and the ensuing depression unearthed substantial evidence for mis- selling of financial products, front-running and other practices that had harmed clients. The Glass-Steagall Act of 1933 responded to these failures by imposing a strict separation of commercial and investment banking. As markets developed and were liberalised during subsequent decades, regulatory arbitrage increasingly undermined the strict separation established by Glass-Steagall. After several changes which modified the Act in response, Glass-Steagall was finally repealed by the Gramm-Leach-Bliley Act (GLBA) of 1999. While GLBA is often seen as the result of successful lobbying by interested parties in the financial sector, it is closer to the truth that Glass-Steagall had become so riddled with holes that it was neither effective any longer nor the resulting banking structures efficient and easy to supervise. GLBA effectively tried to establish a new consistent philosophy on banking structures that corresponded to the realities of modern financial markets. No doubt, after GLBA US supervisory structures continued to display deficiencies, failing to spot emerging trouble. But, if anything, events during the 2007-09 financial crisis have proven that the decision to repeal Glass-Steagall was the right one: had it not been for the reforms of GLBA, one of the key and most effective elements of the US authorities’ strategy to deal with the fallout from the crisis – viz. the absorption of failed investment banks into stable universal banks and the granting of bank holding status (and hence access to emergency liquidity from the Fed) to broker-dealers – would not have been possible. Universal banks: Optimal for clients and financial stability 4 | November 20, 2012 Current Issues Universal banks as just described are the predominant financial institutions in most countries of Europe as well as in some Asian economies. In the US, since the 1930s, the banking sector has traditionally been shaped by specialist banks – either commercial or investment banks, with separate asset managers (see text box 1). Yet since the 1990s, this distinction has become ever more blurred and today, most large financial institutions can indeed be considered to be universal banks. Universal banks often arose when banks which initially were more focused on specific groups of customers i) broadened their operations to cater to the needs of other clients as well or ii) had to develop new expertise in further areas of financial markets when their core clients’ requirements grew in scope and scale. To give a few examples: agricultural banks in France or the Netherlands built up capital markets operations once their customers became more strongly exposed to fluctuating global commodity markets. German commercial banks which started as trade finance houses and corporate banks later opened up to private retail clients. And retail and corporate banks in developing Asian countries expanded into investment banking, established an international presence and became players in private banking when their economies started to catch up with Western countries, foreign trade grew dramatically and households accumulated increasing wealth. Specialist banks became prominent mainly due to regulatory reasons, i.e. when they were forced by law, whereas universal banks usually evolved gradually and independently from each other in many different countries, suggesting that they possess some “natural” advantages over other types of financial institutions which this paper will look at. For analytical (as well as regulatory) purposes it is also necessary to delineate “universal banks” from “large banks” (in the sense of these banks being systemically important) as the two terms are often – and wrongly – used interchangeably. However, there is no 1:1 relation between large and universal banks: universal banks can, depending on their franchise and the scope of business lines they cover, be large or small. Of course, the coverage of more than one business line on an efficient scale will presuppose a certain minimum size; but this scale will, in most business lines, not necessarily entail that the bank in question reaches a systemically relevant size. In turn, financial institutions can well be large and systemically important, but do not necessarily have to be universal banks. The example of a specialised bank running a country’s payment system and central clearing would be a pertinent example. 2. Why universal banks are beneficial to clients (and thus the economy as a whole) I) Benefits for all groups of clients Universal banks provide a number of benefits for all client groups. These benefits are the result of the particular operating model as well as the organisational structure of universal banks. The relative weight of each of these benefits will differ depending on the specific operating model chosen by any given universal bank. a) Convenience Universal banks provide “one-stop shopping” for clients in need of a variety of financial services. In modern economies, the financial needs of both households and certainly firms are diverse, as they try to find financial solutions to a broad spectrum of problems such as payment methods, funding, savings for a variety of purposes (old age, education, real estate, consumption etc.), reducing Universal banks: Optimal for clients and financial stability 5 | November 20, 2012 Current Issues uncertainty and risks. Often, these financial needs are interlinked (consider e.g. a medium-sized firm which i) needs to finance new equipment, ii) must establish a reliable payment system with foreign suppliers providing input used on that equipment and iii) wants to hedge against the risk of FX volatility affecting the price of that input). One-stop shopping lowers search and, hence, transaction costs for clients. In addition, client-bank relationships may be more stable if they cover the whole range of banking services whereas specialist banks would risk turning loyal into “walk-in customers”, with negative effects on the banks’ ability to provide the most efficient financial solutions (because they have only incomplete knowledge of the clients’ financial needs) and potentially higher costs for customers (because of less scope for cross-subsidisation between different product types). 1 b) Tailor-made financial services Due to their comprehensive knowledge of the customers’ financial situation and needs, universal banks may be able to provide superior advice and better suited products than a specialist bank which must rely on the much narrower range of services it offers. This may be particularly relevant for clients requiring a comprehensive set of relatively sophisticated (and complex) corporate finance solutions. c) Lower funding cost due to lower profiling cost for the bank Universal banks only need to establish a customer profile once, including information on credit history, P&L and balance sheet data (for firms) and income and asset levels (for individuals), personal characteristics etc. Single-role or specialist banks have to gather data and other information separately, duplicating efforts the cost of which will ultimately have to be borne by clients. In addition, the above-mentioned argument about (in)complete knowledge of a client’s financial needs pertains to risk aspects too, of course: the more intensive and the broader the interaction of bank and client, the better the bank’s knowledge about the client’s risk profile. d) Lower funding cost due to the bank’s lower funding cost Thanks to their more balanced (and hence stable – see chapter 3a) below) business model, universal banks may also be perceived to be less risky by financial markets than specialist banks (of either nature – commercial or investment banks). This would lead to debt or equity investors charging lower risk premia for universal than for specialist banks and allow the former to pass on at least part of this to their customers, in the form of lower prices or interest rates. Note that this holds true irrespective of any potential implicit subsidy stemming from investors’ assumptions of possible sovereign support that could be forthcoming to stabilise a troubled systemically important institution (the “too- big-to-fail” argument). 2 e) Lower funding cost due to the bank’s lower operating cost Similar to the previous argument, universal banks’ clients may benefit from the banks being able to grant better pricing thanks to their stronger underlying profitability due to synergies on the revenue as well as the cost side (see chapter 3b) below). f) Increased availability of credit to borrowers, higher returns for savers Universal banks are better able than specialist banks to efficiently allocate capital, i.e. channel surplus funds from one side to demand for credit on the 1 Bundesbank Vice President Sabine Lautenschläger has “doubts about a split banking system”, reports Börsen-Zeitung (2012). 2 Furthermore, current regulatory initiatives aim at removing exactly this potential by i) imposing a capital surcharge on systemically important financial institutions and ii) establishing pre-defined procedures for orderly restructuring and possible resolution of failing banks. See chapter 4). Universal banks: Optimal for clients and financial stability 6 | November 20, 2012 Current Issues other side. They are not subject to limitations (beyond the applicable capital and liquidity regimes) about where they invest and which clients they lend to, in contrast to specialist banks which are limited in the range of clients and purposes they can lend to – which can mean that excess liquidity is simply “trapped” inside the bank rather than being deployed to finance an economy’s growth. 