64687 POVERTY THE WORLD BANK REDUCTION AND ECONOMIC MANAGEMENT NETWORK (PREM) Economic Premise SEPTEMBER 2011 • Number 66 JUN 010 • Numbe 18 Trade Finance during the 2008–9 Trade Collapse: Key Takeaways Jean-Pierre Chauffour and Mariem Malouche Trade finance matters for trade, and when financial markets and world trade collapsed three years ago, a shortage in trade finance was hailed as a possible culprit. Because of the potential for global repercussions, world leaders called on the international community to act swiftly to avoid a depression. Governments and international institutions intervened to mitigate the impacts of the crisis. Then the economy bounced back, and trade picked up. But what did we learn from the crisis? In retrospect, what role did trade finance actually play? Did the freeze in the financial markets cause the unprecedent- ed drop in global trade in 2008–9? This note presents evidence on the role of trade finance during 2008–9 and highlights a few takeaways on the data and knowledge gap of trade finance and government interventions during financial crises. The 1997–98 Asian crisis had already illustrated the critical guishing characteristic is that it is offered and obtained not only role that trade finance plays during a financial crisis, especially through third-party financial institutions (trade finance instru- its effects on trade, but that crisis remained regionally confined, ments that vary in terms of risk), but also through interfirm and international institutions and regulators largely blamed transactions, that is, involving contracts between buyers and the opaque financial sector in the affected economies for the suppliers. While banks play a central role in facilitating trade, a crisis. In contrast, the 2008–9 crisis originated in the United large share of trade finance occurs through interfirm, open ac- States, which has one of the most transparent and sophisticated count exchange. As such, it was also not clear whether govern- financial markets, and quickly spilled over to the European ments’ interventions in favor of a specific segment of the finan- Union and the rest of the world. Policy makers, central bankers, cial system—the trade finance market—were justified and and finance ministers from around the world found themselves warranted, and whether they would be effective in filling any in largely uncharted territories. They had to contemplate policy market gap. actions to channel liquidity into the real economy to support Trade Finance during the Great Trade Collapse (Chauffour trade transactions in the absence of data on trade finance. The and Malouche 2011) is an attempt to answer these questions dearth of data complicated the estimation of a possible trade and provides policy makers and analysts with a comprehen- finance “market gap� and of whether trade finance was indeed sive assessment of the role of trade finance in the 2008–9 a major factor driving the fall in global trade. “great trade collapse� (Baldwin 2009) and the subsequent role Trade finance differs from other forms of credit, for exam- of governments and institutions to help restore trade finance ple, investment finance and working capital. Its most distin- markets. 1 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise Figure 1. Trade Finance Arrangements, by Market Share open account cash in advance bank trade �nance 38%–45%, $6.0–$7.2 trillion 19%–22% 35%–40% credit covered by BU arm's-length $3–$3.5 trillion $5.5–$6.4 trillion members inter�rm nonguaranteed $1.25–$1.5 trillion $15.9 trillion in global merchandise trade (2008 IMF estimate) Sources: IMF staff estimates from IMF/BAFT-IFSA surveys of commercial banks (IMF-BAFT 2009; IMF and BAFT-IFA 2010) and Berne Union database. Note: BU = Berne Union. IMF = International Monetary Fund. BAFT-IFSA = Bankers’ Association for Finance and Trade-International Financial Services Association. Evidence Collected on the Role of Trade Large banks proved more cautious than small and medium Finance banks regarding countries seen as posing high financial risks, and they were also more likely to request confirmations or ex- Evidence from banks in advanced and emerging economies port credit insurance. In contrast, small and medium banks Assessment of trade finance conditions is complicated by the ab- were more likely than large banks to manage risk by requiring sence of organized markets for bank-intermediated trade finance greater collateral or stronger covenants. The 2010 Internation- and the proprietary nature of bank information about customer al Chamber of Commerce survey examined Society for World- relationships. To fill this gap, the International Monetary Fund wide Interbank Financial Telecommunication (SWIFT) mes- (IMF) and the Bankers’ Association for Finance and Trade sage data and found evidence of increased risk aversion by (BAFT)—now merged with International Financial Services As- banks and customers, including