Herbert Poenisch: From Domestic to Global Banking-A Survey of Banking Systems in 31 Countries

Herbert Poenisch, Member of IMI International Committee, Former Senior Economist of BIS The BIS recently started publishing the domestic bank lending next to the cross-border bank lending for most of the 46 countries and regions reporting regularly to the BIS locational banking statistics(LBS)[i]. This allows a comparison of domestic versus international business of the banking systems of some 31 countries (see table). Thus a monetary survey variable (domestic lending) can be compared with a balance of payments variable (cross-border lending). However, both variables are stocks, the domestic bank lending and capital inflows and outflows through the banking system[ii]. This analysis looks at banking from a macro perspective rather than from a micro perspective as a recent study in the AIF of Zhejiang University and IMI has shown[iii]. Although the building components are banks of various definitions[iv], this analysis looks at whole banking systems.The details available in the LBS shed light on the role of banks in the intermediationof capital flows[v] and the structure of their intermediation. The paper will present in the first section a short summary of the history since WW2 of the internationalization of banking business, focusing on the major driving forces.  The table in the middle presents a snapshot in mid-2017 of the comparable size of cross-border claims and liabilities compared with the total banking assets and liabilities of the sample countries. This will be followed in the second section by an analysis of the major observations from the table and conclusions. Section 1: Historical phases of bank internationalisation In the postwar period international banking has developed in 5 distinct phases. The first phase was characterized by purely domestic banking systems, providing international finance only for trade, such as letter of credit (LC). In the second phase banks followed their clients abroad, providing mainly finance to the national companies doing business abroad. The third phase was characterised by the growth of the global interbank market, where domestic banks placed surplus funds in and borrowed from this market to balance their daily liquidity. The fourth phase, called internationalisation was setting up branches or subsidiaries abroad or purchasing local banks, financed mainly from the mother banks. The sixth phase called globalization started when overseas branches and subsidiaries started acting like local banks in the host countries. They accept local deposits and provide credit in local currency. 1.1 Domestic banking systems Domestically owned banks, servicing the domestic economies of nation states has been the traditional model, which in its purest form has not existed for a long time. During colonial times banks headquartered in the financial centres of empires provided financial services for the colonies. The main reason for banks going abroad during this period was to follow the clients or serve national interests abroad. Financing trade was the main part of their international business. However, banks were intermediaries of capital flows as they held correspondence accounts with foreign banks and moved funds (working balances) on exchange rate and interest expectations[vi]. 1.2 Following the customer With the expansion of global trade after WW2 and the foreign direct investment (FDI) mostly from the US, banks followed their clients to provide banking services. Other countries’ banks which followed this model were European banks and Japanese banks since the 1980s. In terms of the balance of payments (BOP) these were capital outflows for FDI but also trade finance in the financial account. The current account recorded earnings as interest income and dividends received. This explains why the US until 1990s had surpluses in its international investment position (IIP) although it has run increasing current account deficits since 1980. The lending to national clients happened either directly between domestic banks and overseas companies or through establishment of domestic banks abroad who onlent to national non-banks abroad. In both cases, capital export was created in the BOP. 1.3 Emergence of global interbank market Growth of domestic banks was constrained by the availability of domestic savings. This was alleviated by the emergence of the Eurodollar market in the 1960s, which were mostly available in offshore financial centres. Specialised financial centres such as Luxemburg in Europe and Hong Kong and Singapore in Asia played this role. Normal financial systems such as UK, NL, CH and BE also took on this role, where cross-border business grew faster than domestic business. The main drivers of this internationalization were banking regulations in the domestic financial markets (such as regulation Q in the US), supervisory arbitrage as well as concerns about the political immunity of financial transactions[vii]. Banks from protected domestic banking systems set up offices in financialcentres to tap the international interbank market. Although the entry costs were high, they were able to obtain interbank funding at lower interest rates than in the domestic market. Thus they could escape a restrictive monetary policy at home. This became a major concern for central banks, that the Eurodollar market would undermine domestic monetary policy through increasingly liberalized capital flows. These concerns continue until this very day, and prompted some countries, such as China to retain capital controls to drive a wedge between domestic and international interest rates. Banks mostly used this channel for short term financing. There are some countries which use overseas bank financing for their development goals.  