China Banking

8/2/2019 China Banking

1/29

1

Chinas Changing Business Model of Banking

Prof. Dr. Horst LoechelChina Europe International Business School, Frankfurt School of Finance & Management

German Centre of Banking and Finance at CEIBS699 Hongfeng Road, Pudong, Shanghai 201206, P. R. China

[email protected], [email protected]

Helena Xiang Li(Corresponding author)Frankfurt School of Finance & Management

Sonnemannstrasse 9-11, D-60314 Frankfurt am Main, [email protected]

EUCHINABMTWORKINGPAPERSERIES No.010


2/29

2

Chinas Changing Business Model of Banking

Prof. Dr. Horst LoechelChina Europe International Business School, Frankfurt School of Finance & Management

German Centre of Banking and Finance at CEIBS699 Hongfeng Road, Pudong, Shanghai 201206, P. R. China

[email protected], [email protected]

Helena Xiang Li(Corresponding author)

Frankfurt School of Finance & ManagementSonnemannstrasse 9-11, D-60314 Frankfurt am Main, Germany

[email protected]

Working Paper, January 2010

Acknowledgements: The authors are very grateful for extensive discussions with Prof. Chang Chun and Dr. ZhuHonghui of China Europe International Business School. The authors would like to express deepest thanks also toLudwig Fella (Commerzbank AG, Shanghai Branch), Dr. Stephan Popp (Nord/LB, Shanghai Branch), ConstantinoRiviello (Citibank N.A., Regional Office Asia Pacific Banking Institute), Dr. Hans Schniewind (Helaba, Shanghai),Marcus Wassmuth (LBBW, Shanghai), Julia Wu (Deutsche Bank AG, Shanghai) and Prof. Zhao Xiaoju (ShanghaiUniversity of Finance and Economics) for providing valuable insight into the development of Chinas banking

industry. This research is part of the EU-China Research Fellowship Program, a co-operation between ChinaEurope International Business School and Frankfurt School of Finance & Management in the framework of theEU-China Business Management Training Program, which is generously sponsored by the European Union.


3/29

3

Abstract

The recent turmoil of the global financial system urges both academics and professionals to rethink thefundamental question of the appropriate banking business model. The recent suggestion of President Obama toseparate investment banking from commercial banking recalls the come back of the Glass-Steagall Act, i. e. thereturn to the separate banking business model similar to the reaction after the word economic depression in 1930s.This turn around is particularly of interest for China due to the fact that Chinese banks just moving the other way

around after a track record of more than thirty years reform policy.

Surprisingly enough, academic research about the gradual change of the banking business model in China towardsa specific type of universal banking system is still rare especially in an international context. The current paperaims to shed some light in this regard by conducting a comparative study based on data of banks from China,Europe, the UK and the USA. The analysis reveals that that the profitability of Chinese banks is mainly bolstered

by the still guaranteed interest margin set by the central bank. Moreover, the risk pricing mechanism in loanbusiness as evident in Western peer banks is not yet established well in Chinas banking sector. Therefore, thediversification in fee and commission income is most important for Chinese banks to leverage theircompetitiveness. The high concentration of the current revenue sources on lending in Chinese banks can notsustain if the interest margin narrows in a more liberalized interest and exchange rate environment. The equity

participating of Chinese banks in asset management companies shows already first results in generating synergies

in terms of scale and scope.

However, it will be shown that this transformation is so far not accompanied with an inappropriate leverage ofrisks due to a prudent regulatory environment, which limits non-interest income business of Chinese banks toservice fees and provisions from customers instead of trade for own account. Given the momentum of stricter

bank regulation on an international level as well, a system convergence between Western and Chinese bankingbusiness model could be expected in a foreseeable time.

Keywords:Banking business models, financial crisis, universal banking, China financial system

JEL classification: G01, G20, G21, G28


4/29

4

I. Introduction

In the disastrous crisis year of decades for the global banking industry, Chinese banks celebrated their stunningadvances in the global banking league. The state-owned commercial bank giants ICBC, CCB and BOC

1were

ranked as the world largest banks by assets with the most valuable brands and were the three most profitablebanks over the world in 2008

2. The market capitalization of those three banks also took over the first three places

of global ranking by the end of June 2009, regardless of the fact, that the whole banking industry was deemed as

insolvent with non-performing loan (NPL) ratio of over 20% only five years ago. China has done its home workof banking reform with gradual but persistent and caution steps: from separating policy lending from state-ownedcommercial banks (SOCBs), to the hundred billion dollar bail-out of the whole banking system by transferring

NPLs to separate asset management companies, followed by establishing modern corporate governance structureand introduction of foreign strategic participation as well as bringing banks to capital markets step by step.

3

Despite the long lasting discussion of severe problems including operational inefficiency, lack of risk managementskill, policy lending and so on, the fact can not be neglected that the vanilla business model and strict regulationguarded Chinas banking sector in most instances from the spill-over of the recent financial turmoil.

4

The belief that Western banking model was regarded as sample model for banking reform in emerging marketswas severely shaken after the financial crisis. Lessons are learned that escalating bank return by inflating risks, a

popular model which the success of some Western banks were thanks to before the crisis, is proved to be not

sustainable. A new round of debates broke out among academics and banking managers on banking businessmodel after the crisis time.5

One stream of opinion claims the radical way back to narrow banking in whichbanking business is mainly limited to deposits, lending and settlement. Another stream points out the fact that alimitation of the business scope is unrealistic in modern banking, as disintermediation and global competitionfundamentally undermines the foundation of utility banking with high reliance on interest rate margin and creditexpansion. Market demand for more individual and diversified financial products to meet long-term financialneeds in line with demographic changes and to manage risk complexity requires variety in financial products,markets and channels. The crisis has however shown that revenue enhancement through high-leveraged capitalmarket speculation deviates from the core concept of universal banking.

The implications of this turnaround is of special value for Chinese banks which saved the disruption phase ofsystem shake caused by banks extensive risk taking in trading and leveraging by creating paper gains throughfinancial innovation far beyond the needs of the real economy. The narrow business model in Chinas separated

banking system and strict foreign capital control prevented the infection of the global financial turmoil. Based onthe experiences of the transformation of a separated to a universal banking model

6in Western banking, Lv Suiqi,

Deputy Chair of the department of finance at the Peking University, stressed however that Chinese banks could nolonger rely on the old business model based on interest rate spread which is determined by Chinas central bank.Instead, it is essential to explore other revenue sources such as asset management products to increase the corecompetitiveness of commercial banks.

7However, Chinese banks face the dilemma to adapt their business models

on a more and more liberalizing environment in terms of currency, interest rates and capital markets, without aconvincing paradigm of international universal banking standards. Due to the financial crisis, further deregulationof Chinas banking sector towards an integrated banking business model was even slowed down upon to theresults of reconsideration of the benefits and risks of the Western style of banking. Liu Mingkang, for instance,Chairman of CBRC

8, stated at the Luijiazui Forum in May of 2009 in Shanghai: We dont discuss universal

banking.9

1 ICBC: Industrial and Commercial Bank of China, CCB: China Construction Bank, BOC: Bank of China2 See Financial Times Asia, April 29th, 2009, p. 3.3 See for instance Neftci and Xu (2007), Loechel and Zhao (2006), Wu (2005) and Garca-Herrero, Gavil and Santabrbara (2006)for an overview of the reform process.4 Backed with the strong performance of Chinese banks in the crisis, Liu, Mingkang, Chairman of CBRC, expressed his fullconvince in Chinas banking regulation concept in his newly published article Basic rules helped China sidestep financial crisis inFinancial Times Asia. See Financial Times Asia, June 29th, 2009, p. 7.5 For competing opinions on banking business model after the crisis, see for instance Dobbs and Kuehner-Hebert (2008) and Rizzi(2009).6 In this paper, we define universal banking as the combination of commercial banking, investment banking, asset managementand other related financial services in contrast to the separated banking business model, which focuses banks on commercial bankingactivities alone. The universal banking model can be extended to an integrated financial services provider model by adding

insurance to the service scope.7 See 21st Century Business Herald, May 14th, 2009, p. 10.8 CBRC: China Banking Regulatory Commission, the regulatory and supervisory body for banking in China9 See The Wall Street Journal Asia, May 18th, 2009, p. M2.