3 By contrast, a universal bank may e.g. use retail deposits to fund loans to corporations or invest abroad. Incidentally, this also implies that if many countries were to introduce split banking, the efficient flow of funds between surplus and deficit countries and, by extension, cross-border adjustment processes would be impeded: deficit countries could well be cut off from international funding and would be unable to satisfy their demand for credit. Clearly, this would be to the detriment of both parties: lenders (i.e. savers) would achieve lower returns on their capital while borrowers would be starved of funds and be forced to pay higher interest rates than necessary. A capital market fragmented along national borders would thus be as harmful to individuals, the economy and overall welfare as traditional barriers to international trade in goods and services – which, for good reasons, have steadily been reduced over recent decades. Note that this is not a theoretical problem, but a real-life issue: there is virtually not a single country in the world where domestic supply matches domestic demand – for both goods as well as capital, given different preferences at broadly similar interest rate levels (which naturally evolve in an integrated global economy). By definition, countries with positive or negative trade balances (i.e. current account surpluses and deficits) have a corresponding negative or positive capital account balance, as countries with an export surplus accumulate claims on foreign counterparts, i.e. they accumulate foreign assets and vice versa. German households e.g. tend to be strong savers, resulting in a persistent surplus of funds that are not used entirely for domestic investments but made available in the international capital market for countries where investment prospects are attractive but corresponding savings are insufficient. Under a split banking system, this money would presumably be trapped within the German banking system, at a lower “productivity” (i.e. return on capital for German households), while at the same time useful investments e.g. in a young and quickly growing emerging market (which would also create demand for German exports) could not be undertaken for lack of funds – a disadvantage for both sides. 4 g) Spreading financial innovations Universal banks are the natural conduits to spread financial innovation to a broad spectrum of clients and thus bring the benefits of innovation to society as a whole. Essentially, the benefits from the universal banking model stem from two sources here: first, products initially developed for a specific (usually a sophisticated) client group such as multinational companies or institutional investors can be adapted at low marginal costs to suit the financial needs of other, broader client groups. FX hedging products and cash management services, which were originally targeted at large multinational corporates, but are now used widely, are a good example. Second, universal banks are better suited to help innovations achieve a breakthrough in financial markets as they can transfer their reputation earned with other products onto the new product. To take an example: clients would naturally have been sceptical about the reliability and security of online payments if these had been offered by an unknown internet start-up firm, but were willing to use such products once they were offered by banks with a proven track-record in off-line payments. 3 Deutsche Bank analysts estimate a cash trap of about EUR 2 tr in case Vickers-style subsidiarisation/ring-fencing were to be adopted across the EU. 4 For further analysis of the financial stability implications of restricting the cross-border usage of surplus funds, see chapter 3f) below. Universal banks: Optimal for clients and financial stability 7 | November 20, 2012 Current Issues There is a second reason why universal banks may be able to channel more credit – and particularly in a smoother way – to borrowers: as universal banks generally run more diverse operations than narrower competitors and business cycles for different products never overlap completely (see chapter 3a) below), universal banks tend to achieve more stable results and consequently face fewer funding and capital constraints especially in difficult times. Overall, this allows for a more robust provision of financing to the real economy. Academic research – despite gaps due to considerable data constraints – supports several of the above arguments. Some studies, e.g., argue that the internal allocation of risk and capital within universal banks allows for “natural hedging” between complementary activities, thereby reducing risk-transfer costs and in turn possibly firms’ funding costs. 5 Most of the empirical research, however, has focused on the benefits or disadvantages of universal banks compared with pure investment banks for clients trying to raise equity or debt capital in financial markets. Studies have shown that pre-Glass-Steagall in the US, corporate bonds that were placed by universal banks subsequently had lower default rates than those placed by investment banks – especially when looking at high-yield issuance. At the same time, funding costs (i.e. bond yields) were lower when issuers hired universal banks rather than investment banks to lead-manage corporate bond issues – with the effect being especially pronounced in the presence of large information asymmetries, which are more likely for new rather than frequent issuers (or for smaller firms that are unknown to investors rather than larger enterprises). Both of these results were largely confirmed after the partial repeal of the Glass- Steagall Act in 1987. 6 This suggests that instead of transferring poorer credit risks to the capital markets (and keeping better risks on their balance sheet), universal banks in fact award a seal of quality. Lead management of a corporate bond issue by a universal bank sends a positive signal to the market regarding the issuer’s quality. The underlying reasons are evident: first, investors can trust a universal bank’s signal regarding the issuer’s creditworthiness, because they know that the universal bank has skin in the game itself due to its ongoing relationship and, hence, exposure to the firm in question. Second, universal banks are also more successful than investment banks at screening borrowers, resulting in investors suffering fewer losses. Regarding equity issuance, companies undertaking an IPO are also usually better off with universal banks. Underpricing tends to be lower than for deals which are arranged by investment banks. Furthermore, a previous corporate banking (i.e. lending) relationship benefits the issuer more than a previous relationship in the issuance business. For secondary offerings, hiring a universal bank instead of an investment bank often lowers the client’s underwriting costs (in addition to reducing lending rates), as universal banks in turn benefit by receiving additional business from the issuer – and are able to pass on savings from informational economies of scope (e.g. lower profiling cost). 7 Overall, empirical evidence therefore strongly suggests that universal banks’ advantages in debt and equity issuance may significantly outweigh any potential disadvantages (such as conflicts of interest if banks help companies issue bonds to pay them back their loans or if banks sell overpriced shares from an IPO to poorly informed retail clients) versus pure investment banks – both for investors and issuers. 5 See e.g. Cumming and Hirtle (2001). 6 See also Ang and Richardson (1994), Kroszner and Rajan (1994), Puri (1994) and Puri (1996) as well as Gande et al. (1997). 7 See also Drucker and Puri (2005) and Schenone (2004). Universal banks: Optimal for clients and financial stability 8 | November 20, 2012 Current Issues II) Benefits for specific groups of clients Apart from offering benefits to all client groups, universal banks are also beneficial for a number of very specific client groups, in particular large, multinational firms and institutional investors. a) Benefits for large companies Decades of consolidation due to an increasing role of fixed costs and globalisation have led to the emergence of a number of large corporations which operate on a global scale, in very diverse business areas and have equally diverse funding needs. As they aim for well-balanced balance sheet structures, especially on the liabilities side, they require financial partners that are able to provide them with a large set of options to finance new projects and hedge risk (on the importance of large and international companies, see text box 2). Of course, this covers more than traditional loans, bonds or equity, but includes a large variety of derivatives as well. While a specialist bank may be able to supply either traditional lending products or have the placing power to successfully issue equity or debt securities, only an integrated universal bank can manage to offer the full range of financial services and advise on which instrument would be the most appropriate in a given situation. Specialist banks, on the other hand, due to their limited scope force the company to decide for itself which instrument to use, before searching for the right bank to implement its plan. Granted, many firms run multiple bank relationships, but most use a single “relationship bank” as their main financial partner which continuously provides a comprehensive set of solutions. Only if necessary, and for competitive reasons, other banks are added to the anchor institution. Without the latter, however, having parallel (and competing) relationships with commercial and investment banks will lead to greater complexity and a less efficient use of financial products. Secondly, large companies often require big-ticket funding transactions whose risk most banks are unable or unwilling to bear alone. Hence the impressive size of the syndicated loan market, which has recovered from the financial crisis to reach a total new business volume of about USD 3 tr globally per year (see chart 4 below). While in principle any bank, from a small regional savings institution to a large investment bank, may participate in this form of risk-sharing, in reality large universal banks dominate the market. Of course, this is particularly true of the bigger deals (> USD 1 bn), which account for nearly half of all syndicated loans (see chart 5). The market for those large transactions is highly concentrated: the top 10 bookrunners’ combined share amounts to roughly 80%. Among the top 10, all but one (i.e. No. 10) are large universal banks with both significant corporate as well as investment banking franchises. It is important to note that neither pure commercial banks (without strong capital markets expertise) nor pure investment banks (many of which lack the balance sheet strength to carry a substantial part of the “risk burden” for a longer period of time) rank prominently among these leading syndicated loan arrangers. 