Few countries such as Turkey, India, Indonesia, but also Finland use this channel to borrow from the international banking market. As bank financing is usually more costly than bond financing, using this facility for the purpose of development finance is the exception rather than the rule. The banking sector of China has been a net debtor in the international market for many years, unlike other current account surplus countries such as Germany and Japan. The net debtor position is partly explained by the offshore listings of several of the largest Chinese banks, which could explain some 30% of cross border liabilities. Another explanation is the channeling of offshore RMB deposits to the mainland, such as through HK[viii]. This again makes up 30% of total liabilities[ix]. As this channel of external finance became a permanent feature of bank funding, banks set up in host countries. They established branches and subsidiaries in major financial centres rather than just representative offices. This explains the large number of foreign banks in the major offshore centres, such as Hong, Kong, Singapore, Luxemburg but also in London, Tokyo, New York etc. Source: McCauley et al, BIS 2003 1.4 Internationalisation of banking Once these branches and subsidiaries were set up or local banks acquired through M&A in the 1990s not only infinancialcentres but in other prosperous emerging markets, they started lending to domestic clients in the host countries. The bulk of funding came still from the mother bank during this period (see scheme above). Thus this type of banking was reflected in the BOP as a capital outflows, partially compensated on the current account by interest earnings and dividends received. The main drivers of the internationalization in this period were the push-pull factors[x]. On the push side, the overbanking in many advanced countries which led to banking consolidation with fewer more powerful players ready to expand overseas , were facing small lending margins and falling profitability. On the pull side, emerging markets borrowers became eligible to borrow from the foreign banks, because domestic growth and structural reforms allowed domestic borrowers to service their bank loans. In addition, emerging markets became more transparent which helped the risk assessment by foreign banks. This was tested during the various financial crises in emerging markets, notably the Mexican Crisis of the mid 1990s and theAsian Crisis of the late 1990s. In the aftermath of these crises, authorities in emerging market economies sought to recapitalize their ailing banking systems and improve the overall efficiency of their financial sectors with the help of foreign banks[xi].During this period many large customers in prosperous emerging markets shifted from bank borrowing to capital markets, share issues and domestic bond borrowing[xii]. Although economic growth was good, bank lending did not reflect the full dynamics. 1.5 Globalisation of international banking By the beginning of the new century, prosperous emerging markets became more sophisticated in choosing and regulating foreign banks to provide a level playing field with their own banking champions.The presence of foreign banks has generally led domestic supervisory authorities to upgrade thequality and increase the size of their staff in order to supervise the more sophisticated activities andnew products being introduced by these banks[xiii].At the same time regulatory pressure on the mother banks became an incentive to deal with branches and subsidiaries at arms’ length. International banks became global players when they took responsibility for their own destiny, funding as well as risk management. They behaved like local banks, even running a branch network and accepting deposits from domestic customers[xiv]. While this business continued to flourish, there was no reflection in the BOP because no capital flowed from the mother bank to finance the operations of branches and subsidiaries.  They engage in local borrowing and lending, not classified as international transaction in the BOP accounting framework. As a result, standard BOPbasedmeasures of financial openness tend to underestimate the degree of globalbanking interconnectedness[xv]. Although this type of business largely escapes the BOP, the BIS banking statistics in the consolidated banking statistics (CBS) take this into account by recording the local business in host countries by internationally active banks[xvi]. What happened was fundsrather flowing the other way, once the mother banks ran into problems. The Japanese banks recalled funds already in the 1990s leading to negative capital outflows in Japan; the European banks’ retrenchment followed after the Global Financial Crisis[xvii] and continues until today. Overseas US banks stabilized their operation without major retrenchment. What has happened was rather a restructuring than a retreat from globalization in the following sense. The local or regional operations sold off by international banks were bought by local banks ambitious to internationalise their operations. They in turn became regional important players, such as Singaporean and Malaysian banks in ASEAN and Brazilian and Mexican banks in Latin America. Chinese banks have adopted a similar strategy. Matching their global peers, their funding was originally from the mother bank but is becoming increasing local. When this is achieved there will be less reflection in the financial account of the BOP. Global banking has thus a greater variety of players with more local focus[xviii]. Table: Domestic and cross-border banking claims and liabilities in 31 sample countries In bn USD at the end of 2Q2017 Sources: BIS LBS, PBOC, BNM, TCMB, FED *total is not the sum of components due to unallocated items. LC=local currency, FC=foreign currency Section 2: Observations on the domestic versus cross-border banking statistics The first observation to be made is the absolute size of the total claims (columns 4 and 6) and liabilities (columns 9 and 11). The size of bank lending is largest in China and Japan, both countries dominated by bank finance rather than capital markets. The US figure is much smaller as capital markets play a more important role. The second observation is that lending in big countries is usually dominated by domestic lending, with cross-border business playing a