5/29

5

Against this background, the current paper reflects the revenue diversification process in Chinese banks incomparison with the business scope in European and Anglo-Saxon peer banks. To our best knowledge, this paperhas the unique value to first address the business model transformation in Chinas banking sector in aninternational context.

10The main hypothesis is that Chinas banking business model is gradually moving from a

segregated towards a universal banking system, forced by interest rate liberalization, capital market developmentand further opening of domestic financial markets.

In detail we prove that the profitability of Chinese banks is so far mainly driven by an interest margin above thelevel in Western peer banks. As the interest margin is artificially held high in China through an administratedinterest rate regime, this business model can not sustain. Moreover, we discover that the risk pricing mechanism isstill not well established in Chinese banks lending practice. The guaranteed interest margin and stable loangrowth provide little incentive to diversify revenue sources through non-interest income products. Compared toWestern peers with high investigation in proprietary trading, we notice that fees and commissions instead oftrading income dominate the non-interest income in Chinese banks. We also identify that synergies in Chinas firsttrial of universal banks in the form of banks majority share holding in asset management companies are realizedthrough economies of scale since bank related funds are larger in size and provide favorable management andcustody fee condition.

As the internationalization of the Chinese currency quickens the pace as demonstrated with the ambitious appeal

of Chinese Central Bank governor Zhou Xiaochuan to rebuild the global currency system,

10

the liberalization ofthe domestic interest rate system of China to a more market orientation is simultaneously foreseeable. At the sametime, promotion of direct financing through capital markets to reduce the current high risk concentration infinancing of over 80% through banking will fuel the capital market development in China and thereforedisintermediation. Chinese banks face the unique challenge and opportunity of business model transformation tomore engagement in non-interest income business with balanced fees and commissions, high portion ofinvestment in high-graded government, financial and corporate bonds, limited trading for risk hedging andrestricted exposure in equity and derivative products. There is no reason to believe that this controlled as well ashighly regulated extension of the commercial banking business model of Chinese banks will be acceleratedtowards a Western style high-leveraged trading and asset business model.

The paper is organized as follows: After a short review of current discussions on banking business scope insection II, the evolution of Chinas banking sector from the separated to the universal banking system are

introduced in section III. Section IV analyzes the return and risk impact of interest income and non-interestincome in Chinese banks in comparison with their peers from Europe, the UK and the USA. Section V evaluatesthe first trial of universal banking model in China banks majority holding in asset management companies andidentifies the synergy sources in Chinese universal banks. A conclusion in section VI summarizes the results and

provides outlook for future development of banking business model in China.

II. Business Scope in Banking A Post-Crisis View

The exacerbation of the crisis was triggered with the bankruptcy of Lehman Brothers on September 14th

, 2008.11

The domino effect through interlinks among financial institutions and the large-scale engagement in speculativesubprime real state sector allowed the fast spread-out of the pandemic to almost all universal banks, which ended

up with global dry-up of credit and market liquidity and a recession in global economy. Voices became louder tobring banks back to the basic roots: deposit and loans, the narrow function of a utility bank model, as riskyfinancial innovation in investment banking was deemed as cause of the crisis. President Obama proposed a plan tolimit proprietary trading of commercial banks

12. At the same time, the crisis showed evidence that stand-alone

investment banks without stable financing source from deposits are vulnerable to market disrupts and bankruptcyrisks.

13Leading stand-alone investment banks Bear Stears, Merrill Lynch were sold to universal banks JP Morgan

Chase, Bank of America respectively; Lehman Brothers went bankrupt; both Goldman Sachs and Morgan Stanleywere converted to bank holding companies under the cover of FDIC, conducting commercial deposit taking andlending services. The current crisis renewed the long-lasting academic interest engaging with the expansion of

banks business scope in non-interest income businesses and its impact on risk and return profile of diversified

10 For a comprehensive overview of Chinas financial sector in general see Zhu, Cai, and Avery (2009).10 See Zhou (2009).11 For a judgment of the crisis in historical comparison see Reinhart and Rogoff (2009).12 See Obama beschneidet Freiheit der Banken, Handelsblatt, Jan. 22nd, 2010.13 For discussions about banking business model after the crisis, see for example Schildbach (2009).


6/29

6

banks. The discussion about the justification of the introduction of Glass-Steagall Act in 1933, the legalestablishment of the segregated banking system in the US as a consequence of the financial crisis at that time, hasnot reached conclusive results

14.

In regard to the permissible scope in banking, we first consider functions of banking in the economy. Banksbasically act as settlement agent for money transfer and intermediation in financial markets, enablingtransformation of amounts, risks and maturities of financial assets. The variety of activities include deposit taking,

corporate and retail lending, mortgages, leasing, clearance & settlement, credit cards, custody, sales agency forfunds and insurance products, financial advisory for corporate and private clients, underwriting of debt and equityinstruments, M&A advisory, securities broking and trading etc. Those business lines bear differentiated return andrisk patterns. There are basically two systems to organize the service bundles: the segregated banking system andthe universal banking system

15. The former separates investment banking activities from commercial lending,

deposits and settlement services, the latter unifies commercial and investment banking, either under one roof asthe case in European-style universal banks, or through holding structure in financial holding companies as thecase in the US and the UK. With the expansion of business scope of banking into insurance services in the senseof bancassurance

16, the concept of universal banking is further extended to integrated financial services

providers/financial conglomerates.

The emergence of cross-sector services in the financial industry urged studies on the return and risk pattern of

combination of different business lines in banking. However, academic debates have still not yet broughtconclusive results for our insight in this research area. Previous studies17

generally believe the revenueenhancement and risk diversification gain from expansion in non-interest income businesses, this view has beenhowever not yet conclusively proven on the basis of empirical results. Allen and Jagtiani (2000) find thatcombining banking with securities services and insurance activities reduces the overall risk. DeYoung and Rice(2004) futher reveal persistent outperformance in term of higher sharp ratio of diversified, non-traditional bankingcompared to traditional banking. Ebrahim and Hasan (2004) evaluate the gains from the view of the capital marketand find out that banks with heavier involvement in non-interest income business are credited with more growth

potential and consequently with higher market valuation.

Contradictorily, Stiroh (2004) points out that the volatility in trading revenue largely diminishes the risk reductionthrough diversification in non-interest income businesses and the interest and non-interest income became morecorrelated over time. Laeven and Levine (2007) investigate the diversification discount of financial conglomerates

and show that the benefit gains from economies of scale and scope can not compensate the costs from agencyproblem in financial conglomerate, which leads to a trade discount of financial conglomerates on the capitalmarket. Similar results are found by Schmid and Walter (2008). They reveal comprehensively the existence ofconglomerate discount in the financial services industry and conclude that the total balance of functionaldiversification is value-destroying. However, with more differentiated analysis, they show that the combination ofcommercial banking with insurance and the combination of commercial banking with investment banking

produces significant premium. Kwan and Ladermann (2009) review the portfolio effects of business divergence inbanking and conclude that both securities and insurance services are more profitable and more risky than banking.They emphasize however the necessity to differentiate sub-activities in securities and insurance and point out thatsecurities underwriting and insurance agency businesses for instance can reduce the overall risk of bank holdingcompanies. De Jonghe (2009) explicitly examines the impact of bank diversification on financial system stabilityand concludes that interest income is less risky than non-interest income in crisis time.

The above ambiguous empirical results induce a more differentiated view on banks non-interest income. Wedivide non-interest income in fees and commissions and trading income. Revenue sources for fees andcommissions include both traditional commercial banking services like settlement, credit lines, credit cards,foreign exchanges and investment banking services like brokerage and advisory fees. In recent years, commissionfees from sales agency services for insurance and mutual funds products have gained importance. Trading income

primarily refers to the net gains/losses from banks trading of foreign exchange, fixed-income, equity andderivative products on its own account proprietary trading. Fees and commissions income of banks for

providing brokerage, agency and advisory services remains in most times profits, whereas trading income is

14 See for instance Kroszner and Rajan (1994).15 Saunders and Walter (1994) provide good theoretical and empirical foundation for the universal banking system.16 Nurullah and Staikouras (2008) show the predominant role of banks as life insurance sales agent in European countries like Spain,Italy, France and Belgium,17 For an overview of discussions on benefits and costs in integrated financial services providers, see for instance Loechel, Brostand Li (2008).