8 Sources: Eurostat. 9 Sources: UNCTAD (2012) and EIM (2010). 10 Sources: IfM Bonn, DB Research. 11 Exports accounted for 50% of German GDP in 2011, according to Eurostat, vs. 27% in France, 29% in Italy and 32% in the UK. 12 Sources: Eurostat, DB Research. The role of large and international firms 2 In the EU manufacturing sector alone, there are about 16,000 companies which each employ more than 250 people and in total provide no less than 12 million jobs, thereby accounting for 40% of total manufacturing employment. Even more impressively, those large firms generate combined annual sales of nearly EUR 3.9 tr which is equivalent to 60% of the total output of the EU’s manufacturing industry. 8 But large enterprises not only play an important role because of their sheer size and weight; they are also on average a) more international and b) more innovative than SMEs and therefore better positioned as drivers of future economic growth. — At the world’s 100 largest non-financial transnational corporations (TNCs), foreign sales already account for an average 65% of total sales. On the other hand, for instance, only 25% of European SMEs (< 250 employees) export at all. 9 — Even in Germany – with its globally successful Mittelstand – large enter- prises (annual turnover > EUR 50 m) on average sell a much larger share of their products abroad (24% vs. 9%) than smaller firms, with the export share rising steadily in line with firm size. 10 At the same time, as German companies are generally more export- oriented than their peers in other (large) EU countries 11 , they also rely more on related investment banking (e.g. FX hedging) services and the impact on the German economy of a disruption of the integrated universal banking model would probably be disproportionately bigger than else- where. — Equally, research and development (R&D) in the private sector in Europe is primarily concentrated on large companies (> 500 employees) which account for three quarters of total R&D expenditures as well as 60% of R&D employment. 12 As a consequence of greater R&D intensity, larger firms also tend to be more innovative than smaller ones (see chart 3 below for Germany). Universal banks: Optimal for clients and financial stability 9 | November 20, 2012 Current Issues b) Benefits for international companies The argument for universal banks becomes even more compelling as we consider firms that are active across borders (and which are often also relatively large, see text box 2 above). They have to take into account an additional dimension to their business which comprises issues ranging from i) the use of trade finance and ii) the cross-border transfer of funds to be able to centralise liquidity management to iii) the hedging of exchange-rate risks, in addition to many other things that can have an impact on a company’s financial health such as legal and political risk and cultural differences. Banks with a reasonable local presence and/or sufficient expertise and the necessary capacity to provide solutions to these challenges can claim to make life easier for their corporate customers and offer valuable support. By nature, diversified universal banks (operating across borders) are best suited to serve the diverse financial requirements of such internationally active firms. Foreign exchange services are an instructive case in point. Of course, FX cash and derivatives solutions can also be provided by pure investment banks, without a commercial banking backbone. Yet empirically, it is predominantly large corporate (universal) banks with strong international franchises that play leading roles in the FX market (see table 6 below). It is a rather concentrated market, with the 15 strongest institutions accounting for a combined share of nearly 90%. 10 of these top 15 in 2012 were universal banks, who covered almost two-thirds of the entire global FX business, whereas the five investment banks without a large corporate banking arm reached only 23%. In addition, the universal banks’ share has edged up in recent years, while that of investment banks has tended to move in the opposite direction. There may be three main reasons for this particular shape of the FX market: i) It is a low-margin, high-volume business – a combination that tends to favour large institutions and usually leads to consolidation; ii) A further push for market consolidation comes from large, internationally oriented companies’ need for providers which offer many different currency pairs. iii) The FX business also gives providers with better knowledge about the development of individual economies a competitive edge, as these players will be better positioned to predict currency movements. Universal banks with their broader-based client relationships may thus be better informed and hence more successful in offering FX services than “narrow”, stand-alone players in the FX market. In addition to supplying foreign exchange solutions, large cross-border universal banks also have a strong position in the market for international cash manage- ment and payment services for globally active companies. Pooling corporate cash reserves from diverse local businesses in multiple countries by using central transaction and clearing accounts can bring firms substantial savings, yet only few banks (mostly universal banks) command the resources necessary to organise instant liquidity transfers across international borders and to channel funds to places where they are urgently needed. c) Benefits for institutional investors Institutional investors, such as insurance companies, pension funds, sovereign wealth funds, central banks and hedge funds are a natural client group for universal banks as these offer, all under one roof, many of the financial services that are crucial for the success of these investors. The services include broker- age, securities lending, custody business, lending, repo business, hedging and the provision of research. Being able to source these services from a single institution provides tangible benefits to these institutional investors as it makes an integrated approach to liquidity and risk management easier to implement. This is becoming increasingly important against the background of regulatory changes, such as the EU’s Solvency II or AIFM Directives that are placing greater emphasis on institutional investors’ risk management capabilities. 0 5 10 15 20 25 30 35 40 45 0 10 20 30 40 50 60 70 80 90 5 - 49 employees 50 - 249 employees 250 - 999 employees ≥ 1,000 employees Share of firms with continuous R&D (lhs) Share of new products in total turnover (rhs) Germany: large manufacturing firms more innovative 3 %, 2010 Source: ZEW 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 2005 06 07 08 09 10 11 H1 12 Global syndicated loan market 4 New business volume, in USD tr Source: Thomson 40 50 60 70 80 90 100 30 35 40 45 50 55 60 2005 06 07 08 09 10 11 H1 12 Share of large transactions (> USD 1 bn) in all deals (left) Market share top 10 lead arrangers (right) Large banks crucial for large syndicated loans 5 % Sources: Thomson, DB Research Universal banks: Optimal for clients and financial stability 10 | November 20, 2012 Current Issues In addition, institutional investors may benefit from the unique breadth of universal banks’ distribution channels. To name just two examples: apart from using a universal bank’s securities services, mutual fund companies, e.g., can also have their funds distributed via the bank’s retail and asset and wealth management network; and insurance firms, in turn, may want catastrophe (“cat”) bonds, which they use to pass on insurance risk to financial markets, to be placed with their universal bank’s other institutional clients. Furthermore, many banks provide continuous liquidity in a large range of financial markets – including e.g. currencies, commodities and derivatives – by acting as counterparties to clients wanting to pursue financial investments, supplying or buying desired assets. This so-called market making activity is beneficial for end-users as it reduces transaction costs and, as a consequence, financing costs. Not every universal bank needs to be a market maker, of course, but there are benefits to clients if their bank is an active market maker. Market making ensures not only that orders can be executed with a smaller time-lag, but, more importantly that the spread between buy and sell prices (bid- ask spread) is lower. From the issuers’ point of view, i.e. firms and public-sector entities seeking finance, this is positive as (i) it tends to raise the issuance price, because potential buyers would otherwise lower their bidding price to make up for higher transaction costs; and (ii) it tends to increase demand for these assets, because potential buyers know that there is less price risk in a liquid secondary market should they wish to sell the asset subsequently. From an institutional investor’s point of view, lower spreads reduce transaction costs, thereby enhancing portfolio returns, and allow for better risk management as the portfolio’s composition can be adjusted at lower costs. 3. Why universal banks are beneficial to financial stability (and thus taxpayers) Beyond providing benefits for their customers, universal banks also contribute positively to financial stability through a number of channels. a) Diversification of revenues, assets and liabilities, supporting a bank’s resilience Universal banks by definition have a more diversified revenue structure and balance sheet than specialist banks, which may allow universal banks to better withstand a downturn in one specific market segment. Asset quality, for example, may be less volatile if a bank is invested in a greater variety of loans, debt securities and equities. Universal banks usually also derive a larger share of their revenues from fees and commissions which are partly driven by other factors than interest income. Equally, the total funding base may be more stable if different sources and kinds of refinancing can be tapped, from retail deposits to (un)secured bonds and repo markets. All in all, this diversification reduces the potential for extreme losses at the group level and thus the “tail risk” of an institution failing in spite of all preventive measures. 13 It is no coincidence that during the financial crisis of 2007-09 large universal banks were the only institutions – apart from governments – that were able to add to instead of subtract from financial stability by taking over failing competitors (mostly specialist banks). Examples can be found especially in the US, where prior to the crisis specialist banks had been more common than in Europe: investment banks Bear Stearns and Merrill Lynch, retail banks Washington Mutual and Countrywide Financial were taken over by JPMorgan and Bank of America, respectively. Of course, universal banks got into trouble as well: e.g. Citigroup in the US; RBS, Fortis and Dresdner Bank in Europe. Yet 13 See also Templeton and Severiens (1992). Global FX market leaders 6 Rank Bank Market share in 2012 (%) 1 Deutsche Bank 14 . 6 2 Citigroup 12. 3 3 Barclays 11. 0 4 UBS* 10. 5 5 HSBC 6. 7 6 JPMorgan 6. 6 7 RBS 5. 9 8 Credit Suisse* 4. 7 9 Morgan Stanley* 3. 5 10 Goldman Sachs* 3. 1 11 BNP 2. 6 12 Bank of America 2. 4 13 Soc Gen 1. 8 14 Nomura* 1. 3 15 Commerzbank 1. 3 * Investment bank without a sizeable international corporate banking franchise Source: Euromoney 6 Universal banks: Optimal for clients and financial stability 11 | November 20, 2012 Current Issues this came despite the stabilising effect of their broad and diversified operations, not because of it. Generally, management mistakes (incl. poor risk manage- ment) were the main reason for bank failures during the crisis, and no business model proved completely immune. However, for any given environment, the universal banking model tends to produce more stable and less volatile results (both to the upside and the downside) due to its more balanced business and funding mix and therefore greater resistance to both external shocks and internal errors (see empirical evidence below). This implies that had specialist banks been more common in Europe, the crisis would surely have resulted in many more bank failures. The predominance of the universal banking model in the EU prevented even more damage to the continental financial system – and lowered the cost of bank bailouts to taxpayers as it increased the industry’s loss absorption capacity. 14 Furthermore, given the asynchronous timing of the financial and the economic crisis (losses on structured credit products and other toxic assets occurred mainly in 2007 and 2008, while commercial banking was still flourishing; in turn, the surge in loan losses from the global recession hit in late 2008 and 2009, whereas investment banking revenues already surged again in 2009), universal banks were in a much stronger position to weather the downturns in the different market segments and to compensate setbacks by a relatively strong performance in other areas. Charts 7 and 8 provide empirical evidence for the beneficial effects of a diversified but integrated business model. Looking at post-tax return on equity of a representative sample of large banks in both Europe and the US, two results become relatively clear: 15 i) In the boom years as well as through the crisis since 2007, average ROE has been broadly similar for all three groups of banks. However, it seems as if in the new post-bubble and low-growth world, diversified universal banks can manage best. ii) The volatility of results (measured by the standard deviation of annual ROEs) is lowest at universal banks, followed by commercial and investment banks. This demonstrates that universal banks are indeed better able to balance different business cycles in banking and financial markets than specialist banks. In addition, given the ongoing crisis in some southern European countries which are home to several large commercial banks in our sample, the gap between universal banks and other types of banks may also grow rather than shrink in the next few years. Chart 8 shows how this counter-balancing effect has worked for some of the most prominent global universal banks – two from the US and two from Europe. From 2008 to 2009, in all four cases the result of the investment banking division(s) improved enormously, mostly turning from a loss into a large profit. At the same time, at all four banks the result of the other business lines combined (primarily retail and corporate banking, plus asset management) fell consider- ably, sometimes even shifting into negative territory. Hence, the combination of a broad range of banking services under one roof proved stabilising, smoothing profits and reducing volatility. Put differently: in times of crisis, universal banks provide an intrinsic “first line of defence”: profits (and surplus capital) of the group’s other business lines. Consequently, having separate “trading entities” within a bank holding structure such as those recently proposed by the European Commission’s High-level Expert Group under Erkki Liikanen 16 , may actually make matters worse because it forbids the rest of the bank to offer help. 14 See also Altunbas et al. (2011). 15 Further support comes from an analysis of risk-adjusted returns on assets – to eliminate potentially distorting effects from greater (balance sheet) leverage at some banks – which yields very similar results. 16 See also HLEG (2012). -15 -10 -5 0 5 10 15 20 25 30 2000 01 02 03 04 05 06 07 08 09 10 11 H1 12 Investment banks Universal banks Commercial banks Average post-tax ROE by type of bank 7 %, selection of large European & US banks Sources: Company reports, DB Research Universal banks: Optimal for clients and financial stability 12 | November 20, 2012 Current Issues Furthermore, most proposals for the separation of banking operations seem to be based on the fundamental assumption that trading activities are inherently more risky than traditional lending. This would be a short-sighted conclusion with potentially dangerous consequences: throughout history, the vast majority of financial crises was in fact caused by traditional lending booms, very often directed at real estate markets. And while it may be true that securities valuations are more volatile than loans that are kept on balance sheets, this makes risks involved in securities investments at least much more transparent, while credit risks from lending are mostly disguised by accounting practices which create a “stability illusion” that only falls apart when a credit bubble suddenly bursts. Universal banks’ advantages may be all the more pronounced if they operate not only in a wide range of business segments in banking and financial markets, but also in geographically different markets. Global economic cycles in the past few decades have shown ever stronger convergence in line with globalisation yet there still is much variation – which in addition may become wider again in the post-crisis years as some (western) countries need to deleverage, while growth in the emerging markets has not yet run its course (see chart 9). Hence, universal banks that are also globally active may benefit most from different strengths in different financial markets and different countries. 17 Beyond this general line of argument, which would apply to virtually any sector of the economy, the benefits of diversification play a particularly important role with respect to the banking industry. The reasons are the distinct characteristics of a bank’s business model and the handling of risk: risk is inherent in financial markets, or more precisely: financial markets are all about the creation of exposures and the distribution of the associated risk. Hence, it is impossible by definition to eliminate risk in the financial system, and policymakers and regulators can only decide about the system’s shape and set incentives for the level and distribution of risk within it. 18 17 Again, lower volatility also implies less upside in the boom. Diversified institutions will not be able to produce the same returns as (geographically limited) specialist banks that conduct all of their business in a market experiencing e.g. a credit bubble. 18 See also Weistroffer (2011). - 10 - 5 0 5 10 15 JPM 08 JPM 09 BoA** 08 BoA 09 BNP 08 BNP** 09 HSBC 08 HSBC 09 Investment bank Other operations Net profit* by business line, selected universal banks 8 Reported currency, bn Sources: Company reports, DB Research * For BNP and HSBC: pre - tax profit ** BoA adjusted to include Merrill Lynch; BNP adjusted to exclude Fortis - 28 - 6 - 4 - 2 0 2 4 6 8 10 12 1990 95 00 05 10 15 17 Advanced economies EMs: Asia EMs: CEE EMs: LatAm EMs: MENA Real GDP growth 9 % yoy Source: IMF Universal banks: Optimal for clients and financial stability 13 | November 20, 2012 Current Issues As regards the distribution of risk within the financial system, there seem to be arguments supporting the view that allocating risk in universal banks is likely to be beneficial to financial stability: first, as shown above, universal banks have a higher risk absorption capacity due to their diversified and hence self-stabilising revenue flows. Second, universal banks represent a closely supervised (by authorities) and closely monitored (by investors) and thus more transparent sector than other parts of the financial system. Third, universal banks should enjoy better risk management capabilities due to their first-hand insight into several or even all segments of the financial market: to the extent that universal banks have an integrated risk management, insights gained from market risk management can be used to control risk in the banking book (and vice versa), thereby lowering overall risk costs. For instance, if a universal bank, through its market making and brokerage activities, becomes aware of rising risk aversion amongst investors, this could signal a general economic downturn requiring a tightening of lending standards in commercial banking activities. b) Higher bank profitability, supporting a bank’s resilience Universal banks can achieve revenue and cost synergies, partly through economies of scale and scope (including the exchange and adoption of best practices), which specialist banks have to forego. Revenue synergies relate e.g. to cross-selling and cross-referencing between different business units: more sophisticated products such as derivatives for hedging risks that have been developed for large corporate clients may also be offered to small and medium- sized enterprises (SMEs) at no additional cost, whereas a specialist (retail & commercial) bank would have to duplicate the corporate and investment bank’s effort. Equally, business units may be able to easily point customers with growing needs for diverse financial advice/products to experts in other business areas inside the bank, thereby increasing an institution’s revenue potential. A successful entrepreneur, e.g., whose firm has already taken loans from a bank and used its cash management services may at some point start looking for ways to effectively manage and further expand his or her personal wealth – in which case the universal bank’s private banking organisation should be best positioned to gain a new client. 19 Cost synergies mainly stem from additional economies of scale which universal banks can realise compared with specialist banks: 20 IT costs, some product development costs, risk management (e.g. issuing internal ratings) and treasury expenditures and in part even costs for regulatory compliance and for market and economic research are generally fixed costs which, within a universal bank, can be spread across a larger revenue base and need not be incurred twice, so their relative (though probably not absolute) burden decreases. Taken together, these synergies will ceteris paribus strengthen a universal bank’s profitability, and hence its ability to withstand adverse business and broader economic conditions which in turn should add to financial stability and lower potential risks for taxpayers. Even the UK’s Independent Commission on Banking which (among other things) called for a ring-fencing of banks’ domestic retail business acknowledged that its recommendations would cost the banking sector about GBP 4-7 bn p.a. and the economy as a whole an additional GBP 1-3 bn p.a. 21 On the other hand, whereas a number of research papers have looked at specific aspects of universal banks’ business model such as benefits 19 For recent research results on economies of scale, see e.g. Wheelock and Wilson (2009). 20 Of course, it is also possible for specialist banks to raise economies of scale simply by becoming larger. Here instead we want to highlight the potential for further savings as certain infrastructure expenses (mostly) – that otherwise both a pure retail/commercial and a corporate/investment bank would have to swallow – can be reduced within one integrated organisation. Several subsidiaries operating independently under a bank holding company, with few central functions shared at group level, as envisaged e.g. by Britain’s ICB obviously would not be able to reap those benefits. 21 See also ICB (2011). Universal banks: Optimal for clients and financial stability 14 | November 20, 2012 Current Issues of diversification and economies of scale and scope (see references cited above), to the best of our knowledge, in-depth and comprehensive studies analysing universal banks’ sustainable profitability in a long-term and cross- country comparison are still by and large lacking. 22 c) Lower counterparty risk, greater transparency Growing sophistication of financial markets has led to the emergence of many specialists for certain activities along the value chain. This involves two main dimensions – i) business operations themselves and ii) infrastructure and support functions. The first dimension plays a prominent role especially in the US where e.g. traditional on-balance sheet lending by commercial banks nowadays only accounts for a share of less than 30% of the total credit market. 23 With regard to infrastructure and support, outsourcing and off-shoring of operations from IT to the back office of research are common across the industry. But whereas this has not shown to produce significant additional risks for the financial system as a whole (so far at least), the splitting up of core banking activities into small slices amongst a multitude of different providers definitely increases inter-dependencies and the number of potential breaking points in the value chain. Think, for example, of the complex securitisation deals prior to the crisis, which easily involved up to ten different independent parties – from the original borrower, the mortgage broker, the bank, the loan servicer and the MBS/CDO structurer to the investment bank trader, the structured credit fund and in the end the retail investor – each posing a default risk to their counterparts (in addition to accumulating large transaction costs). Counterparty risk has been highlighted dramatically during the financial crisis, which is leading to the implementation of substantially higher regulatory capital charges. Yet in a straightforward way, much of that risk can be eliminated directly, while also creating the potential for simpler and more transparent structures every supervisor is yearning for: larger legal entities, namely universal banks, can combine several steps in the value chain under one umbrella and thereby reduce the potential for fractions which might endanger crucial functions of the entire financial system. Put differently, a universal bank essentially replaces a multitude of external market transactions and the resulting inter-dependencies with an internal capital market under a uniform risk management. This reduces interlinkages and complexity and, thereby, overall risk in financial markets. d) Diversity in banks’ business models, supporting the sector’s resilience Of course, there are several other business models for banks besides that of a universal bank. Savings banks and cooperative banks often operate as pure retail banks, serving only private customers and small enterprises and referring larger (corporate & institutional) clients to “umbrella organisations”. Some financial institutions solely conduct asset management activities, a few others offer corporate finance and advisory services only. In addition, there are significant cross-country differences, with distinct national features often due to reasons that go far back in history. 24 Overall, this plurality has served financial 22 Busch and Kick (2009) show for Germany that income diversification increases banks’ risk- adjusted return on equity and return on assets. 23 See also Deutsche Bank (2012a). 24 Not least because of this differing evolutionary background, based again on substantial differ- ences in economic structures between countries, a “one-size-fits-all” approach to banking structures may not be justified nor appropriate. Other principles, however, are (quite rightly) commonly accepted such as how to measure risk best and what increase in capital requirements may be needed – the Basel III process. Revealingly, the US, i.e. the country which most pro- minently followed a split banking path for several decades, is not returning to this model. Some- what puzzlingly then, the continental European debate on the contrary has recently focused a lot on how to fundamentally disrupt a system that has served European prosperity well for a very long time. In the end, this could have the strange result of US universal or investment banks taking over much of the investment banking business from their European competitors – creating additional risks as evidence from the recent crises shows that foreign financial institutions’ Universal banks: Optimal for clients and financial stability 15 | November 20, 2012 Current Issues markets well. As the recent crisis has clearly shown, there is no single business model which is immune to the build-up of substantial risks. But vulnerabilities seldom turn into real losses all at the same time. Hence, while some business models may function better in a given environment, others may be in crisis. A regulatory push towards making all banks similar in terms of their business model – e.g. by enforcing the separation of commercial and investment banking (or at least the trading part of it) or excessively tightening risk weights for certain activities – may lead to a more uniform banking sector where individual institutions behave very much the same, and indeed suffer from this herding behaviour later on once a bubble bursts. In banking as in nature, (organisational) biodiversity has its clear merits. 25 In addition, the established universal banking system allows for considerable competition in most areas of investment banking which could change if these would have to be spun off: rising funding costs would probably be followed by an exit of smaller, “sub-scale” service providers, leaving behind only a handful of very large stand-alone investment banks. Concentration in this market would increase with all the well-known negative effects on competition, pricing – and the likelihood for an orderly exit of a troubled provider without plunging the entire market into turmoil. Corporate and institutional clients’ dependence on the few remaining investment banks/market makers would rise, as would the risk for taxpayers. 26 Finally, this issue might well gain particular importance over the next few years as capital markets (will have to) become more relevant providers of funding for the corporate sector whereas the traditional banking sector will shrink (or grow much less strongly than it used to) because it needs to adjust to much higher capital levels (Basel III) and other stricter regulatory provisions. e) Better detection of unsustainable (systemic) risk accumulations Risks to financial stability can, among other things, stem from the failure of large, systemically important institutions. For that reason, micro-prudential supervision looks at and subjects individual banks to rules which aim at securing viability even under difficult circumstances. Yet supervising individual institutions is not enough, as the recent crisis and the run-up to it have clearly shown. System-wide developments need to be monitored as well, with macro-prudential supervisors taking appropriate action if aggregate risk-taking across the financial sector exceeds acceptable levels. 27 How effective the respective bodies that have been newly set up – the ESRB in Europe (and additional peers in national Member States), the FSOC in the US – will be, however, remains to be seen. There are grounds for caution, not least i) because of the fact that the ability to enforce change once undesired developments have been recognised is fairly confined and ii) because of the politically difficult task to stop the party while it is still in full swing. Hitting the brake on new lending, e.g., would encounter resistance not only from households and companies that want to continue borrowing, but also from politicians with a view to winning the next election, and from the financial industry which – in the short term at least – may have to forego additional profits. Yet spotting the build-up of risk and ultimately unsustainable bubbles across the entire array of financial markets can also happen within large financial commitment to their clients tends to be more volatile than that of domestic ones, especially in times of distress (see e.g. Schildbach (2011) and Fitch (2011)). 25 See also CEPS (2009) and CEPS (2010). 26 If regulators, however, were to establish thresholds (as proposed e.g. by the HLEG) below which trading activities could still be conducted in an integrated model, other unwelcome effects could be the result: due to potential funding advantages, the deposit-taking entities engaging in investment banking activities would have an incentive to expand their operations up until the threshold, increasing the risk embedded in these presumably “safe” institutions. 27 See also Weistroffer (2012). Universal banks: Optimal for clients and financial stability 16 | November 20, 2012 Current Issues institutions, and primarily universal banks which have deep insights into traditional commercial banking (especially lending) developments as well as an eye on capital markets with their vast variety of innovative financial products. Thanks to their integrated risk management and an easier flow of information between diverse business lines, these universal banks may sometimes be able to detect emerging systemic risk before the official macro-prudential supervisor does so, and both actively reduce their exposures and provide this information to the market – including clients, supervisors and also competitors. The potential for overall better risk-taking decisions is even greater if there are internationally active banks which by definition accumulate knowledge not only about risk levels in a specific region, but also about cross-border linkages and indeed global exposures. The financial institutions that are best positioned to discover (though not inevitably successful in doing so) evolving systemic risk early on – thus possibly contributing to financial stability – are therefore large, inter- nationally operating universal banks. f) Efficient handling of loan-deposit mismatches Finally, financial stability is also supported because universal banks are better able to deal with loan-deposit mismatches and do not have the problems which specialist banks, and by extension a split banking system as a whole, have in dealing with such mismatches. Apart from depriving potential (“outside”) borrowers of credit, a ring-fencing of deposits would also leave banks with a deposit surplus (see chart 10) with essentially only two main options: either lending more to customers inside the fence or discouraging the inflow of deposits for lack of profitable investment opportunities (a third possibility would be the purchase of liquid, “low-risk” assets – which currently provide near-zero returns, though). Both ways, a prohibition to use deposits for virtually any other purpose than household and corporate lending would thus eliminate a major advantage of the universal banking system and could cause considerable damage: customers have a strong interest in being able to hold a safe and liquid asset in the form of bank deposits. Equally, the health of the overall economy requires a financial system which neither embarks on a sudden relaxation of credit standards followed by a (usually unsustainable) lending spree, nor actually pushes for a reduction in what is arguably its most stable funding source – viz. client deposits. Of course, a similar argument can also be made for banks with a deposit deficit (the more common type of bank in Europe). Those would have an incentive to reduce lending or – the more risky option from a financial stability point of view – to expand their wholesale funding base. Interbank loans, e.g., would be permitted to finance SME and mortgage loans under the HLEG proposals, even though the financial crisis has forcefully demonstrated the dangers of a business model which depends on short-term capital markets funding (the British retail bank Northern Rock being one of the worst examples). 4. Probability of government support for failing banks So far, our analysis has focused on the multiple advantages of universal banks over specialist banks both from an operating perspective in a normal business environment and in times of financial stress. We have argued that there are abundant reasons why the presence of (integrated) universal banks may overall be beneficial for their clients as well as for financial stability. But what happens in case a bank, despite all precautionary measures, indeed suffers a deep crisis threatening its very existence? Would it not be easier for a government to let a stand-alone investment bank fail, rather than a universal bank which also has a lot of retail and business clients? Therefore, would a banking system that is split into commercial banks on the one hand and investment banks on the other not pose a lower risk to taxpayers as public support would be ruled out explicitly and 0 50 100 150 200 250 HSBC Deutsche Bank** Crédit Agricole UBS** Standard Chartered Credit Suisse** KBC Deposit surplus* of selected European banks 10 Sources: Company reports, DB Research EUR bn; as of June 30, 2012 * Customer deposits - customer loans ** For UBS & CS: total loans; for DB: total deposits - total loans Universal banks: Optimal for clients and financial stability 17 | November 20, 2012 Current Issues ex ante for investment banks (or the separately organised investment bank operations under a universal banking holding structure, with a ring-fence around those parts “serving the real economy”)? At first sight, this seems to be a convincing argumentation. Taxpayers would only have to stump up funds – if at all – for those parts of the financial system that, allegedly, really support the economy, small- and medium-sized enterprises as well as households. Those financial institutions that live in a “parallel world” and only engage in “speculative activity” could fail without causing any serious harm. However, this simplistic world is a far cry from reality: 1) Modern finance does not distinguish strictly between traditional, balance sheet-based commercial banking and investment banking. Over the past few decades there has been a remarkable degree of convergence and there is virtually no commercial bank anymore which would not provide some sort of capital-markets related services to their clients. Capital markets funding, e.g., has become more important for the corporate sector in recent years – and will probably become even more relevant in future due to the limited growth prospects for traditional bank lending. Likewise, a substantial part of banks’ traditional lending today is based on corresponding access to capital markets: even retail-only banks nowadays rely on being able to hedge interest rate risk by means of swaps or to insure themselves at least partly against losses and reduce portfolio concentrations by securitising loans, buying credit default swaps – or using pledged collateral to receive funding from the markets or the central bank. Without this close connection between a bank’s balance sheet and easily accessible, open financial markets, the banking sector’s lending capacity would be curtailed massively. 2) A forced split of universal banks may not only reduce the overall availability of funds for companies (and make them more expensive) but distinguishing between “the good” and “the bad” parts of the banking system would be highly difficult – and the results contestable. The two cases where lawmakers have already decided to broadly follow this road (implementing the recommendations of the Vickers Commission in the UK and the Volcker rule in the US) highlight the dimension of the challenge 28 : US authorities mandated to develop concrete rules for institutions to comply with Volcker have received no less than 17,000 comment letters on their 300-page proposal (which includes more than 1,400 questions) from the industry, raising a huge range of issues that have yet to be dealt with. In the UK, very few activities are assigned a fixed position on one side of the ring-fence, while the majority of business areas can sit either inside or outside the fence – depending on the structure a bank prefers to choose. 29 3) Obviously, this optionality creates a lot of room for manoeuvre and the potential for far-reaching changes particularly to the costs of certain banking services that are located outside the ring-fence. In principle, given the universal banking model’s manifold advantages for all business lines, a split- up would probably reduce the efficiency and profitability of both commercial and investment banking operations. However, whereas policymakers have made it clear that activities inside the fence are considered essential for the functioning of the entire economy, those outside would be left to fend for themselves and could be wound down if necessary. On the one hand, this would surely result in significantly lower credit ratings for the “non-protected” entity. But at the same time, operations within the fence may well be perceived as having de facto an explicit government guarantee – which 28 The HLEG proposals would face similar difficulties, e.g. in defining “hedging services to non- banking clients” which are supposed to be exempted from the general requirement that derivatives transactions must take place outside the ”deposit bank”. 29 Deutsche Bank analysts estimate that RBS, e.g., could place anything between 27% and 82% of its loan book inside the ring-fence (see also Deutsche Bank (2012b)). Universal banks: Optimal for clients and financial stability 18 | November 20, 2012 Current Issues would probably overcompensate the loss of benefits from the integrated universal banking model and lead to artificially low funding costs (enabling excessive risk-taking in this unit). 4) Moreover, despite all public statements to the contrary, it is still an open question whether authorities would indeed decline support for a stumbling stand-alone investment bank, especially in a systemic financial crisis such as the one unfolding in 2008/09. Investment banking activity – particularly those businesses that rely on a strong balance sheet such as securities trading (on the behalf of clients), market making, position-taking in derivatives and underwriting capital issues – is an indispensable, integral part of a functioning, diversified financial system in a modern economy, whether this is politically en vogue or not. 30 In a systemic crisis, which by definition affects large parts of the financial system simultaneously, the decision to wind down a trading entity as envisaged by the HLEG will remain a difficult call, even if an effective restructuring regime were in place. At the same time, in such a crisis a bail-in of creditors, too, is much more difficult than under “normal” circumstances. 5) Furthermore, potential spill-over effects from a failing investment bank can be greater than for other banks, as investment banks tend to be more closely connected with other financial institutions (though new regulations aim at reducing the degree of interconnectedness). 31 It is therefore not unlikely (all the more after the devastating effect of letting Lehman Brothers fail in 2008) that public support of some sort would still be forthcoming in an existential crisis of a systemically important institution (even if that entity conducted no commercial banking business). Separating trading entities from deposit banks – and prohibiting mutual assistance – may well increase rather than reduce the dangers for domestic taxpayers. In the end, it is above all a political decision – and one depending on a very specific situation – whether to extend help to a troubled bank. That may be more likely if the institution in question is a universal bank rather than a pure investment bank, and even more likely if it is a pure commercial bank. Yet this is definitely not just a matter of the bank’s business model but also of its size and importance, the overall state of financial markets and the economy, and of political preferences at this particular moment. Hence, claims that all failing investment banks would simply be allowed to go under may well turn out to be a (self-)delusion. 6) Against this backdrop, the pressing need for establishing a sound restructuring and resolution regime becomes very clear. However, it does not mean that a financial crisis would not hurt anyone – losses would nonethe- less be incurred by investors, but hopefully by a much larger community and in much smaller individual slices than otherwise. A workable, efficient resolution regime would substantially reduce the pressure on the sovereign to possibly act as an “investor of last resort”. And well-designed resolution procedures should equally apply, with the same pre-defined mechanisms and loss-sharing effects, to commercial banks, universal banks and investment banks. They would definitely do much more for improving financial stability and lowering costs to the public than any arbitrary scheme trying to isolate “worthy” from “dangerous” financial activity. In that context, the bank restructuring laws that have already been adopted in Germany and the UK as well as the draft EU directive are sensible steps in the right direction. 30 See also ZEW (2012). 31 In fact, it could be argued that, as a result of the regulatory-induced consolidation, the inter- connectedness of any remaining investment bank with the rest of the financial system will probably rise rather than shrink. Universal banks: Optimal for clients and financial stability 19 | November 20, 2012 Current Issues 7) As in medicine, however, prevention is usually better (and cheaper) than ex- post treatment. Rather than having to wind down stumbling banks, the preferred approach should be to require them to hold larger buffers in the first place. And if potential risks associated with the existence of large banks are considered to be bigger than those of smaller institutions, an additional buffer for “systemically important banks” may be the right choice. Again, increasing banks’ loss absorption capacity and reducing the likelihood of failure from the beginning – irrespective of the type of bank – would indeed address a possible problem for financial stability, much more than a forced split of universal banks based on mistaken beliefs about their inherent risk. Compared with that, the SIFI buffer agreed under Basel III (though surely not perfect) is a targeted and reasonable measure. 5. Conclusion There are two overarching reasons why a universal banking system may be preferable to a split banking system: i) it provides greater financial stability and ii) it produces better solutions for its customers. On financial stability: to decide on the superiority of either the split or the universal banking system, policymakers should consider two simple, but critical questions: 1) Would a split banking system have prevented the recent financial crisis (or, for that matter, the still ongoing debt crisis)? 2) Would a split banking system prevent future financial crises? Based on the analysis above, the answers to both questions are quite clearly “most probably not”. In general, financial crises have occurred time and again, and most of them in fact involved traditional commercial banks that fell after having grown too quickly and too strongly during a preceding credit bubble. The root causes of all financial crises – asset bubbles that burst sooner or later – are completely unrelated to the structure of the banking system. The financial and debt crisis demonstrates this clearly: it originated in countries with stand-alone investment banks (such as the US, where the collapse of a large specialist bank – Lehman Brothers – triggered the global meltdown) as well as in countries with large universal banks (such as the UK) and in countries with commercial banks only (such as Spain). Most crucial for an individual institution’s performance has been the quality of its management, much more than the type of bank. Yet it is revealing that one of the most important solutions to the 2008/09 crisis in the US has been the transfer of weaker investment banks under the umbrella of stronger universal banks. Thus, while having universal banks alone does not guarantee financial stability, it is surely an important, helpful ingredient for a robust banking system that can efficiently serve its clients’ needs and the overall economy. On the benefits for clients: universal banks are not only more convenient and more effective in tailoring financial services to their customers’ specific needs than “narrower” specialist banks. They can also provide more credit by channelling surplus deposits directly to borrowers, at lower cost. Empirical evidence reveals that universal banks’ actions in the underwriting market are more credible signals to other investors, while they also reduce clients’ funding costs. Universal banks’ advantages may be most pronounced in their relation- ships with large and internationally oriented companies as well as with institutional investors. Finally, by focusing primarily on organisational matters, policymakers and regulators risk getting distracted from the “real” issues that would substantially contribute to creating a more sound and stable financial system which can still perform its role to serve the real economy. Most crucial in this respect are three Universal banks: Optimal for clients and financial stability 20 | November 20, 2012 Current Issues measures where progress has been made to some extent but more clearly needs to be done: i) strengthening the banking sector’s loss absorption capacity, ii) establishing workable restructuring and resolution regimes iii) supplementing micro-prudential with effective macro-prudential (i.e. system- wide) financial supervision. In line with its nature, losses occur in the financial system all the time. However, policymakers need to prevent them from potentially spilling over to the rest of the economy. Large-scale public bailouts (“negative externalities”) such as those during the 2007-09 financial crisis must be avoided, of course. The right way to do this is first of all to require banks to hold substantially larger capital and liquidity buffers. Basel III thus is an important step, with timely implementation and broad application across all types of banks and on an international scale crucial for its effectiveness. A second decisive change needs to come in the form of credible and sensible “bail-in” mechanisms for bank creditors; here, regulatory proposals are far less advanced, however (“living will” concepts alone are not sufficient). Thirdly, supervisors need to look more carefully at longer- term developments in financial markets as a whole to spot (and act prudently upon) the potentially risky build-up of asset bubbles which may seem to be beneficial to individual financial institutions but mostly turn out to be unsustain- able and cause major disruptions when they suddenly self-correct. Given that there is still a lot to do on these issues which are indeed crucial for the future of the European (and global) banking system, proceeding on the wrong path towards splitting up universal banks would be even more dangerous. Universal banks provide a whole set of advantages for their customers, financial stability and thus society at large, and they should by all means be allowed to continue doing so. Jan Schildbach (+49 69 910-31717, jan.schildbach@db.com) Universal banks: Optimal for clients and financial stability 21 | November 20, 2012 Current Issues Literature Altunbas, Yener, Simone Manganelli and David Marques-Ibanez (2011). Bank risk during the financial crisis: Do business models matter? ECB Working Paper 1394. Ang, James S., and Terry Richardson (1994). The underwriting experience of commercial bank affiliates prior to the Glass-Steagall Act: A reexamination of evidence for passage of the act. Journal of Banking and Finance, 18, pp. 351-395. Börsen-Zeitung (2012). Zweifel an Trennbankensystem. No. 206 (25 October 2012), p. 5. Busch, Ramona, and Thomas Kick (2009). Income diversification in the German banking industry. Deutsche Bundesbank Discussion Paper (Series 2: Banking and Financial Studies) 09/2009. CEPS (2009). Investigating Diversity in the Banking Sector in Europe. CEPS (2010). Investigating Diversity in the Banking Sector in Europe. Cumming, Christine M., and Beverly J. Hirtle (2001). The Challenges of Risk Management in Diversified Financial Companies. Federal Reserve Bank of New York Economic Policy Review, 7 (1), pp. 1-17. Deutsche Bank Markets Research (2012a). Is Private Sector De-Leveraging Almost Done? Bank Cheat Sheets. Deutsche Bank Markets Research (2012b). Retail banking 2012: past, present and future. Drucker, Steven, and Manju Puri (2005). On the Benefits of Concurrent Lending and Underwriting. Journal of Finance, LX (6), pp. 2763-2799. EIM (2010). Internationalisation of European SMEs. Fitch Ratings (2011). U.S. Money Funds and European Banks: French Exposure Down. Gande, Amar, Manju Puri, Anthony Saunders and Ingo Walter (1997). Bank Underwriting of Debt Securities: Modern Evidence. Review of Financial Studies, 10 (4), pp. 1175-1202. HLEG (2012). Final Report. ICB (2011). Final Report. Kroszner, Randall S., and Raghuram G. Rajan (1994). Is the Glass-Steagall Act Justified? A Study of the U.S. Experience with Universal Banking Before 1933. American Economic Review, 84 (4), pp. 810-832. Puri, Manju (1994). The long-term default performance of bank underwritten security issues. Journal of Banking and Finance, 18, pp. 397-418. Puri, Manju (1996). Commercial banks in investment banking: Conflict of interest or certification role? Journal of Financial Economics, 40, pp. 373-401. Schenone, Carola (2004). The Effect of Banking Relationships on the Firm's IPO Underpricing. Journal of Finance, LIX (6), pp. 2903-2958. Schildbach, Jan (2011). Home, sweet home? Deutsche Bank Research. EU Monitor 80. Frankfurt am Main. Templeton, William K., and Jacobus T. Severiens (1992). The Effect of Nonbank Diversification on Bank Holding Company Risk. Quarterly Journal of Business and Economics, 31 (4), pp. 3-17. Universal banks: Optimal for clients and financial stability 22 | November 20, 2012 Current Issues UNCTAD (2012). World investment report. Weistroffer, Christian (2011). Identifying systemically important financial institutions (SIFIs). Deutsche Bank Research. Current Issues. Frankfurt am Main. Weistroffer, Christian (2012). Macroprudential supervision. Deutsche Bank Research. Current Issues. Frankfurt am Main. Wheelock, David C., and Paul W. Wilson (2009). Do Large Banks have Lower Costs? New Estimates of Returns to Scale for U.S. Banks. Federal Reserve Bank of St. Louis Working Paper 2009-054E. ZEW (2012). The Role of Investment Banking for the German Economy. European banking markets With its dozen main indicators, the Interactive map European banking markets offers a broad overview on structures and developments in the banking systems of EU countries. Data are available on, e.g., market size and lending business, capitalisation levels and profitability measures. Our publications can be accessed, free of charge, on our website www.dbresearch.com You can also register there to receive our publications regularly by E-mail. Ordering address for the print version: Deutsche Bank Research Marketing 60262 Frankfurt am Main Fax: +49 69 910-31877 E-Mail: marketing.dbr@db.com © Copyright 2012. Deutsche Bank AG, DB Research, 60262 Frankfurt am Main, Germany. All rights reserved. When quoting please cite “Deutsche Bank Research”. The above information does not constitute the provision of investment, legal or tax advice. Any views expressed reflect the current views of the author, which do not necessarily correspond to the opinions of Deutsche Bank AG or its affiliates. Opinions expressed may change without notice. Opinions expressed may differ from views set out in other documents, including research, published by Deutsche Bank. The above information is provided for informational purposes only and without any obligation, whether contractual or otherwise. No warranty or representation is made as to the correctness, completeness and accuracy of the information given or the assessments made. In Germany this information is approved and/or communicated by Deutsche Bank AG Frankfurt, authorised by Bundesanstalt für Finanzdienstleistungs- aufsicht. In the United Kingdom this information is approved and/or communicated by Deutsche Bank AG London, a member of the London Stock Exchange regulated by the Financial Services Authority for the conduct of investment business in the UK. This information is distributed in Hong Kong by Deutsche Bank AG, Hong Kong Branch, in Korea by Deutsche Securities Korea Co. and in Singapore by Deutsche Bank AG, Singapore Branch. In Japan this information is approved and/or distributed by Deutsche Securities Limited, Tokyo Branch. In Australia, retail clients should obtain a copy of a Product Disclosure Statement (PDS) relating to any financial product referred to in this report and consider the PDS before making any decision about whether to acquire the product. Printed by: HST Offsetdruck Schadt & Tetzlaff GbR, Dieburg Print: ISSN 1612 - 314X / Internet/E - mail: ISSN 1612 - 3158