refusals to honor letters of sociation (BAFT-IFSA)—conducted four surveys of banks during credit because of discrepancies in documents (ICC 2010). the crisis to gather information on the market shares of financing The survey evidence on pricing is also consistent with a de- products. The surveys suggested that about 40 percent of trade mand-driven theory in which the decline in trade finance plays finance was bank intermediated and that the rest was split be- no more than a modest role in the decline in merchandise tween cash in advance (about 20 percent) and open account trade. The survey results indicate some increased pricing for (about 40 percent), including interfirm financing (figure 1). trade finance, at least relative to banks’ cost of funds. Other While in times of financial crises, interfirm trade credit is things being equal, the increased pricing should have reduced often more resilient than bank-intermediated trade finance, at the use of bank-intermediated trade finance as a share of trade. least in the short-term, the bank-intermediated component The increased share of bank-intermediated trade finance de- was relatively resilient during the 2008–9 crisis. To be sure, spite increased pricing also suggests domination by demand banks were increasingly cautious with real sector customers factors such as exporter risk aversion. and counterparty banks, and pricing margins often increased. They adopted stricter risk management Table 1. Reasons for the Decline in Value of Trade Finance practices in response to higher risks. They differenti- Percent of respondents ated more, depending on the individual client, the All Small Medium Large business segment (trading, retail, commodities, and so Reasons banks banks banks banks on), and home country. Banks also limited their own Fall in the demand for trade activities 85 81 90 80 risk through expanded insurance, shorter loan matur- Fall in the price of transactions (for example, 38 25 24 56 ities and stronger covenants, and by requiring higher commodity prices) cash deposits or other collateral from clients. But these Less credit availability at your own institution 30 19 24 40 factors were more than offset by an increase in risk Less credit availability at counterparty banks 30 6 24 48 aversion on the part of exporters seeking protection from risk. As a result, the share of world trade sup- Shift toward open account transactions 23 19 33 16 ported by bank-intermediated trade finance appears Shift toward cash-in-advance transactions 21 31 14 20 to have increased during the crisis. The causes of the Decline in support from export credit agencies 8 0 5 16 increased price and decreased value of trade finance Decline in credit from multilateral institutions 0 0 0 0 appear to be mostly spillovers from broader financial Other reasons 18 31 10 16 markets and the recession-induced decline in the val- Source: IMF/BAFT-IFSA Trade Finance Survey, March 2010. ue of international trade rather than specific problems Note: This reflects only the views of the 61 respondents that reported a decline in value of trade finance in at least one geographic region presented and that subsequently marked at least one in the trade finance markets themselves (table 1). option for the current question. 2 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise In sum, bank-intermediated trade finance largely held up than large firms because of their weaker capital base and bar- during the 2008–9 financial crisis even as it came under several gaining power compared to global buyers as well as banks. sources of strain. The value of trade finance fell at the peak of Also, SMEs experienced higher increases in the cost of trade the crisis, but it fell by consistently smaller percentages across finance instruments. Many SMEs operating in global supply regions than did the export declines in the same regions. As a chains or in the sectors most affected by the slow global econ- result, the share of bank-intermediated trade finance in world omy, such as in the auto industry, reported being constrained trade increased during the crisis. This larger share developed both by the banking system and by the drop in export reve- despite considerable headwinds. Banks supplying trade finance nues and buyer liquidity. shared the general increase in risk aversion observed in broader Banks in developing countries became more cautious, risk financial markets, and they restricted their supply of trade fi- averse, and selective in light of the drastic reduction in global nance to certain countries or sectors and otherwise tightened financial liquidity and in the number of intermediary players. credit conditions. Banks also increased pricing margins, driven Interviews with banks confirmed the increase in pricing and by both increased perceptions of default risk and higher capital drop in trade credit volume. Yet the drop in volume seemed to requirements, the latter