small role. Examples are Brazil, China, India, Indonesia and Mexico. Other big countries have individual explanatory factors. The US has traditionally, through its window to the world, New York, played a major role in international finance. Russia, after the collapse of the Soviet Union quickly opened its financial systemwhile domestic lending expanded slowly. The third observation is the share of cross-border claims (column 12) and liabilities (column 13) which are largest in the financial centres, such as SG and HK, but also in some countries where financial systems play an international role such as the UK, NL, CH and BE. As this is a BOP view, there are banks of different nationalities which perform the international business as residents. It can be assumed that Chinese owned banks in HK perform substantial cross-border business from HK rather than from mainland China as the domestic banking sector in HK is rather small. The fourth observation is that there are still a few countries’ banks which do not have a balanced share of their external claims and liabilities. They continue net borrowing from the international banking market (column 14) as outlined above in 1.3. Countries with a high ratio are Turkey, India, Indonesia but also Finland. Another indicator, which is not reflected in this table is the participation in the international interbank market for short term lending and borrowing[xix]. This varies greatly from country to country and can reach a substantial part of banks’ funding. While this is considered as stable funding during normal times it has been known to dry up during stress periods, such as in the aftermath of the Global Financial Crisis when banks were not sure of other banks’ exposure to the subprime market. As a result, interbank lending dried up and rates (such as overnight interest swap rate) shot up. Finally, this article does not offer an analysis over time. This would allow a statement on the role of domestic bank financing versus cross-border bank financing over time. Indicators in the BIS study on banking in emerging markets suggest that domestic bank financing (also through foreign banks’ establishments) has expanded faster than financing from abroad[xx]. Thus bank lending continued to expand without causing cross-border capital flows. Conclusion While the evolution from domestic banking systems into internationally active banking systems can be documented well from the BOP, measured by cross-border business as share of total banking business, the transformation into global banks cannot be monitored through BOP statistics. It needs further statistics which are collected by BIS from the consolidated banking statistics, which presently are provided only by 31 reporting, mainly advanced economies. In spite of these limitations, as international financial criseshave shown, banks located abroad can be significant – and volatile – sources ofcredit. Therefore, the LBS can provide a useful signal regarding potential fragilitiesin the financial system. In particular, the LBS can help monitor the build-up ofvulnerabilities associated with cross-border and foreign currency bank credit[xxi]. [i]BIS (2017): Local Banking Statistics. www.bis.org/statistics. Beforehand, the BIS statistics had to be complemented by the IMF International Financial Statistics. [ii]In the IMF methodology, this variable is not easy to calculate as the IMF does not single out sectors but rather types of transactions. There are three categories: Foreign direct investment, portfolio investment and other investment. Most of the cross-border banking transactions are to be found in the third, but also in the second category. See IMF (2010): Balance of Payments Manual no 6. www.data.imf.org [iii]AIF of ZJU and IMI (2017): East or west, home is best?—Are banks becoming global or local. In: IMI Review Vol 4, No 4, October www.aifzju.edu.cn [iv]In this context deposit taking institutions [v]The role of banks in capital outflows from China has been analysed by McCauley, R and Shu, Chang (2016): Dollars and renminbi flowed out of China. In : BIS Quarterly, March www.bis.org [vi]The close link between the Deutsche Mark and Austrian Shilling up to the introduction of the Euro in 1999 provided such arbitrage opportunities. [vii]One of the factors for the emergence of the Eurodollar market was the Soviet fear of a freeze of their deposits in the US. Thus they moved USD deposits to Europe, the Banqued’Europe du Nord in France. [viii]Hu, Hong and Wooldridge, Philip (2016): International business of banks in China. In: BIS Quarterly Review, June www.bis.org/publications [ix]See table LBS A5-CN [x]Jeanneau, S and Micu, M (2002): Determinants of international bank lending to emerging market countries. BIS Working Papers No 112, June www.bis.org/publications [xi]Goldberg, Linda S (2008): Understanding Banking Sector Globalisation. www.newyorkfed.org [xii]Turner, Philip (2006): The banking systems in emerging economies: how much progress has been made. In: BIS Papers No 28, August www.bis.org/publications [xiii]Mihaljek, Dubravko (2008):  Privatisation, consolidation and the increased role of foreign banks. In: BIS Papers no 28, August www.bis.org/publications [xiv]McCauley, Robert N, Ruud Judith S and Wooldridge, Philip D (2003): Globalising International Banking. In: BIS Quarterly Review, March www.bis.org/publications [xv]BIS (2017): Understanding globalization. In: Annual Report 2017, chapter VI www.bis.org/publications [xvi]BIS(2017): Consolidated banking statistics. www.bis.org/statisticsThe reporting is limited as presently only 31 countries report their banks’ cross-border business on a consolidated basis. [xvii]Claessens, Stijn and van Horten, Neeltje (2014): The Impact of the Global Financial Crisis on Banking Globalisation. In IMF Working Papers, WP/14/197 www.imf.org/publications [xviii]Claessens and van Horten, ibid. [xix]BIS (2017): BIS LBS captures 93% of cross-border interbank business. In: Statistics Bulletin, September www.bis.org/publications [xx]See table 4 in Mihaljek, Dubravko, ibid. [xxi]BIS (2017): ibid

DOWNLOAD