7/29

7

highly volatile subject to capital markets mood and can turn to be huge losses.

Due to the high volatility and loss potential in non-interest income, especially in trading income, voices are louderto limit the business scope of commercial banks to loans and deposits. We emphasize however that the limitationof the business scope is unrealistic in modern banking as the operating environment has gone fundamentalchanges in last years:

Innovation in information and communication technology has changed the fundamentals in banking. Moderninformation and communication technology challenges banks role as information producer. Internationalconvergence in disclosure requirements enables higher transparency and comparability about lending firmsand the internet facilitates seamless communications between investors and lenders. Large firms, the corecorporate clients for most internationally operating banks, turn directly to capital markets by issuingcommercial papers and bonds for short-term and long-term financing. Calculations of Royal Bank ofScotland show that direct financing through bonds already accounts for 64% of external financing of largefirms, compared to (syndicated) loans.

18

Global competition has depressed interest rate spread in banking. Not only the loss of large corporate lendersdue to disintermediation, but also the pressure on interest margin due to fierce international competition as aresult of financial markets deregulation and information transparency shakes the dominance of interest

income as revenue source in banking.

Banks are facing more substituting products and service providers. Alternative investment in money markets,investment funds and insurance products has channeled deposits directly into capital markets or to insurers.Lower cost structure through large-scale data processing in internet banking and on-line brokerage firmsfurther challenges banks market position as sole financial intermediators. The settlement function of bankshas been more and more outsourced to external IT-specialists. Rating agencies and financial information

providers have been increasingly replacing banks role as credit monitor;

Financial advisory services for corporate clients have become more sophisticated. Capital market knowledgeand risk management expertise of banks are increasingly demanded by internationally operating largecorporations with decreased needs for bank loans.

Demographic changes generate demand for individualized products for long-term private wealthaccumulation. Asset management, insurance and pension have gained weight in financial markets.

As a consequence, global operating banks have adjusted their business model, expanding the business scope morein investment banking, asset management or even insurance and non-interest income has gained weight as revenuesource. Backed by an upward trend of business circle in overall economy and capital markets, the diversificationin Western banking especially in proprietary trading granted years of double-digit capital return and induced theradical transformation of banks role from lenders to traders. The excessive engagement in profit making fromspeculative asset bubbles beyond the needs in real economy uncovered the risk side of the business model and is

blamed as one of the main causes of the financial crisis. Some banks went too far, moving from capital markettrade facilitator to risk taker

19, which brought the whole financial system in fragility. We believe that lessons

are learned that there will be a return from the capital-market focused universal banking model to the

customer-centric universal model in banking with a balance of commercial and investment banking activitiesdetermined by the needs of the real economy instead by greed for short-term profits.

Regulatory reforms are of necessity in the post-crisis time. The radical cut-off of proprietary trading fromcommercial banks is however subject to closer look, since proprietary trading is actively engaged in banksasset-liability and risk management. The exclusion of commercial banks as market makers would haveundesirable consequence of the overall liquidity on capital markets. Barth, Caprio and Levine (2008) even proveempirically that activities restriction in banking increases the probability of banking crisis. If activities restrictionis unrealistic or expensive in the time of disintermediation, global convergence of regulation on integratedfinancial services providers has become an urgent task. Research in this field is already under way:Blundell-Wignall, Atkinson and Lee (2008) propose to bundle diversified services in banking only undernon-operating financial holding to prevent risk transfer. Change of paradigm to integrated risk management

urges, with emphasize on interactions of various risks from interest and non-interest income within a financial

18 See Hennes and Metzger (2010).19 See BCG Creating Value in Banking 2009.


8/29

8

conglomerate group.20

Compulsory disclosure of non-interest income business lines has become necessary toevaluate the aggregated risk profile of a universal bank.

The above discussions are of special value for Chinas banking reform, since China is on trace of thetransformation from the segregated to the universal banking system. Chinese scholars are increasingly analyzinglessons of Western universal banks to develop a universal banking model suitable for the Chinese reality. Chen(2009), Professor for economics at Renmin University of China, recently points out that the restriction of

permissible services in banking leads to banks concentration on lending business, which in turn causespro-cyclical credit expansion. Mao (2008) addresses that the high risks from proprietary trading diminished thediversification benefit of universal banking in the crisis and emphasizes that the non-interest income business inuniversal banking should concentrate in fees and commissions services.

The transformation of Chinas banking sector towards the universal banking model has become an urgent task inrecent years, as:

Foreign competition and integration of China in global financial markets put pressure on traditional bankingbusiness model. After Chinas accession to WTO, foreign banks were granted to full license of local currencyalso to retail clients after December 2006

21. Over the world, functional separation of commercial and

investment banking were lifted, in 1986 in the UK, 1987 in Canada, 1992 in Japan and in 1999 in the USA

for example. Foreign competitors in China are not only involved in deposits and lending with strongposition in foreign currency lending and trade financing, but also are active in high value-added investmentbanking business like derivatives trading, M&A advisory and asset management. With the entry of foreignuniversal banks, Chinese banks bear risk of the loss of business opportunities in cross-sector services toforeign competitors, if Chinese banks miss the chance to build up capacities for cross-sector services due tostrict activities permission in a separated banking system. Currency and interest rate liberalization in nextyears is anticipated in line with the State Councils plan to build up Shanghai to an international financecentre till 2020

22. The business model of concentration of revenue source on lending of the current over 80%

on average in Chinas banking sector can not sustain in the more liberalized operating environment withlower margin. Bolstering revenue through loan expansion bears the risk of excess market liquidity and asset

bubble.23

Chinese banks face the challenge to change the quantity-based competition to quality-basedcompetition with diversified products, markets and channels

24.

Leveraging banks transformation provides the key solution for solving the co-existence of high excessliquidity in the banking sector and limited financing access of private SME companies. As in figure 1depicted, over 80% of financial assets in China are concentrated in the banking sector, which bears highconcentration risk for the entire financial system. Extending banks scope to securities services, assetmanagement and insurance can broaden financing access for corporates and build up social securities for

private households.

20 See for instance the recent research results of BIS on interaction of market and credit risk, BIS (2009).21 Till the end of 2008, seven local incorporated foreign banks were granted license for RMB retail business; 51 banks were grantedlicense for derivatives trading, see CBRC (2008), p. 44.22 For details of the plan to build Shanghai as an international finance centre, see Loechel/Boeing (2010).23 Foos, Norden and Weber (2009) show that abnormal loan growth leads to higher credit losses, lower bank profitability and higherbank insolvency risk.24 Jamie Dimon, CEO of JPMorgan Chase, observes the need for diversification in Chinese banks and puts the evolution ofJPMorgan as an example for Chinese banks expansion into fee business, see Dimon (2009).


9/29

9

Figure 1. Distribution of Savings in China

Despite the relevance and urgency of the topic on banks non-interest income business, relevant research in thisfield, especially for the Chinese case, is rare.

25The current paper aims to fill this gap by evaluating the business

model of Chinese banks in an international context and to contribute to the general debate on banking businessmodel after the financial crisis.

III. Transformation of Segregated to Universal Banking System in China

As if by a miracle, Chinese banks escaped almost unaffected from the crisis year, in greater extent as a result ofthe lucky fact that Chinese banks still follow the narrow banking model with strict restriction in engagement ofinvestment banking and the strict foreign currency control policy in China prevents the risk transfer frominternational capital markets. This segregated banking system was established in 1993 with the issuance ofResolution on Financial System Reform of the State Council and legally established with Article 43 of the Lawof Peoples Republic China on Commercial Banks issued in 1995:

Commercial banks are not allowed to make trust investment, trade in shares or make investment in fixed

assets of non-self use within the People's Republic of China. Commercial banks are not allowed to make

investment in non-banking institutions and enterprises within the People's Republic of China.

Correspondently, the separation of securities from banking is regulated through Article 6 of the Securities Law ofthe Peoples Republic of China enforced in 1998:

Securities business shall be engaged in and administered as a business separate from the banking business,

trust business and insurance business. Securities companies shall be established separately from banks, trust

companies and insurance companies.

The integration of banking in insurance institutions is restricted according to Article 104 of theInsurance Law ofthe Peoples Republic of China issued in 1995: :

The use of fund of the insurance company is restricted only to bank deposit, trading of government bondsand financial bonds and other forms of fund utilization stipulated by the State Council. The fund of the

insurance company may not be used to set up securities operation organizations or to invest in enterprises.

According to the above articles, combination of banking with securities services, real estate and insurance, whichare usually included under one roof in European style integrated financial services providers, are prohibited bylaw in China. Even in international comparison with developing countries, China still belongs to the 10% minorityof countries with the most strict activity restriction in the banking sector

26. Furthermore, the Provisional Rules

Governing Money Brokerage Firmsissued in 2005 grants only non-bank institutions permission to set up moneybrokerage firms. Chinas choice for the segregated banking model is all other than accidental. Chinese Academy

25 Sauders and Walter (1996) are among the first to address the development of universal banking in the Asia Pacific context.26 See IIB Global Survey (2009) and Barth, Caprio and Levine (2001).


10/29

10

of Social Science (2001) argues that this decision made in the 1980s was in accordance with the acknowledgementof the draw-back of conflicts of interests in universal banks and the lacked skill in managing risks in diversifyinginstitutions in the immature stage of Chinas financial institutions, after thorough comparative studies on theAmerica style separated banking system and the European style universal banking system.

With the pace of global deregulation of activities restriction in banking and market demand for diversified services,financial conglomerates with holding subsidiaries in banking, securities and insurance emerged in China

27and

some banks like BOC actually by-passed the restriction by setting up subsidiary for cross-sector services insecurities and insurance abroad like in Hong Kong. In 2004, an opening clause...with the exception stipulated bythe State was introduced to Article 43 of the commercial bank law and article 6 of the securities law regarding

business restriction, which marked the first step of the loosening of activities restriction. Banks should leveragetheir dominant position in the financial system to foster diversified financing sources and channel funds to capitalmarkets through providing products and services in securities and asset management.

With the overall market enthusiasm for banks diversification in other revenue sources than lending and topromote capital market development, PBC

28, CBRC and CSRC jointly issued in February 2005 Administrative

Rules for Pilot Incorporation of Fund Management Companies by Commercial Banks, which established the legalpermission for SOCB and JSCB as primary shareholder in equity-holding of fund management companies. Banksare permitted to sell fund products of affiliated fund management companies on the commission basis, however

only under third-party condition. The aim was to channel the high liquidity in bank deposits to capital marketinvestment, to increase competition between fund management companies, as well as to diversify banks revenueresources. This step was regarded as a significant step to transfer Chinas banking system from the segregated tothe universal banking model with a holding structure. Pilot permissions were granted to three of the big fiveSOCBs: ICBC, CCB and BOCom

29. Already before this deregulation, BOC circumvented the sector separation by

setting up its majority holding fund management company through its holding subsidiaries in Hong Kong in 2004.After the liberalization, the shares were transferred to BOC in 2008. As the last SOCB, ABC got the license in2008 and founded its own fund management company. All five SOCB-held fund management companies wereestablished as joint ventures with foreign financial institutions. The permission to cross-sector bank holding infund management companies granted to five SOCBs was expanded to two JSCBs - SPDB

30and China Min Sheng

Bank, setting up joint venture fund management companies in 2007 and 2008 respectively. As depicted in figure 2and figure 3, the liberalization of banks participating in fund management companies triggered an expansion inmarket share of bank related asset managers in the form of bank subsidiary, bank affiliation and financial

conglomerate affiliation31. The fifteen fund management companies with bank-linkage occupied 38.70% of totalmarket share in total assets under management among total sixty fund management companies only three yearsafter the liberalization in 2005.

Source: Wind

27 For the development of financial conglomerates in China, see for instance Lin (2003).28 The Peoples Bank of China (PBC) is the Central Bank in China. The regulation and supervision of financial institutions is carriedout through the central bank plus three commissions (Yi Hang San Hui) system in China. Three commissions include the sectorsupervisory bodies CBRC, CSRC and CIRC.29 BoCom: Bank of Communications30 SPDB: Shanghai Pudong Development Bank31 Bank subsidiary is defined in this paper as a fund management or securities company majority-held by a bank, bank affiliation is afund management or securities company minority held by a bank and financial conglomerate affiliation is a fund management orsecurities company with a financial conglomerate as major shareholder which also has equity participation in banking within thegroup.

Figure 2. No. of Fund Management Companies

7

26

45

05

10

15

20

25

30

35

40

45

50

Bank subsidiary Bank affil iation Financial

conglomerate

affilation

Stand alone

Figure 3. Market Share of Fund Management

Companies per Total Assets 2008

8.16%6.41%

24.14%61.30%

Bank subsidiary Bank affiliation

Financial conglomerate affilation Stand alone


11/29

11

Moreover, the equity link of banks with fund management companies promotes the agent and custody services inbanking. Although banks with fund management affiliates are only allowed to conduct business contracts withtheir affiliates under the same condition as with third party, the business cooperation with affiliated parties is moreintensified and the scope of cooperation is broader and deeper, especially that bank as sales agent may favorablychoose affiliated funds, securities and insurance companies as partner. The cooperation with affiliated companiesis however restricted through Article 8 of the Law of Peoples Republic of China on Securities Investment Fundissued enforced in June 2004, A fund trustee and a fund manager should not be the same party, and should not

make capital contribution to or hold the shares of each other. Since eleven of the total fourteen banks with acustody license are listed on the exchange, the above Article is in current discussion to be removed in order toenable fund management companies to optimize their investment portfolios including shares of their custody

banks. The manifold possibility of cooperation of banks with their affiliates can be illustrated with the followingexample: Industrial Bank Co. Ltd., Industrial Securities Co. Ltd. and Industrial Fund Management Co. Ltd., forinstance, signed a MoU for strategic alliance in 2007 to share client base, distribution channel, productinformation and training capacities. The lifting of the separation of banking from fund management in 2005changed dramatically the landscape of the distribution model in fund products. As shown in figure 4, bank gainedsignificant ground in fund distribution and took over almost 80% of the market share within two years after the

policy liberalization.

Figure 4. Fund Sales Channel

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

2004 2005 2006 2007

Fund Bank Security

Source: Securities Association of China

This policy change opened new revenue resources with high growth potential alongside with the booming capitalmarket. Total bank custody fee skyrocketed in 2007 and almost reached the mark of five billion CNY (see figure5). With the vast branch network, broad customer base and capital strength, the big five SOCBs dominate the fundcustody market and took over 90% the market share, as illustrated in figure 6.

Figure 5. Bank Custody Fee

0

10

20

30

40

50

60

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

in100mC

NY

0

50

100

150

200

250

300

350

400

450

500

Rebased1998=100

Fee SH Com posit Index

Source: Almanac of China's Securities Investment Funds, wind


12/29

12

Figure 6. Market Share of Fund Custody 2007

ICBC, 31%

CCB, 22%BOC, 16%

ABC, 14%

BoCom, 8%

Other banks, 9%

Source: Securities Association of China

Compared to the rapid rise of banks in the fund management market, banks equity holding in securities firms islimited. There is only one securities subsidiary of a bank in China BOC International (China) Ltd. founded as asubsidiary in 2002 by Bank of China International Holdings, a Hong Kong-based wholly-owned subsidiary ofBank of China Ltd. Securities affiliates of financial conglomerates have long track record in securities services

and hold 13.86% of total market share. However, 96 stand-alone securities companies still dominate the market,with 77.47% of total assets in the securities market, as shown in figure 7 and figure 8. From the policy side, it isnot expected to see a rapid liberalization and the consequent expansion of banks market share in securitiesservices in the near term, as what happened in the fund management market with a strong supporting policyfavoring banks engagement. The cautious approach of the Chinese government in dealing with securities servicesdemonstrates its great concern about the transfer of volatile capital market risks to the banking system.

Source: Securities Association of China, China Securities Yearbook

The financial crisis seems not to distract China from its cautious plan of gradual loosening of activities restrictionin banking. In November 2009, CBRC issued Provisional Rules on Commercial Banks Share Holding in

Insurance Companies, granting banks equity participation in insurance companies.

Over years, especially the big five SOCBs and JSCBs have build up financial holding emporia with business linescovering banking, securities, asset management, leasing, insurance etc. with subsidiaries at home and abroad. Asshown in Figure 9 as an example, the financial group BOC Ltd. covered financial services in banking, insurance,investment and leasing. Similar to the situation in many of its domestic peers, Hong Kong SAR

32, with its liberate

regulatory boundary in business diversification for financial institutions, is a preferred location for Chinese banksto explore the expansion in cross-sector services. The investment-banking arm of BOC Bank of ChinaInternational (China) Ltd. was founded as a subsidiary of BOCs wholly owned Hong Kong subsidiary BOCInternational Holdings Ltd. to enable the best knowledge and personnel transfer in securities services from HongKong to Mainland China.

32 SAR: Special Administrative Region

Figure 7. No. of Securities Companies

3 5 1

96

1

0

20

40

60

80

100

120

Bank

subsidiary

Bank affil iation Financial

conglomerate

affilation

Insurance

subsidiary

Stand alone

Figure 8. Market Share per Total Assets 2007

1.25% 6.08%13.86%

1.35%

77.47%

Bank subsidiary Bank affiliation

Financial conglomerate affilation Insurance subsidiary

Stand alone


13/29

13

Figure 9. Group Structure of BOC

Source: Bank of China Annual Report 2008

China has chosen a gradual and cautious pace in relaxing activities restriction in banking. Compared toEuropean-style universal banking with commercial and investment banking under one roof, China follows theAnglo-Saxon approach of financial holding company. The most remarkable difference compared to universal

banks in Western style is the limited scale in proprietary trading. Trading products are in majority less riskyfixed-income products as well as interest rate and foreign exchange derivatives for risk hedging. For eachcategory of trading instruments, a separated license should be applied. The licenses are linked with strict capitaland human resource requirement. For instance, the Rules on Proprietary Trading in Foreign Currency ofFinancial Institutions issued in 1993 set US$20 million as the minimum capital for foreign currency trading. Themanagement and trader should have gained working experiences in trading for five and three years respectively.With the belief that banks have gained knowledge over time, the scope of trading securities is gradually expandedto stocks, commodities and derivatives. Permissions are however predominantly granted to preferential SOCBsand advanced JSCBs which can meet the strict requirements in capital strength, risk management capacity andspecialist know-how.

The overall result shows a rapid expansion of Chinese banks into non-interest income sector since the policychange in 2005. Universal banks in China benefited from the capital boom from 2005 to 2007 and gainedconsiderable market shares in fund distribution, custody and insurance agent services. With the trend of the

internalization of the Chinese currency and the subsequent liberalization of interest rates as well as thedevelopment of Chinas capital markets and disintermediation to more direct financing, the move of Chinese

banks from the narrow banking model to the universal model seems to be irreversible. The meltdown of theWestern banking model with excessive leveraged trading is seen as a confirmation for Chinas caution inliberating Chinese banks engagement in proprietary securities trading. In the following section, we compare thecurrent business model of Chinese banks with banking models of Western peer banks, try to understand the returnand risk impact of revenue diversification in non-interest income in Chinese, European and Anglo-Saxon banksand to prognosticate the further pace of this transformation.


14/29

14

IV. Chinas Banking Business Model in International Comparison

4.1. Data description

In this section, we compare the business models of Chinese banks with their European, UK and US peers, usingdata both for top banks defined as the top five banks per total assets in respective regions/countries and panel dataof bank annual financial reports over the time frame from 2003 to 2008.

33We try to understand how differently

Chinese banks operate compared to their Western peers and evaluate the sustainability of the current model. Dataare drawn for commercial banks from the database Bankscope

34.In order to capture the whole business rage in

universal banking, we choose the consolidated financial statements of the selected commercial banks.35

Figure 10 provides a comparison of the development in total assets in top banks. As shown in figure 10, the largestbanks are concentrated in the Euro area. In Europe, seven banks passed the total assets threshold of one trillionUSD (Deutsche Bank, BNP Paribas, Socit Gnrale, Banco Santander, UniCredit, ING Bank and Calyon) at theend of 2008. All big four Chinese SOCBs (ICBC, CCB, ABC and BOC) reached the level of their international

peers with over one trillion USD total assets. Barclay, RBS, HSBC from the UK and Bank of America, Citibankfrom the US play in the same league of large banks with over one trillion assets. With the increasing financialdepth in China (currently 278% of GDP, 314% for the Euro zone, 326% for the UK and 385% for the US)

36, a

further expansion of assets in Chinese banks in next years is expected. Backed with similar operating scale as

internationally operating banks, Chinas SOCBs have the potential to upgrade theirs business model to competewith their international peers.

Figure 10 . Average Total Assets - Top 5

0

500,000

1,000,000

1,500,000

2,000,000

2,500,000

2003 2004 2005 2006 2007 2008

inmillionUSD

China Europe UK US

Source: Bankscope

The overall statistics of the sample are provided in appendix B. For the whole banking sector, Chinese banks withthe average asset return of 0.68% still lie behind the level of European, UK and US banks with asset return of0.89%, 0.85% and 0.97% respectively. On the aggregated level, Chinese banks are still weak in equity strength(with average equity ratio of 4.98% compared to 9.63%, 11.85% and 11.10% in Europe, the UK and the USArespectively) and asset quality (with average NPL ratio of 5.05% compared to 3.22%, 2.20% and 0.88% in Europe,

the UK and the USA respectively). Regarding the business scope, revenue resources are highly concentrated ininterest income with non-interest income as percentage of gross revenue of average 14.25%, compared to 45.22%,36.14% and 27.99% in European, UK and US banks respectively. Chinese banks benefit from stable savingsdeposit as main financing source with 80.78% of assets from customer deposits (compared to 40.87%m 55.57%and 70.89% in European, UK and US banks respectively) and enjoy high asset growth with average growth rate of26.21% (compared to 15.33%, 14.11% and 16.26% in Europe, the UK and the USA respectively). Remarkable isthe very limited involvement of Chinese banks in trading with trading asset ratio as percentage of total assets of

33 The total sample size is summarized in appendix A. The short time frame is chosen since the development of universal banking isa relatively new phenomenon in line with the recent loosening of activities restriction in China, and data availability is limited forChinas banking sector before 2003. Over the sample years, Chinese banks underwent fundamental restructurings. Both factors couldlimit the explanatory power of the results.34 Bankscope universal format is applied to mitigate the deviation in term definitions.35 Consolidated reports are only available for 17 Chinese banks. For the rest of Chinese sample banks, unconsolidated bank reportsare applied. Since the majority banks with unconsolidated reports do not have subsidiary entities, the results will not be altered withthis limitation.36 See McKinsey Global Capital Markets: Entering a New Era (2009).


15/29

15

1.97% (compared to 8.54%, 8.70% in European and UK banks respectively). The composition of interest andnon-interest income of Chinese banks in comparison with their Western peers as well as the return and risk impactof the differences in business scope are of our special interest in the analyses in this section.

4.2. Business model comparison

Revenue pattern - interest income

In the following, we explore the pattern of revenue resources in banking. We first take a look at the composition ofrevenue sources in top five banks. Figure 11 manifests the high reliance on lending business in top Chinese banks.The average portion of non-interest income in gross revenue of 14.8% lies far below the level of peer banks inEurope with 51.8%, in the UK with 46.5% and 41.0% in the US. The figure clearly indicates that Westerncommercial banks have more balanced revenue sources between interest income and non-interest income

businesses to cope with disintermediation in their operating environment, while Chinese banks still struggle for abreak-through in creating value with high value-added financial services beyond the lending business. Based onthis observation of top banks, we assume:

Hypothesis 1: Compared to Western banks with balanced portfolio of loans and high value-added non-interest

income products, asset return in Chinese banks depends highly on lending business with net interest margin and

asset quality as profitability determinates.

Figure 11. Non-interest Income / Gross Revenue - Top 5

0

10

20

30

40

50

60

70

2003 2004 2005 2006 2007 2008

Percentage

B5 EURO UK USA

Source: Bankscope

To test hypothesis 1, we conduct in the following regression analyses for the testing fields based on panel data37

.The OLS model is used for the following regression equation (Basic model: China, Europe, UK and USA model1):

ROAAit = + 1 *[Ln (total assets)] it + 2 * [equity/total assets] it+ 3 * [net interest margin] it + 4 * [cost income ratio] it+ 5 * [growth of total assets] it + 6 * [impaired loans/gross loans] it+ 7 * [non-interest income/gross revenue] it + it

With the above equation, we try to test the influence of seven explanatory variables for i bank at time t onprofitability measured with ROAA

38. We choose asset return ROAA instead of equity return ROAE as

profitability indicator, because Chinese banks experienced higher fluctuation in equity through asset injection andIPO in last years, and some banks even had negative equity in early sample years before the capital injection. Theseven explanatory variables include the natural logarithms of total assets, equity ratio, net interest margin, costincome ratio, asset growth, impaired loan ratio and non-interest income ratio. The total regression results are

37 We conduct separated regressions for the four countries/regions instead use pooling data with country/region dummy variables,with the belief that the profitability of banks in those countries/regions follow different patterns.38 ROAA: return on average assets, ROAE: return on average equity.


16/29

16

summarized in appendix C.

The seven variables in the basic models jointly explain 43%, 24%, 51% and 51% (adjusted R) for the returnpattern in China, Europe, the UK and the USA respectively. Comparing the results of regression constants, theinsignificance of the constant factor in China reveals the high heterogeneity of asset return among diverse types ofcommercial banks in China, whereas the asset return of Western banks converges to a level of over 1% at 10%significance level (1.0107, 1.0660 and 1.8608 for European, UK and US banks respectively). In China model 2

where the sample concentrates on the biggest commercial banks in China, the regression constant with 1.5896 atthe 1% significance level shows that the profitability in strongest Chinese banks, including five SOCBs and listedJSCBs, lies on an international comparative level.

Across the research fields, banks with equity strength (with significantly positive regression coefficients of equityration of 0.0197, 0.0747, 0.0364 and 0.0256 for China, Europe, the UK and the USA respectively) and favorablecost income structure (with significantly negative regression coefficients of cost income ratio of -0.0178, -0.0123,-0.0073 and -0.0233 in China, Europe, the UK and the US respectively) have significantly higher asset return. Theincrease of 1% in net interest margin in China contributes to 0.01882% higher asset return, compared to 0.0748%in the UK and 0.01689% in the USA. Margin increase in Europe has however no significant impact on asset return.In comparison, the increase in 1% of non-interest income ratio as percentage of gross revenue leads tosignificantly higher asset return of 0.0144% in Europe, compared to 0.00795 in China and 0.0100 in the USA.

Summarizing the balance of interest income and non-interest income as revenue sources, European banks exhibitthe highest level of disintermediation. Valverde and Fernndez (2007) explain that European universal bankscharge low net interest margin for client access and create value through bundling and cross-sell high value-addednon-interest income businesses. Asset expansion in both Chinese and UK banks contributes to higher return (withregression coefficients of 0.0026 in China and 0.0241 in the UK). In China, larger banks demonstrate significantlyhigher asset return (with the regression coefficient for the natural logarithms of total assets of 0.0293), whereas thesize affect is not significant in Western peers. In Western banks, asset quality predominantly determines the rate ofasset return. The regression coefficients of impaired loan ratio are -0.0626, -0.2049 and -0.2884 in Europe, the UKand the USA respectively. In contrast, the regression shows no significance of the impact of impaired loan ratio inChina. This could be explained by the fact that there were structural changes of NPLs in Chinese banks due torestructuring programs in last years. For example, the impaired loan ratio in top five banks dropped dramaticallyfrom the average of 17.6% in 2003 to 2.6% in 2008, as shown in figure 12. In the subsequent years, banks wereforced to take tremendous efforts to improve lending practice by replacing policy lending, establishing more

market oriented lending mechanism and better risk management39.

Figure 12. NPLs / Gross Loans - Top 5

0

2

4

6

8

10

12

14

16

18

20

2003 2004 2005 2006 2007 2008

Percentage

B5 EURO UK USA

Source: Bankscope

We future examine the level and impact factors of net interest margin, another determinant in lending businessbeside asset quality. Ho and Saunders (1981) reveal that various factors have impact on net interest margin: banksrisk perception, size of banks transactions, banking market structure and market interest rate variance. Saundersand Schumacher (2000) further recover that regulatory components like restrictions on deposits, reserverequirement, capital ratio etc. also have influence on banks net interest margin. As exhibited in figure 13, Chinese

39 For an overview of the restructuring programs in Chinas banking reform, see for instance Wang (2009)


17/29

17

top five banks enjoyed an average interest margin of 2.6%, compared to the European average of 1.1% and 1.3%in the UK. The low margin in continental Europe and in the UK of large banks can be partially explained by thefierce competition for large European banks and the dominant universal banking model in which interest margin iskept low to maintain long-term client relationship and revenues from cross-selling non-interest income productsand services to those clients play an important role. Only in the US, large banks achieved a slightly higher averagemargin of 2.8% in last years. For the whole banking sector, the average net interest margin of Chinese banks with2.78% is relatively low compared to 3.38% in Europe, 2.98% in the UK and 3.53% in the USA (see attachment B).

However, the extreme low standard deviation of 0.94% in Chinas banking sector (compared to 16.92%, 4.00%and 1.30% in Europe, the UK and the USA respectively) reveals in some extent a more differentiated pricing

policy in Western banks with broader client scope of diverse risk profiles.

Figure 13. Net Interest Margin - Top 5

0.0

0.5

1.01.5

2.0

2.5

3.0

3.5

4.0

2003 2004 2005 2006 2007 2008

Pe

rcentage

B5 EURO UK USA

Source: Bankscope

In China, policy component PBC could play a significant role in explaining the magnitude and variance of banksinterest margin. Under the current interest rate regime, the central bank of China PBC sets a ceiling of deposit rateand a floor of lending rate, which for example guarantees an official net interest margin of current 3.06% for

one-year loan in local currency, as illustrated in figure 14. This guided interest rate system was establishedthrough Interim Measures of the Administration of RMB Interest Rates in 1990 and the revised versionAdministrative Provisions on RMB Interest Rates issued in 1999, which granted the central bank PBC theauthority to set the ceiling deposit rates and floor lending rates. The interest rates on the inter-bank markets andrepo rates are allowed to be negotiated among institutional market participants. A floating of the lending ratewithin the range of 0.9 and 2.0 is permitted from 2004 on, the ceiling deposit rate is however obligatory. The goalis to use guided interest rate as macro-prudential measure for monetary control as well as to prevent banks fromlosses in fierce competition through high interest rate promise and dumping lending rates

40. For foreign currency

depositing and lending, the price mechanism is more market oriented since only the interest rate for deposits underthe limit of 3 million USD is ceiled with a guided interest rate issued by PBC. However, the strict capital controlin flows of foreign currency diminishes the freedom in market pricing. The guaranteed profit margin in RMBlending keeps Chinese banks away from the level playing field with international peers which price loans solelyon market condition. The artificial higher profit partially conceals the hiding inefficiency in pricing and riskmanagement and impedes innovation in new products and services in Chinas banking sector. The higherofficial-set margin can only be held with the current system of capital control for foreign currency capital account.With the internalization of the Chinese currency and more openness of its financial market in the foreseen decade,it is expected that the interest rate regime will be reformed in the direction of more market orientation. Till then,Chinese banks, which fail to build up risk pricing capacity and balanced revenue portfolio, will risk their ability instanding the competition with international peer on a level playing field.

40 Saunders and Schumacher (2000) also argue that policy settings in interest rate is used as instruments to balance the trade-off ofbank solvency (high equity ratio, high margin for example) with lower financing costs (lower margin).


18/29

18

Figure 14. Net Interest Margin from Guided Interest Rates 1991-2008

0

2

4

6

8

10

12

14

1991

.04.21

1993

.05

.15

1993

.07

.11

1995

.01

.01

1995

.07

.01

1996

.05

.01

1996

.08

.23

1997

.10

.23

1998

.03

.25

1998

.07

.01

1998

.12.07

1999

.06

.10

2002

.02.21

2004

.10

.29

2006

.04.28

2006

.08

.19

2007

.03

.18

2007

.05

.19

2007

.07

.21

2007

.08

.22

2007

.09

.15

2007

.12.21

2008

.09

.16

2008

.10

.09

2008

.10

.30

2008

.11

.27

2008

.12.23

Date of PBC Interest Rate Adjustment

Percen

tage

Deposit Loan Margin

Source: PBC

Based on the above observation of interest rate setting, we assume:

Hypothesis 2: Due to the administrated interest rate regime, the risk pricing mechanism in lending is distorted in

Chinese banks.

We test the risk pricing mechanism in banking by exploring the correlation of various profitability determinateswith net interest margin. The results are summarized in appendix D. The striking observation is the negative signof the correlation between impaired loan ratio and net interest margin for Chinese banks compared the positivecorrelations in Western banking. While the risk pricing rule lower asset quality in terms of higher NPL ratio,higher margin functions well in Europe and in the UK (with correlations of 0.2403 and 0.5388 respectively), thenegative correlation in China with -0.1997 is statistically significant at the 1% level. This observation manifests

partially that the government administrated interest rates distort risk pricing mechanism in lending practice of

Chinese banks.

Furthermore, the close relationship between cost income ratio and net interest margin in China with -0.4608confirms that the high interest margin instead of operational efficiency explains partially the lower cost incomeratio in China. The consistent negative correlation between non-interest income ratio and net interest margin in alltesting fields (-0.6316, -0.1025, -0.1662 and -0.3229 in China, Europe, the UK and the USA respectively)documents that the decline in net interest margin forces banks to expand the business scope in non-interest income

business. The highest correlation between net interest margin and non-interest income ratio in China (-0.6316)demonstrates that the administrative guaranteed high net interest margin reduces in large extent the incentive ofrevenue diversification in Chinese banks. The high negative correlations between trading securities ratios with netinterest margin (-0.1570, -0.5239 and -0.1136 in Europe, the UK and the USA respectively) reveal that the thinmargin in lending business further drives Western banks to diversify securities trading. In respect of the size effect,

the negative sign of correlation with total assets shows that large banks in all testing fields are not able to use themarket power to push through higher margin, which dilutes the concern of the monopoly power of current largesize banks.

To summarize the results from the above analysis of interest income as revenue source, the profitability pattern inChinese banks has the highest dependence on net interest margin. Especially in European universal banks, thecorrelation of non-interest income ratio with between net interest margin is statistically not significant, whichshows the highest disintermediation rate in European universal banks. Furthermore, the risk pricing mechanism isstill not well established in Chinese banks due to the administrated interest rate regime. Western peer banksdemonstrate in comparison consistent risk pricing pattern: lower loan quality, higher net interest margin. Theguaranteed high margin diminishes partially the incentive of Chinese banks to expand the business scope innon-interest income businesses. Facing margin pressure, Western banks are diversifying in trading business. Thefindings from the comparison cause much concern about the sustainability of the current business model in theChinese banking sector: At the latest when big SOEs turn directly to capital markets for fund raising and theinterest rate is liberalized and adjusted to an international conventional level, Chinese banks which fail to build upcapacities for risk pricing in lending business and to balance interest income and non-interest income business will


19/29

19

suffer from a sharp decrease in profitability.

Revenue pattern - non-interest income

Having recognized the backlog in revenue diversification, Chinese banks took tremendous efforts to developnon-interest income businesses, backed by supporting policy and taking advantage of the capital market boom forinsurance and funds agency and custody services in the last upward cycle. In absolute term, the scale of net

income from fees and commissions lies however half to two-third of the level of comparable peers from the UK orEurope, as shown in table 1. And the potential in bancassurance products is not fully explored in China. It isremarkable that the revenue scale of securities trading in Western top banks almost reached or even surpassed thelevel in fee and commission business, whereas the scale in trading of Chinese large banks amounted to only asmall fraction of net income from fees and commissions. Measured by the fraction of trading assets in total assets,the limitation of Chinese banks trading activities is evident, with an average rate of 1.1% (20.1% in Europe,13.8% in the UK and 9.3% in the US), as illustrated in figure 15. The immature stage of revenue diversification inChinese banks results on the one side from the fact that the loan business generates sufficient profit thanks to highinterest spread and loan growth rate, on the other side from the underdeveloped stage of capital markets as well aslimited experience and resources in cross-sector services. Another remarkable fact is that only a small fraction ofinvestment securities (BOC for example 5.3% in 2008) belongs to the active trading securities designated as theasset class financial assets at fair value through profit and loss and the majority of the investment securities are

high-graded bond securities of the public sector (BOC for example with 66.7% in government and public sectorbonds, 27.1% in bonds of financial institutions, 5.5% in corporate bonds and 0.8% in equity and other securities in2008). Trading securities in Chinese banks are mainly fixed-income products of lower risk class and in someextent foreign currency and interest rate derivatives for risk hedging.

When we compare the results from 2008 with those of 2007 shown in table 2, it is noticeable that the highvolatility in trading income is in great contradiction to the stable income flows from the most positive fees andcommissions income. In crisis time, losses from trading can even diminish the total net fee and commissionincome, as the case of Deutsche Bank in 2008. Fees and commissions remain as stable and positive revenuesources, since banks act mostly as value creator through providing financial market related services. Thecomposition of non-interest income in fees and commissions as well as in trading of each bank shows the extentof a banks transformation from a capital market service provider and advisor to capital market risk taker.

in million USDNet G/L on Trading

and Derivatives

Net G/L on Other

Securities

Net G/L on Assets at

FV through Income

Statement

Sub-totalNet Insurance

Income

Net Fees and

Commissions

China

ICBC 271 (65) (101) 106 n.a. 6,330

CCB 352 (324) 110 138 n.a. 5,531

ABC (647) 71 n.a. (576) n.a. 3,424

BOC (1,017) 289 164 (564) 1,010 5,747

BoCom 224 33 n.a. 257 n.a. 1,271

Europe

Deutsche Bank (49,756) (969) 34,530 (16,195) n.a. 14,339

BNP Paribas (16,146) (478) 19,028 2,403 4,604 8,617

Socit Gnrale 6,750 (249) (410) 6,091 775 10,906

Banco Santander 1,764 1,557 893 4,215 370 12,429

UniCredit (3,710) n.a. (773) (4,482) 61 13,364

UK

Barclays Bank 2,327 392 61 2,780 1,576 15,528

Royal Bank of Scotland (10,312) n.a. n.a. (10,312) n.a. 10,621

HSBC Bank 5,480 151 (2,026) 3,606 1,951 7,309

Bank of Scotland (5,473) n.a. n.a. (5,473) 392 3,373

Lloyds TSB Bank (17,095) 35 n.a. (17,060) 15,277 4,686

USA

Bank of America (346) (1,671) n.a. (2,017) n.a. n.a.

Citibank (4,058) (1,900) n.a. (5,958) n.a. n.a.

HSBC Bank USA n.a. n.a. n.a. n.a. n.a. n.a.

Sallie Mae-SLM Corporation n.a. (186) n.a. (186) n.a. 461

RBS Citizens 79 78 n.a. 158 n.a. n.a.

Source: Bankscope

Table 1. Overview of Non-interest Income Business in Top 5 Banks in 2008


20/29

20

Figure 15. Trading Securities / Total Assets - Top 5

0%

5%

10%

15%

20%

25%

30%

2003 2004 2005 2006 2007 2008

B5 EURO UK USA

Source: Bankscope

in million USDNet G/L on Trading

and Derivatives

Net G/L on Other

Securities

Net G/L on Assets at

FV through Income

Statement

Sub-totalNet Insurance

Income

Net Fees and

Commissions

China

ICBC 178 56 (186) 47 n.a. 5,046

CCB 111 171 47 328 n.a. 4,119

ABC 227 5 n.a. 231 n.a. 3,025

BOC 406 (429) (368) (392) 1,206 4,675

BoCom 33 86 n.a. 119 n.a. 933

Europe

Deutsche Bank 5,370 1,087 4,438 10,894 n.a. 16,842

BNP Paribas 8,984 1,173 1,765 11,922 3,942 8,664

Socit Gnrale 13,143 22 (388) 12,777 1,035 10,317

Banco Santander 3,182 n.a. n.a. 3,182 438 11,019

UniCredit 772 1,762 (5) 2,530 45 12,924

UK

Barclays Bank 7,528 1,121 587 9,236 1,039 15,445

Royal Bank of Scotland 2,287 n.a. n.a. 2,287 n.a. 12,062

HSBC Bank 6,983 1,105 252 8,341 495 8,379

Bank of Scotland 370 n.a. n.a. 370 298 5,261

Lloyds TSB Bank 6,280 10 n.a. 6,290 (4,190) 5,261

USA

Bank of America (3,183) 236 n.a. (2,947) n.a. n.a.

Citibank (2,804) 468 n.a. (2,336) n.a. n.a.

HSBC Bank USA n.a. n.a. n.a. n.a. n.a. n.a.

Sallie Mae-SLM Corporation n.a. (96) n.a. (96) n.a. 492

RBS Citizens 54 82 n.a. 137 n.a. n.a.

Source: Bankscope

Table 2. Overview of Non-interest Income Business in Top 5 Banks in 2007

To test the impact of securities trading on asset return, we extent the above regression analyses by adding threeregression variables trading securities ratio and available for sale securities ratio as well as customer deposits ratioto the regression equation (China, Europe, UK and USA model 2). The results are summarized in appendix C.

The surprising finding is that proprietary trading, either classified as trading securities or available for salesecurities, does not generate significant higher asset return over the testing period in all testing fields, as indicated

by the insignificant coefficients of securities ratio.41

This could result from the fact that the high volatile tradingresults from one year to another are only averaged out over years and do not generate constant higher asset return.

The coefficients of net interest margin in extended models reconfirm that the profitability in Chinese banks ismainly driven by higher interest margin (0.1328 in China model 2), whereas the effect of margin in large Western

41 We replicate the extended model regression by excluding data sets from 2008 for the four testing fields to factor out the impact ofthe extreme market scenario in the crisis. Neither the coefficients of trading securities ratio nor those of available securities ratio arestatistically significant at 10%.


21/29

21

banks42

is not significant. Non-interest income contributes constantly to higher asset return in European and UKbanks with coefficients of 0.0109 and 0.0122 respectively.

Risk pattern

In the following, we investigate the risk reduction effect through diversification in non-interest income. To realizerisk reduction through diversification, it is expected that:

Hypothesis 3: The volatilities of interest income and non-interest income are not or negatively correlated.

Regarding non-interest income (NII), we further separate the effects of fees and commissions (FC) and tradingincome, since we believe fees and commissions business and trading business have different risk pattern. Wedetermine the risk term as the standard deviation of the outcome, nominated with the respective average value

43.

The risk factor for interest income (II) of i bank is calculated with the standard deviation of the banks outcome ofinterest income for the sample years, nominated with the average interest income of the bank:

Risk IIi = Standard deviation ( IIi t ) / Average IIi

The same method is applied to calculate the risk factor for fees and commissions as well as for total non-interest

income. We tried to determine the risk factor also for trading income, came however to only a few data sets, sincethe average income for the sample years usually turned to be negative. We test the correlations of the risk factorsin the three business lines. As shown in table 3, fees and commissions income is significantly correlated withnon-interest income in China, the UK and the USA with correlations of 0.2362, 0.4501 and 0.4823 respectively,which indicates that fees and commissions are driving forces of non-interest income in those countries. In alltesting fields, the correlations between interest income and non-interest income are statistically not significant,which implies the potential of risk reduction through combining interest income with non-interest income.

Noticeable, fees and commissions income is positively correlated with interest income in UK and US banks (withcorrelations of 0.2877 and 0.4834 respectively), but not significant in China and in Europe. In the UK and theUSA, both interest income and fees and commissions could be more influenced by movement of macroeconomicdeterminants like interest rate level or general capital market mood, whereas fees and commissions in Chinese andEuropean banks could more stem from stable sources like credit card fees, custody fees which are less sensible todeterminants of interest income.

II FC NII II FC NII II FC NII II FC NII

II - - - -

p-value

obs.

FC 0.0846 - -0.0041 - 0.2877 ** - 0.4834 ** -

p-value 0.4078 0.9435 0.0450 0.0226

obs. 98 305 49 22

NII -0.1257 0.2362 ** - -0.0091 0.0316 - 0.0247 0.4501 *** - 0.1208 0.4823 ** -

p-value 0.2174 0.0192 0.8734 0.5831 0.8608 0.0012 0.2515 0.0230

obs. 98 98 311 305 53 49 92 22

II: interest income; FC: fees and commissions income; NII: non-interest income.

USA

Table 3. Risk Correlation

*** / ** / * Statistically significant at 1% / 5% / 10%

Source: Bankscope

China Europe UK

We further compare the risk factors for interest income, fees and commissions income and non-interest incomeseparately

44for the sample period to compare the magnitude of the risks. As shown in table 4, the fluctuation of

revenues from fees and commissions as well as from total non-interest income business in Chinese and UK banksis statistically significant higher that that in traditional interest income business (0.3068, 0.5387 and 0.2917,0.3417 higher for China and the UK respectively). Non-interest income in US banks also has a higher volatilitythan interest income (0.2030 higher). However, the variation of fees and commissions business in Europe and in

42 By adding two variables trading securities ratio and available for sale securities ratio, the sample size is reduced. We believe theremaining sample banks in regression model 2 are mainly large banks, since the financial reporting are more detailed.43 In case the average value turns to be negative, the risk term is not included in the analysis, with the assumption that long-termaverage return should not be negative.44 The risk factor for trading income is not used here since most mean values in trading income turn to be negative as the result ofsharp down-ward movement in capital markets in the financial crisis.


22/29

22

the USA and that of total non-interest income in Europe are statistically not significantly higher than that ofinterest income. The overall results show that especially European banks have build up stable non-interest incomesources.

II FC NII II


23/29

23

distribution ability due to reputation of established brand and vast distribution network in bank branches.

Bank Affiliate Conglomerate Stand-aloneBank >

OtherP value

Affiliate >

OtherP value

Conglomerate >

OtherP value

Total assets 2008 in bn CNY 5.8746 5.3785 5.8679 3.4274 1.8748 *** 0.0080 1.3057 * 0.0849 2.1151 *** 0.0000

obs. 29 19 73 284 405 405 405

Net assets 2008 in bn CNY 4.9928 5.1457 5.5560 3.2632 1.2007 ** 0.0437 1.3226 * 0.0683 2.0230 *** 0.0000

obs. 33 20 74 298 425 425 425

*** / ** / * Statistically significant at 1% / 5% / 10%

Source: wind

Table 6. Fund Total Assets Comparison

Variable

Mean Difference

The favorable fee condition could be another explanation for the attractiveness of bank related funds. Thanks tothe larger size of assets managed by a single fund and synergies with banks own asset management operation,especially bank funds are able to offer fund products with favorable condition by lowering management andcustody fee, which in turn attracts more assets and aggravates the economic of scale. In average, bank fundsrequire 0.12% and 0.01% lower management and custody fee respectively, as results in table 7 show.

Bank Affiliate Conglomerate Stand-aloneBank

China Banking

Documents

Transcript of China Banking