partially due to Basel II requirements. more reflect lack of demand due to the global recession than Impact in developing countries the increase in pricing. Moreover, lack of liquidity in local cur- The World Bank sponsored a unique bank and firm survey in rency did not appear to be an issue. By April 2010, trade fi- 14 developing countries across five regions in 2009 and 2010. nance value and volume—particularly for interfirm trade cred- Survey findings confirmed that the global financial crisis con- it—also bounced back. Firms’ revenues picked up along with strained trade finance for exporters and importers in develop- the economic recovery, as did interfirm trade finance. However, ing countries. Yet the drop in demand emerged as firms’ top banks remained relatively risk averse because they needed to concern (figure 2). The lack of export revenues put pressure on deleverage and reassess underwriting risks. As a result, prices of firms’ cash flow and, therefore, on their capacity to fund their trade finance instruments and spreads, although narrowing, export and import transactions. The survey revealed some styl- remained higher than precrisis levels. A large number of firms ized facts at the firm, bank, and country-income levels. in Egypt, the Philippines and South Africa, where bank inter- Firms that relied largely on the banking system for trade fi- mediation is important, claimed that banks were still imposing nance suffered from more risk averse and selective local banks. stringent eligibility criteria for trade finance transactions. Over- In contrast, firms that relied mostly on interfirm financing all, 45 percent of surveyed firms reported that banks were as and self-financing were most affected by the slowing global risk averse in the last quarter 2009 as in they were in the last economy, the lack of and cancellation of orders, delays in buy- quarter of 2008 (figure 3). ers’ payments, and shorter maturity imposed by suppliers. The impact of the crisis was also heterogeneous across coun- Small and medium enterprises (SMEs) were more affected tries. The three low-income African countries where the survey was conducted (Ghana, Kenya, and Sierra Leone) seemed rela- tively more insulated from the financial crisis as of March– Figure 2. Lack of Orders Main Reason for Export Decline (2009) April 2009. Their primary trade finance constraints originated from more structural problems, such as poorly developed 60 banking systems and trade finance institutions as well as mac- roeconomic imbalances. Many of the African exporters have 51 50 traditionally relied on self-financing and cash in advance; there- fore, their exports were also hit by the drop in commodity pric- number of respondents: 190 es and global demand from their main export markets. The 40 percent drop in their cash reserves further constrained their trade fi- 30 nance. The financial crisis mainly added strains on the coun- 25 tries’ domestic financial systems and hindered SMEs and new 18 firms seeking to diversify away from commodity exports. 20 Looking at the impact of the financial crisis on African ex- 7 porters, most interviewed firms in Africa did not experience 10 direct difficulties with trade finance (Humphrey 2011). Yet, 0 indirectly, the financial crisis—through its effects on global de- no new cancelled Lack of Lack of mand and price volatility—led to deterioration of firms’ credit- orders orders �nance on �nance on worthiness and a decline in their access to trade finance. More- buyers’ part company’s part over, the survey underscores the differentiated impact the crisis Source: Malouche 2009. may have had by firm type: scarce bank finance reportedly was 3 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise Figure 3. Are Banks Still Imposing More Stringent Credit Eligibility As such, there was a compelling institutional and economic Criteria for Trade Finance Transactions? case to support trade finance cooperation. Indeed, one clear les- 100 son from the Asian financial crisis was that—in periods prone to 90 lack of trust and transparency as well as to herd behavior—all 80 actors, including private banks, export credit agencies, and re- 70 percent 60 gional development banks, should pool their resources to the 50 extent practicable. Cooperation among players is particularly 40 important because of the lack of a comprehensive and continu- 30 20 ous data set on trade finance flows. 10 Government and institutional response to the 2008–9 0 financial crisis ile of a ia a ru a sia Uk y e s e ne an ny ric in d Pe Ch rk p. In response to the financial crisis, many governments and inter- ni In ra Ke pi Gh Af Tu Re Tu ilip h national financial institutions (IFIs) put in place programs that ut b Ph ra So A either injected liquidity into banks or provided fiscal and mon- t, yp yes no Eg etary stimulus to the economy, sometimes directly in support Source: Malouche 2011. of affected exporting firms. Central banks with large foreign Note Country breakdown by number of responses, for all respondents = 223. exchange reserves could supply foreign currency to local banks and importers, generally through repurchase agreements. And government intervention was not reserved to developed coun- channeled mainly to firms with established exporting records tries. The central banks of Argentina, Brazil, India, Indonesia, and regular customer relations, leaving SMEs and new entrants the Republic of Korea, and South Africa, to name a few, also that lacked relationships with banks and customers in a dire massively supported their local banks. The measures helped situation. mitigate the global decline in output and trade flows and di- Government and Institutional Intervention rectly and indirectly supported the provision of trade finance— stimulating more confidence in the outlook of individual coun- Notwithstanding uncertainty about the size of the trade fi- tries, reducing risk premiums, and providing more direct nance gap and its potential role in the drop in trade, govern- financing to financial institutions. However, many developing ments around the world were compelled in the fall of 2008 to countries were not in a position to extend credit or expand ex- intervene to mitigate the impacts of the crisis on their domestic isting trade finance facilities, and therefore needed support. economies. The exceptional nature of the crisis called for im- The World Bank Group, through the International Finance mediate action. A critical question was whether the supply of Corporation (IFC), was quick to act by strengthening its trade trade finance declined because of market or government fail- facilitation programs between November 2008 and April ures, and, hence, whether there was a rationale for public inter- 2009. The IFC Global Trade Finance Program (GTFP) doubled vention to address such failures. Two broad cases that could its revolving ceiling to $3 billion in late 2008 to support emerg- create a real trade finance gap are: (i) insufficient supply (“miss- ing markets’ trade finance. Leveraging the experience gained ing markets�) and (ii) supply at prices temporarily too high to from the GTFP, the IFC launched the Global Trade Liquidity meet demand (“overshooting markets�)—both of which may have had temporary relevance in fall 2008. Program (GTLP) in July 2009 to rapidly mobilize and channel Government actions supporting trade finance were ratio- funding to support underserved developing-country markets nalized on the basis of the potential damage shrinking trade fi- by providing trade credit lines and refinancing portfolios of nance might have on the real economy. International supply trade assets held by selected banks. Additionally, the new pro- chain arrangements globalized not only production, but also gram’s premise was to leverage IFC funding by creating a his- trade finance. Sophisticated supply chain financing opera- toric collaboration with other IFIs that also contributed their tions—including those for SMEs—rely on a high level of trust financial resources to the GTLP. Both programs successfully and confidence in global suppliers that they will deliver their facilitated trade during the crisis period. As the world economy share of the value added and have the necessary financial means slowly recovers from the crisis, the IFC will bring the GTLP to to produce and export in a timely manner. Any disruption in an end, starting in 2012. the financial sector’s ability to provide working capital or pre- Regional development banks also quickly responded by scal- shipment export finance, issue or endorse letters of credit, or ing up their trade finance facilities. The European Bank for Re- deliver export credit insurance could create a gap in complex, construction and Development increased the overall program outward-processing assembly operations and lead to a contrac- limit of its Trade Facilitation Program from €800 million to tion in trade and output. €1.5 billion. The Asian Development Bank ramped up the ac- 4 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise tivities of its Trade Finance Program to support $2 billion in been a moderate factor in the sharp drop in trade flows in trade in 2009, an increase of more than 300 percent over 2008. 2008–9. Trade finance and trade volumes dropped most- Further enhancements of these programs were agreed on at the ly as a result of the spillover of the financial crisis to the real G-20 summits, particularly, as already noted, IFC’s establish- economy, including through lower activity and destock- ment of a liquidity pool to allow cofinancing of operations with ing. The demand effect was further amplified for firms banks in developing countries. From this perspective, the Afri- operating in global supply chains or in sectors that were can Development Bank established a $1 billion Trade Finance most affected by the slow global economy, such as the auto Initiative in January 2009 as part of its broader package of crisis industry. response initiatives. For its part, the Inter-American Develop- 5. SMEs have been particularly vulnerable to the tightening ment Bank (IDB) had already put in place its Trade Finance of trade finance conditions. The lack of access to afford- Reactivation Program (TFRP) when the crisis hit. TFRP sup- able trade finance has been especially detrimental to cer- ported IDB’s fast response in Latin America and the Caribbean, tain firms (for example, SMEs and new exporters), partic- strengthening supply-side capacity and trade-related infra- ularly firms in developing countries with underdeveloped structure. In addition, the Trade Finance Facilitation Program financial systems and weak contractual enforcement sys- (TFFP), implemented in 2005, proved an effective fast-delivery tems. SMEs have been more affected than large firms be- vehicle for not only mitigating the effects of the liquidity crisis, cause of their weaker capital base and bargaining power in but also for expanding trade finance for financial intermediar- relation to global buyers and banks. Also, SMEs have been ies and their clients. subjected to more large increases in the cost of trade fi- nance instruments, with banks being more risk averse and Lessons Learned preferring to work with large, sound, multinational firms. The mostly original and unpublished contributions to Trade 6. Bankers and some international institutions consider Ba- Finance during the Great Trade Collapse suggest the following 10 sel II regulations to have further constrained the supply of takeaways. trade finance during the crisis and in the postcrisis envi- 1. The lack of trade finance data is impeding policy formula- ronment, especially banks based in low-income countries tion. The absence of data covering all aspects of trade fi- (as well as second- and third-tier banks in middle-income nance (bank-intermediated and interfirm) has proven to countries). They have called on regulators to carefully be a major constraint to measuring the extent of the trade study the potential unforeseen impacts of proposed Basel finance shortfall and its effect on trade flows during the fi- III changes on trade finance. In particular, banks argue that nancial crisis. New initiatives such as the International the increase in the new liquidity and capital prudential re- Chamber of Commerce’s build-up of the Trade Finance quirements and the nonrecognition of trade assets as high- Register are a significant step forward because they will ly liquid and safe would lead to a significant increase in the create a living database of the trade finance market and cost of banks providing trade finance, which in turn will may help demonstrate the resilience of trade finance. lead to a lower supply, higher prices, or both. Conversely, 2. Trade finance matters for trade. Results from bank and regulators have maintained the view that under Basel II firm surveys conducted during the crisis as well as postcri- and III, the increase in capital for trade finance exposures is sis empirical analyses all indicate tighter trade finance con- not any greater than for other exposures. The new leverage ditions during the crisis and significant adverse effects on ratio and the new liquidity rules will not have any system- trade flows. atic impact on trade finance, though they may affect a few 3. Not all forms of trade finance are equal. Although the cri- large, complex, or wholesale-funded banks, albeit for rea- sis constrained both bank-intermediated trade finance and sons unrelated to their trade finance activities. Even in interfirm trade credit, empirical findings suggest that in- these cases, the impact on trade finance is not expected to terfirm trade credit may be more resilient than bank-inter- be greater than on any other class of asset. Given the di- mediated trade finance in times of crisis. Trade credit of- verging views, the Basel Committee on Banking Supervi- fers features that make it safer, given the better information sion (BCBS) has established a working group to study the that buyers and suppliers have on clients’ creditworthiness impacts of regulation on trade finance, and—at the request and the liquidating feature of trade credit. Although trade of the World Bank and the World Trade Organization—the credit (particularly among supply chains) could be a conta- G-20 will take stock of the situation at its November 2011 gion leading to sharp drops in trade during crises, it also meeting. contributes to a quicker rebound when economies recov- 7. The international community responded swiftly to the er—a pattern observed in Southeast Asia during the crisis. trade finance crisis. The G-20 orchestrated the quick and 4. Trade finance was not the main driver behind the 2008 collective actions of governments and the international fi- trade collapse. The shortfall in trade finance seems to have nancial community. This led to a set of cofinancing ar- 5 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise rangements among development banks, export credit continuing analysis of the issues by academics, practitio- agencies, foreign commercial banks, private insurance un- ners, and other interested stakeholders. derwriters, and investment funds. While part of the G-20 About the Authors support was directed mostly at a handful of large banks and international banking groups, the support from the Jean-Pierre Chauffour is a Lead Economist and Regional Trade Co- IFC and regional development banks—in terms of both in- ordinator in the Middle East and North Africa Region, and surance and liquidity—has targeted mainly smaller banks Mariem Malouche is an Economist in the PREM Trade Group. and banks in developing countries. The authors would like to thank Bernard Hoekman for his com- 8. A timely exit for the large-scale trade finance support pro- ments on this note. grams implemented to mitigate the impacts of the crisis is key. As the global economy slowly recovers and demand Note rises, some governments appropriately cut back their trade 1. Brazil, Chile, and Peru in Latin American and the Caribbe- finance programs to avoid displacing legitimate private sec- an; Indonesia and the Philippines in East Asia and the Pacific; tor activity. Similarly, beginning in 2012, the IFC will India in South Asia; Ghana, Kenya, Sierra Leone, and South wind up the GTLP, which was set up in response to the Africa in Africa; the Arab Republic of Egypt and Tunisia in the crisis. Setting clear time limits, planning exit strategies for Middle East and North Africa; and Turkey and Ukraine in Eu- intervention programs, and sharing, rather than fully un- rope and Central Asia. derwriting, risks are important considerations to limit moral hazard and the crowding out of commercial banks References in times of financial crises. Baldwin, Richard, ed. 2009. The Great Trade Collapse: Causes, Consequences 9. There is a case for implementing and maintaining specific and Prospects. VoxEU.org, London: Centre for Trade and Economic Inte- gration. programs to support vulnerable segments of the trade fi- Chauffour, Jean-Pierre, and Thomas Farole. 2009. “Trade Finance in Crisis: nance market. For example, continued uncertainty in Market Adjustment or Market Failure?� Policy Research Working Paper some markets (such as low-income countries with under- 5003, World Bank, Washington, DC. developed financial systems and weak contractual enforce- Chauffour, Jean-Pierre, and Mariem Malouche. 2011. Trade Finance during the Great Trade Collapse. Washington, DC: World Bank Group. ment) or among some firms (such as SMEs and new ex- Humphrey, John. 2011. “Trade Finance in Africa: A Survey of Firms.� In Trade porters) calls for vigilance on the suitability and timing of Finance during the Great Trade Collapse. Washington, DC: World Bank the retrenchment of governments and international orga- ICC (International Chamber of Commerce). 2010. “Rethinking Trade Finance 2010: An ICC Global Survey.� ICC, Paris. nizations’ trade finance programs. At the Seoul G-20 IMF and BAFT-IFSA (International Monetary Fund and Bankers’ Association meeting in November 2010, the international communi- for Finance and Trade International Financial Services Association). 2010. ty expressed particular concern for low-income countries “Trade Finance Services: Current Environment & Recommendations: that may still be facing severe difficulties in accessing trade Wave 3.� Report by FImetrix for IMF and BAFT-IFSA, Washington, DC. ———. 2010. Trade Finance Services: Current Environment & Recommendations: finance at affordable cost, particularly in import finance. Wave 3. Report by FImetrix for IMF and BAFT-IFSA, Washington, DC. 10. Finally, there is still an important knowledge gap regarding Malouche, Mariem. 2009. “Trade and Trade Finance Developments in 14 De- the effect of trade finance on trade and the role of trade fi- veloping Countries Post September 2008.� Policy Research Working Paper 5138, World Bank, Washington, DC. nance during crises, as well as on the appropriate banking ———. 2011. “World Bank Firm and Bank Surveys in 14 Developing Countries, regulations and supervisory standards for banks’ trade fi- 2009 and 2010.� In Trade Finance during the Great Trade Collapse. Wash- nance portfolio exposure. This knowledge gap requires a ington, DC: World Bank. The Economic Premise note series is intended to summarize good practices and key policy findings on topics related to economic policy. They are produced by the Poverty Reduc- tion and Economic Management (PREM) Network Vice-Presidency of the World Bank. The views expressed here are those of the authors and do not necessarily reflect those of the World Bank. The notes are available at: www.worldbank.org/economicpremise. 6 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise