3. Looking at the Last Credit Cycle to Better Manage the Next One

International Monetary Fund. Western Hemisphere Dept.

Jump to Content Jump to Main Navigation

Author: International Monetary Fund. Western Hemisphere Dept.

Language:: English

Keywords:: REO; IMF staff calculation; country; ECCU country; economy; credit market; government bond; private bank; credit decline; exchange rate; Foreign banks; Securities markets; State-owned banks; Global

Download PDF (1.8 MB)

Abstract

The main financial challenge in Latin America is to increase financial depth safely, without exacerbating economic cycles. In recent years, firms and households increased their access to bank credit, in part as governments found in domestic bond markets an alternative source of funding. The bank credit cycle since 2005 shows that credit growth continues to be procyclical, but this cycle avoided the collapses in credit seen in the past. This was possible because banks funded credit mostly from a stable domestic base. Also bank liabilities to nonresidents—the most volatile source of funding—were relatively small. In terms of the role of bank ownership, foreign banks tended to amplify this time the cycle, while public banks played a countercyclical role. Although banks have successfully passed the recent “real live stress test,” this should not give room for complacency, especially as a new credit cycle begins.

This chapter examines selected aspects of domestic financing in Latin American economies over the most recent credit cycle, with an emphasis on the events surrounding the global crisis. The focus is on bank credit and, to a lesser extent, on bond financing. The main goal is to derive lessons from the region’s experience during the last financial cycle to better manage the next one.

The chapter starts with a brief overview of the main structural features of the banking system, while highlighting some relevant financing trends in Latin America. It then examines the behavior of banks and bond markets during the 2005–09 cycle and the incipient current recovery. Finally, policy implications for managing the new financial cycle are drawn from the analysis.

An Overview of the Financial System

The global financial crisis has highlighted the need to improve the understanding of credit dynamics to build safer and more stable banking systems. To do so, it is useful to understand the market structure and balance sheet composition of the banking system.

Bank credit continues to be the main source of funding in Latin America, in particular for the private sector (Figure 3.1). However, the depth of domestic bank financing continues to be relatively shallow in the region, as is evident when compared with other emerging markets. The average size of total bank credit has barely exceeded 40 percent of GDP over the past decade, half as large as in Asia and smaller than in emerging Europe (Figure 3.1). Exceptions include Brazil and Chile, where total bank credit to GDP exceeds 75 percent.

The credit market in Latin America is dominated by private, domestically owned banks (Figure 3.2). However, foreign-owned banks and crossborder flows do play an important role in many countries (Box 3.1). The provision of credit by public banks had been declining across the region before the Lehman events, but has since played a significant role in some economies such as Brazil, Chile, and Costa Rica.

A closer look at bank credit in Latin America reveals that most domestic bank credit is funded from resident sources. These mainly took the form of relatively stable deposits, as opposed to wholesale funding (Figure 3.3). In recent years, however, funding has increasingly relied on the issuance of securities (for example, bonds, including subordinated debt, and short-term instruments).

Box 3.1.

Structure of Private Sector Credit Market: Role of Foreign Banks

Latin American firms and households get bank credit from domestic and foreign institutions. For the average country, domestic and foreign banks that report to the Bank of International Settlements (BIS) provided about 30 percent of GDP in credit to the nonfinancial private sector by end-2009. About 90 percent came from banks with a physical presence in the country, whether domestic banks or foreign branches and subsidiaries. The local presence of foreign banks in domestic credit is very large in some countries, but for the median country in the region, it accounts for about 20 percent of domestic bank credit. Crossborder bank credit amounts to about 3 percent of GDP.

The extent of direct involvement of foreign banks varies considerably across countries. Crossborder bank financing is particularly large in Chile (about 10 percent of GDP) and about 4 percent of GDP in Brazil and Mexico. Foreign bank branches and subsidiaries of BIS reporting banks provide credit up to 25 percent of GDP in Chile, 19 percent of GDP in El Salvador, and about 15 percent of GDP in Mexico. In Brazil, this share has increased significantly over the last few years, but is still relatively small at about 6 percent of GDP.

Spanish banks play a dominant role. They account for about 50 percent of the total claims of global banks on Latin America. In turn, U.S. banks account for close to 30 percent, and British banks for close to 20 percent. Canadian, Dutch, and German banks account for less than 5 percent each. Nevertheless, the importance of global banks from individual advanced countries varies across countries in Latin America. The Spanish Banco Santander and Banco BBVA have an important presence in the two largest countries in the region, with Santander playing a large role in Brazil and Banco BBVA focusing its operations in Mexico.

The share of Brazil, however, has been increasing since late 2005, at the expense of that of Mexico’s. Because of their size, Brazil and Mexico receive the bulk of global bank claims on Latin America. The portfolio shift favoring Brazil may have reflected the view that economic conditions in Mexico might deteriorate more than in Brazil from the stress in the U.S. economy since 2007, given its close connection to U.S. manufacturing. The shares of Brazil and Mexico, however, appear to have stabilized since late 2009, probably reflecting the favorable impact of the U.S. recovery on Mexico.¹ Chile has received a relatively large and stable share of global bank financing over the last five years.

Credit to the private sector is the main asset on banks’ balance sheets. Banks also hold a large share of government and central bank assets (Figure 3.4). Foreign assets average about 10 percent of assets, although a few of the more dollarized banking systems also hold a nontrivial share (10–20 percent) of their assets abroad.¹

Against this background, the financing landscape has witnessed an important transformation in the main economies of the region: the expansion of domestic bond markets, from a very limited base in 2004. This is especially significant for the government and financial sectors (Box 3.2), while corporate bond markets have remained largely underdeveloped (Figure 3.1). As a result of this transformation, the public sector has been able to diminish its funding from banks, crowding in funds now available for the private sector.

Learning from the Last Financial Cycle

How Big Was the Recent Cycle?

Looking over the last 15 years, credit growth has been procyclical in Latin America, a common pattern in other emerging markets (Figure 3.5).² The amplitude of the cycle typically exceeds—by far—that of domestic demand or GDP. Thus, credit-to-GDP ratios tend to increase notably in upswings and decline during recessions.

Box 3.2.

Government Bond Markets in Latin America

A key financing transformation before the global crisis in the largest economies of the region (Brazil, Chile, Colombia, Mexico, and Peru) was the expansion of markets for domestic government bonds in local currency (Figure 3.1). Fostered by improvements in debt management practices, macroeconomic policies, and favorable external conditions, the process was similar to that seen in Asia.¹ The size of liquid domestic government debt in the region increased from 29 percent to 36 percent of GDP (US$450 billion to US$1.1 trillion). This expansion led to a fall of vulnerabilities associated with exchange rate and rollover risks. Exchange rate-linked debt was phased out, and fixed-rate instruments in local currency expanded rapidly. The average maturity of marketable debt also lengthened from 3 to more than 5 years. By early 2008, most governments in the region had built up relatively liquid bond markets with yield curves in local currency that spanned in some cases up to 15 years and in others up to 30 years or more (for example, Mexico and Peru), providing with it important pricing information for the economy. This transformation was supported by domestic agents (for example, banks and pension funds) and to a much lesser degree by foreign investors, who had a limited participation—the main exception was Mexico (CGFS, 2008).

How dependable have these markets been? Bond markets in Latin America were not immune to the jump in risk aversion and the global process of flight to dollar liquidity and quality. The Lehman Brothers’ episode led to a rapid and sharp increase in government yields, especially pronounced in the long end of the curve (see figure to the right, in particular the shaded area).³ However, the widening of domestic government bond yields proved to be short lived. By early 2009, yields were back down to precrisis levels. Econometric analysis indicates some decline in the sensitivity of domestic government bond yields to global risk aversion, as captured by the VIX index. Thus markets are slowly graduating into adulthood, reflecting improvements in the overall management of the economies and public finances (as also suggested by upgrades of sovereign credit ratings over the years, and by declining spreads). Still, the results confirm that these markets are not fully immune to contagion, and that they do not yet behave as mature markets (for example, Australia, Canada, New Zealand or Norway) where bouts of global risk aversion tend to lower bond yields.

However, the evidence does suggest that there is a declining trend in the sensitivity of bond yields to risk aversion in tranquil periods. Indeed, Chile has displayed negative sensitivity to risk aversion over the past year (light blue bar in figure to the right). This contrasts with the positive sensitivity that characterized these markets in the tranquil period that preceded the Lehman event (dark blue bar in the figure). Even in countries where markets reacted strongly during the crisis, such as Brazil, there are signs of a declining sensitivity during the past year. These results are encouraging in that they include the recent new round of turbulence in global markets that originated in Europe earlier this year (see Box 2.1).

Domestic bond markets saw an important decline in issuance in the fourth quarter of 2008. However, this partly reflects that government financing needs had already been met. Long-term improvements in government debt management and reduced fiscal deficits meant that the decline did not generate an immediate funding gap.⁴ When needed, governments in the region were able to issue, roll over, and swap debt in domestic bond markets.

Overall, while the evidence is not complete, bond markets did remain available as a source of funding, especially for governments. This was also true to some extent for high tier firms, but not for small and medium-size enterprises.

Note: This box was prepared by Camilo E. Tovar.¹ For an account of the policy challenges in developing local currency bond markets, see IMF (2002), Hausmann and Eichengreen (2005), BIS (2008), CGFS (2007), and Mathieson and others (2004). See Ocampo and Tovar (2008) for a discussion of how this transformation is linked to changes in the external financing patterns of the region.² See Borensztein and others (2008) for an overview of corporate bond markets in Latin America before the crisis.³ Chile displayed a behavior that diverged from the rest of the region, possibly suggesting that the local market is starting to mature.⁴ As an example, in September 2008, Colombia had completed its Annual External Financing Plan and 96 percent of its domestic financing needs.

During the last credit cycle in Latin America, real credit growth accelerated from 2005 onward and remained high, only to fall, with economic activity, following the Lehman Brothers event. That said, there has not been a collapse of real credit levels, as banking systems this time proved relatively resilient to external shocks.

During the upswing (June 2005 and September 2008), the average ratio of private credit to GDP increased by 10 percentage points.³ However, there were significant differences across countries in the region. For instance, in Brazil, Chile, Costa Rica, and Honduras the total increase in bank credit exceeded 15 percentage points of GDP. Crossborder lending to firms and households was a key part of the picture only in few countries, mainly in Central America (Figure 3.6).⁴

During the slowdown, credit declined on average about 1 percentage point of GDP between the peak of the cycle reached in September 2008 and the trough, the timing varying by country. The most significant declines occurred in El Salvador, Guatemala, Honduras, and Uruguay. As of March 2010, credit in several Central American countries was still contracting. Despite their low weight in total credit, crossborder flows explain a significant part of the credit decline in these countries (Figure 3.7).

Shifts in Funding and Asset Structure

In the upswing, banks funded their credit expansion with an increase in the domestic deposit base, but also by intermediating funds from the government and nonresidents into firms and households. By saving part of the fiscal revenue boom and having access to bond financing, central governments reduced their net credit from banks, making more funds available for bank lending to firms and households. In some countries, banks also experienced a moderate increase in liabilities to nonresidents, part of which they lent to firms and households.

In some countries, banks also lent to firms and households funds received from other nonbank financial institutions, including pension funds (Figure 3.8).

During the slowdown, a sharp deceleration in the growth rate of bank liabilities—including to a lesser extent domestic deposits—took place. Credit to firms and households decelerated even more as balance sheet trends seen during the upswing reversed direction. In this phase, banks ended up transferring (net) funds to nonresidents (deleveraging their exposure abroad or building up foreign assets) and to the government (Figure 3.9). Ultimately, the crisis showed that domestic deposits were the most stable source of bank funding, while nonresidents were the most volatile liabilities. In general, financial institutions were not a source of instability, as bank soundness was preserved and the supply of credit did not collapse as it did in the late 1980s and early 1990s. The slowdown in credit growth, however, did affect the most leveraged firms (Box 3.3).

Part of the reduction of credit reflects a flight-to-quality asset management response by banks. This is captured by the increase in the share of the government and central bank securities in the asset side of the bank’s balance sheet. The economic slowdown increased uncertainty on loan performance, which led to tighter credit standards. Thus banks shifted from riskier loan portfolios toward safer and more profitable securities (for example, government bonds).⁵ The dry-up of liquidity in global markets led to a partial selloff of foreign investors’ fixed income portfolio holdings in emerging markets, including Latin America (Figure 3.10).

In the large economies of the region, bond and other securities issuance provided banks with an alternative source of funding. This was possible because, unlike in past episodes, capital flight did not take place. On the contrary, some domestic agents such as pension funds repatriated capital. During 2009 and 2010, several banks in the region issued bonds to support their working capital and sustain credit growth (for example, BBVA Colombia, BBVA Bancomer, Banco Santander Peru, Scotia Bank Peru). Long-term bonds provided banks with a stable source of funding, unscathed by the crisis.

Box 3.3.

Financial Structure and Corporate Performance during the Global Crisis: Microevidence for Latin America

In Latin America, more leveraged firms suffered more during the global financial crisis. This is what results from an econometric exercise based on balance sheet data for all nonfinancial publicly traded firms of Argentina, Brazil, Chile, Colombia, Mexico, and Peru.

Using difference-in-difference cross-sectional regression techniques, Kamil and Sengupta (forthcoming) estimate the effect on real sales between 2008:Q3 and 2009:Q2 of financial indicators of firms as they stood before the crisis. In particular, they explore for a systematic relationship between differences in sales growth across firms and their (a) capital structure (share of short-term debt, share of dollar-denominated debt, and leverage); (b) dependence on bank credit; (c) cash holdings; and (d) export orientation. The study controls for other factors that may affect firms’ real economic activity, such as firm size, access to international capital markets, and country and sector-specific effects.

The study finds that higher leverage and more rapid bank credit growth were the main financial factors explaining the drop in sales following the Lehman events (see table).¹ The degree of dollarization—a source of vulnerability during previous crises—did not seem to play a key role. This is consistent with the findings of previous Regional Economic Outlooks (REOs) that Latin American firms are now less exposed to exchange rate risk arising from currency mismatches in their balance sheets. In addition, the study finds that large cash buffers helped mute the impact of financial frictions at the peak of the crisis. Exporting firms seem to have been disproportionately hit, as might be expected given that global trade was strongly disrupted during this crisis.

Effect of Predetermined Financial Conditions on Firms’ Postcrisis Sales Performance^1,2,3

Sources: Kamil and Sengupta, forthcoming.

^¹Based on cross-sectional estimations using firm-level data for Argentina, Brazil, Chile, Colombia, Mexico, and Peru. Predetermined refers to values as of 2007:Q4 and postcrisis refers to the time period between 2008:Q3 and 2009:Q2.

^²Leverage is defined as the ratio of total liabilities to total assets. Share of bank debt is defined as ratio of total bank debt to total liabilities.

^³Country and Industry dummies as well as dummies controlling for firm size are included in the estimations, but not reported. ***, ** significant at the 1 percent and 5 percent levels respectively.

⁴The liquidity cut-off is a cash to asset ratio of 9.6 percent-the average across firms of all countries in the sample.

Effect of Predetermined Financial Conditions on Firms’ Postcrisis Sales Performance^1,2,3

Explanatory Variables			Differential Effects
Explanatory Variables	Average Effects		Low-Liquidity Firms		High-Liquidity Firms
Growth in share of bank debt	-0.11	**	-0.08	*	-0.14	*
Share of short-term bank debt in total bank debt	-0.04		-0.06	**	-0.04
Share of dollar debt in total debt	-0.03		-0.02		-0.01
Observations	481		321		160

Sources: Kamil and Sengupta, forthcoming.

^²Leverage is defined as the ratio of total liabilities to total assets. Share of bank debt is defined as ratio of total bank debt to total liabilities.

^³Country and Industry dummies as well as dummies controlling for firm size are included in the estimations, but not reported. ***, ** significant at the 1 percent and 5 percent levels respectively.

⁴The liquidity cut-off is a cash to asset ratio of 9.6 percent-the average across firms of all countries in the sample.

Effect of Predetermined Financial Conditions on Firms’ Postcrisis Sales Performance^1,2,3

Explanatory Variables			Differential Effects
Explanatory Variables	Average Effects		Low-Liquidity Firms		High-Liquidity Firms
Growth in share of bank debt	-0.11	**	-0.08	*	-0.14	*
Share of short-term bank debt in total bank debt	-0.04		-0.06	**	-0.04
Share of dollar debt in total debt	-0.03		-0.02		-0.01
Observations	481		321		160

Sources: Kamil and Sengupta, forthcoming.

^²Leverage is defined as the ratio of total liabilities to total assets. Share of bank debt is defined as ratio of total bank debt to total liabilities.

^³Country and Industry dummies as well as dummies controlling for firm size are included in the estimations, but not reported. ***, ** significant at the 1 percent and 5 percent levels respectively.

⁴The liquidity cut-off is a cash to asset ratio of 9.6 percent-the average across firms of all countries in the sample.

Note: This box was prepared by Herman Kamil and Rajeswari Sengupta.¹ The median leverage (precrisis liability to asset ratio) across all firms in the sample is about 47 percent, with Brazil leading the group (55 percent) and Chile, Colombia, and Peru below the median.

Differences in Bank Behavior: Foreign and Public Banks

Local domestic banks proved to be a more stable source of funding, as foreign banks displayed more pronounced swings around the cycle. In particular, credit from foreign-owned banks that report to the Bank of International Settlements (BIS) declined much faster than that extended by domestic-owned banks (Figure 3.11).⁶ Looking closer, crossborder flows contracted markedly; credit from branches and subsidiaries of foreign banks funded locally proved to be less volatile, but also slowed. Available data for Brazil, Chile, Colombia, Costa Rica, Peru, and Uruguay show that foreign branches and subsidiaries usually slowed their credit earlier, and often by more, than did domestically owned banks. This pattern is consistent with other findings that global banks are usually more sensitive to changes in worldwide financial conditions than domestic banks.⁷ By contrast, on average, some public banks played a modest discretionary countercyclical role during the slowdown. They either expanded credit in real terms or reduced its growth rate at a much slower pace than domestic private and foreign banks. This was especially significant in Brazil, Chile, and Costa Rica (Box 3.4), but also noticeable in Argentina, El Salvador, and Mexico.⁸ This countercyclical role was possible, in part, because these countries had fiscal space.

Box 3.4.

Public Banks as a Countercyclical Tool: Experience in Brazil, Chile, and Costa Rica

In a number of countries, including Brazil, Chile, and Costa Rica, public banks played an important role in offsetting the drop in lending from private banks (both domestic and foreign). The impact of the measures depended in part on the initial share of public banks in overall credit.

In Brazil, the deceleration in real credit growth to the private sector following the Lehman event was much smaller in public financial institutions than in private and foreign ones. In particular, real credit growth by public institutions remained above a 20 percent rate, while foreign institutions actually contracted their real credit in 2009. This behavior was an explicit component of the countercyclical policy response in the country, heavily implemented through the state bank BNDES (Barbosa, 2010). The significant impact of the measures partly reflects that public banks are still very large in Brazil, managing about 15 percent of GDP in credit to the private sector in 2009. The share of public financial institutions in the private sector credit market had declined to 35 percent by late 2008 from 60 percent in the early 1990s. After the recent expansion, its share has increased to about 40 percent.

In turn, Banco Estado de Chile also increased heavily its credit to the private sector following the Lehman events. Its effect on total domestic bank credit to the private sector, however, was lower than in Brazil because its market share is lower (about 10 percent). Yet, its macroeconomic impact was not trivial because financial depth is larger in Chile than in Brazil. In 2009, Banco Estado managed about 11 percent of GDP in credit to the private sector.

In Costa Rica, public banks played a similar role. Given high dollarization in private banks, real credit increased on impact in the last quarter of 2008 with the currency depreciation, an effect that tended to erode over time. The public bank kept its real credit growth stable for almost a year, while that in private banks declined rapidly. Public banks manage about 25 percent of GDP in credit to the private sector, about half the total credit to firms and households.

Note: This box was prepared by Jorge Iván Canales-Kriljenko.

Asset Quality

The shock that hit the banking system during the global crisis was not as damaging for the financial system as in the past. The swings in exchange rates, commodity prices (affecting commodity exporters, benefiting commodity importers), and economic activity, in some cases significant, were short lived. Moreover, adverse effects were partly offset by the decline in international interest rates.

Standard financial indicators weakened but banks in the region remained sound. The global crisis and resulting recession moderately lowered their profitability, although not considerably (Figure 3.12). This was associated with the reduction in the loan portfolio (which typically earns higher returns than liquid assets) and the increase in nonperforming loans (which doubled from low levels), thus triggering the need for higher provisions. The slight decline in profitability is consistent with the general view that banking systems in the region did not have a meaningful exposure to the toxic assets that disrupted financial systems in advanced economies, at least in systemic amounts. Capital adequacy ratios remained high.

Market estimates of probabilities of bank default (as measured by Moody’s expected default frequency—EDF—and defined as failure to make a scheduled payment) remain low but are higher than before the crisis. After the Lehman event, EDFs increased significantly, especially in Brazil. More recently, bank strength in Venezuela has continued to weaken (Figure 3.13). The EDFs often move with the VIX, which suggests that they reflect not only bank-specific developments but also changes in risk aversion over time. Another piece of evidence of the relative resilience of profitability of regional banks was the behavior of bank-market valuations. Unlike in advanced economies at the center of the crisis, bank equity prices remained stable relative to other stock prices in each country (Figure 3.14).

A Complementary Financing Role of Bond Markets

Despite the tightening of credit conditions, major players (for example, governments and large firms), in the absence of capital flight, were able to find support in the expanding local bond markets. When needed, governments in the region were able to issue, rollover, and swap debt in domestic bond markets (for example, Brazil, Colombia, Mexico, and Peru) (Box 3.2). Certainly, it also helped that government finances were not overstretched because their financing needs had already been met, thanks to their progress over the years in debt management and reducing fiscal deficits.

Domestic corporate bond markets played some limited role. Their expansion between 2003 and 2008 had been limited and were often only accessible to top tier companies (financials or conglomerates). During the contraction in economic activity, the outstanding stock of bonds issued abroad declined, which was offset by bonds issued domestically.

Policy Implications for the Ongoing Cycle

As reviewed above, banking systems in the region fared the global crisis relatively well. However, having successfully passed the recent “real live stress test” should not give rise to complacency. This is particularly relevant as the new financial cycle begins.

In some countries, credit growth is already under way. The pace of recovery has varied, with strong growth in South America. Domestic bank lending in local currency has so far led the way (Figure 3.15). Taken together, the last credit cycle increased real credit to firms and firms and households, despite a temporary and relatively small reversal (Figure 3.16).

Looking forward, with external financial conditions now again conducive to a credit expansion (low interest rates and increasing risk tolerance), the risk of excess has to be taken seriously.

What lessons does the experience with the last financial cycle leave for managing the new financial cycle?

Using Macroprudential Regulation to Smooth out Financial Cycles

Authorities should seek to smooth out the financial cycle by avoiding excessive cyclical swings in private credit and excessive risk taking behavior of banks. Furthermore, authorities need to take into account the complex interaction between the financial system and the economy at large to preserve financial stability. All this calls for a broader view of macroeconomic stability, complementing countercyclical macroeconomic policies with macroprudential measures. (This theme is explored in Chapter 4.)

Keys to Financial Deepening in a Safe Manner

The desirability of increasing financial depth cannot justify unsound credit growth. Many countries around the world have been able to increase their credit-to-GDP ratio. But doing this too quickly can put the financial sector in a vulnerable position, which increases the probability of a financial collapse. The key to success appears to hinge on a stable source of funding and a well-grounded capacity to manage bank risks on the asset side.

The most secure form of funding is likely to be the resident deposit base. Increasing financial intermediation through domestic deposit growth, however, is a longer-term process. The level of domestic deposits critically depends on macroeconomic stability, but also on the legal, regulatory, and judicial systems. Indeed, countries perceived to have stronger respect for the rule of law usually have higher domestic deposit bases (Figure 3.17). Therefore, improvements in this area may help attract funds that Latin American residents now hold abroad, while keeping financial savings at home and fostering intermediation. For the sake of stability, policies should encourage banks to limit reliance on volatile forms of funding, including nonresident deposits, especially short-term, and wholesale deposits. Nonresident liabilities are typically small in Latin America (Figure 3.18). Authorities should also be careful that regulations or other policies do not encourage direct crossborder borrowing by firms, and have systems in place—as many countries already do—that allow authorities to monitor crossborder corporate debt. Although crossborder borrowing provides access to needed external savings, it can be a source of vulnerability because it tends to be significantly more volatile than domestic savings and feed currency mismatches.

The degree of safety of using market (that is, nondeposit) financing for bank credit to the private sector depends on its compatibility with the maturity (and currency) structure of banks’ balance sheets. Reliance on short-term instruments creates clear risks; longer-term bonds can give access to institutional investors (including those from other countries) and help banks more safely finance long-term investment and housing. This can complement the long-term lending that banks may provide through their traditional “maturity transformation” role (preferably from a base of deposits that is relatively stable).

With some exceptions, domestic banks in most Latin American countries fund their credit to the nonfinancial private sector out of deposits from resident firms and households. In particular, in most countries, the amount of these deposits exceeds the amount of credit granted to the private sector. The main exceptions are banks in Chile, Colombia, Costa Rica, and Paraguay, which also rely on securities (included in the IMF’s definition of broad money).⁹ These securities appear to be forms of domestic resident savings, which should be fairly stable over time. In Chile, two-thirds of those securities are held by domestic pension funds, and the other third by households. In Colombia, Costa Rica, and Paraguay, more than 80 percent are held by firms and households. (See also Spring 2009 Regional Economic Outlook, and Kamil and Rai, 2010.)

Promoting safe funding has become an important prudential issue in international fora. The Basel Committee on Banking Supervision has proposed a liquidity rule for Basel III requiring that banks fund less liquid assets with a more stable financing source. In practice, the net stable funding ratio sets the minimum in stable funds that a bank should hold given the liquidity structure of its assets. Banks would need to fund fully their lending to firms and households from stable sources. The rule defines stable funding as capital, preferred stock, and liabilities with remaining maturity exceeding one year, plus a share of demand and term deposits expected to remain with the institution even under conditions of stress. The Committee, however, does not propose implementation of this rule until 2018. Latin America would be well served by incorporating as soon as possible the main elements of Basel III, especially as the new credit cycle builds momentum.

Along similar lines, New Zealand’s central bank introduced in 2009 a quantitative limit on banks’ short-term and offshore funding. The authorities plan to tighten this limit over the next few years. They currently envisage that the “core-funding” requirement of 65 percent of total loans and advances will increase to 75 percent by 2012. Core funds are defined to include funding that is stable because it has at least one year maturity or because it is from sources that are not likely to pull out their money quickly (in contrast to short-term offshore funding, for example). (See Hoskin and others, 2009.)

Increasing Crowding in, where Possible

One route to increasing credit availability, even with the existing deposit base, is to reduce the share of bank credit that goes to the government and central bank. In Argentina, Brazil, and Mexico, for example, the central government accounts for about one-fourth of total bank assets. Claims on central banks, including required reserves and central bank securities, also appear prominently in banks’ balance sheets in many countries in the region.¹⁰ Of course, holdings of central bank and government paper do provide interest-bearing liquidity, the need for which is country-specific. (In some cases, the size of bank holdings of such paper may also be a reflection of the scarcity of safe and profitable investment opportunities.¹¹)

The largest countries in the region have successfully reduced public debt and improved its composition. Significant room for improvement, however, still remains. Besides the obvious fiscal consolidation policies to further reduce public debt levels, it is also possible to reduce their share in bank balance sheets by active public debt management that secures alternative sources of credit and instruments for government financing. These alternative sources can include foreign funding, which will be macroeconomically safer if denominated in domestic currency. This may lead to the further development of instruments and markets for hedging the exchange rate risk taken by foreign investors.

Foreign and Public Banks

Besides the usual roles that foreign and public banks play in the banking system, the recent credit cycle highlighted another dimension associated with their cyclical behavior.¹² During the recent experience, foreign banks tended to amplify the effect of external financial conditions. On the other hand, some public banks played a seemingly useful countercyclical role. This aspect needs to be taken into account in making monetary and banking policy decisions.

Decisions about foreign bank market entry and the existence or scope of public banks should carefully consider their pros and cons, and be consistent with the country’s obligations under the General Agreement on Trade in Services (GATS) or other trade arrangements. Foreign banks usually bring technology, innovation, and efficiency. The literature has also documented that foreign banks have played a stabilizing role during episodes of homegrown financial stress.

Although public banks may not be best suited for bank asset allocation, they may have a role to play in providing some social and financial services, including creating markets that would otherwise not develop.

The seemingly helpful countercyclical role of some public banks during the recent global crisis needs to be viewed in perspective. First, to operate in a truly countercyclical manner, they would have to significantly slow down (or contract) credit growth during boom episodes, something that needs to be seen. Fiscal risks of public bank lending are always present, and strong corporate governance is needed to keep them under control. Political pressure for subsidies and direct credit allocation often build up. While lending by public banks can play a useful countercyclical role in rare, extreme events originated abroad such as the global financial crisis, in general, they are not the best countercyclical tool.¹³ During normal cyclical fluctuations, monetary policy is more flexible and cost effective.

Strengthening Other Financial Markets

The global crisis has highlighted that financial markets are strongly interconnected and that failures in any one of them can have adverse effects in other markets, magnifying their macroeconomic impact. A lesson has been that the perimeter of regulation should cover any financial institution or market that could have systemic implications. Financial oversight should be comprehensive and consolidated as much as possible.

Nonbank financial markets have important benefits although their development can bring tradeoffs and new risks. For example, securities and derivative markets can increase welfare and financial stability by helping individuals in an economy better manage and share risk. They can also help channel long-term savings directly into productive ventures, while providing liquidity to investors by allowing their negotiation in the market. Financial instruments for managing and transforming risk also help attract foreign investors to domestic currency assets by allowing them to hedge the currency risk. These features can result in deeper and more stable markets that help reduce currency and maturity mismatches, improving the capacity of the economies to absorb large and sustained capital inflows. This is possible by helping channel capital inflows into long-term domestic currency assets, rather than into short-term domestic foreign-currency ones.

Yet such markets can also be a source of new risk. More sophisticated and complex securities and derivative markets often make it harder to identify who is bearing the ultimate risk, which under certain circumstances can amplify stress and increase the challenges involved in policy decision making. As the global crisis has shown, this can be a serious problem if the bearers of risk are systemic financial institutions or nonresidents who leave the market in a hurry. Thus, the quality of securities and derivatives that make it into banks’ balance sheets or create off-balance sheet exposures need to be closely monitored and understood by supervisory authorities.

During the recent global crisis, Latin America did not experience major macroeconomic stress arising from disruptions in these markets. The region entered the global financial crisis with securities and derivative markets present in only a handful of countries. Most of the existing domestic bond markets financed mainly the public sector, while the corporate bond market was small and mainly limited to large rated companies. Derivative markets and asset securitization were at their infancy.¹⁴ Although Latin America did not experience major banking problems originated in these markets, the misuse of foreign exchange derivatives transactions in the corporate sector created some stress in Mexico and Brazil after Lehman’s events.¹⁵

Based on the experience in advanced countries, authorities in Latin America should carefully oversee developments in nonbank financial markets to get a comprehensive view of financial stability. The challenge is to encourage their development while containing their risks. For this, authorities need to assess the degree of interconnectedness between financial institutions and other financial markets, with a view that more complex structures are usually riskier, or at least that they are more difficult to monitor, and so require more attention.

Final Remarks

Latin America made significant progress in strengthening its financial sector over the last decade. The broad challenge for the region is to continue increasing its financial depth in a safe manner, avoiding excessively rapid credit expansions that may end up in a financial crash. For this, macroeconomic stability needs to be preserved, prudential regulations enforced, and financial supervision further strengthened. Legal, regulatory, and judicial changes that improve creditor rights can increase the security of domestic deposits and bonds. Ultimately, these will expand credit opportunities in the economy. If properly managed and supervised, domestic securities and derivative markets can help deepen domestic financial intermediation and strengthen financial stability by allowing financial institutions and firms to better manage risk. In doing so they improve the capacity to absorb large and sustained capital flows. However, financial authorities need to closely oversee and supervise these markets because they may also be a source of risks. Supervisors need to understand how they affect the risk exposures taken by systemic financial institutions and to detect early when firms or financial institutions are using these markets to increase, rather than decrease, their risk profiles. In the immediate future, the regulatory challenge is to improve the set of prudential tools that would help better manage the incipient credit cycle.

Note: This chapter was prepared by Jorge Ivan Canales-Kriljenko and Camilo E. Tovar.

There is no evidence that banks in the region held the “toxic assets” associated with the global financial crisis, at least not in significant amounts.

The causal relationship between economic activity and credit probably goes both ways, although this is ultimately an empirical issue that is difficult to disentangle precisely. In this chapter, we assume that expansions and contractions of credit have a causal effect on economic activity, at least to some extent. This does not preclude the existence of a reverse effect, from activity to credit.

Changes in credit-to-GDP ratios are computed from the evolution of real credit and GDP. Moreover, the real credit figure controls for real exchange rate fluctuations. Base ratios are those for December 2007.

The expansion did not qualify as a full “credit boom” under the methods used in the literature (Mendoza and Terrones, 2008).

In some large economies, banks acquired government bonds that foreign investors left behind.

Foreign banks that do not report to the BIS International Banking Statistics are important in Central America, where local banks operate in several countries in the region.

Looking back over a 12-year period, Galindo, Izquierdo, and Rojas-Suarez (2010) find that foreign-owned banks operating in Latin America—with the notable exception of Spanish-owned banks—have tended to magnify the impact of changes in risk conditions in the international capital markets on real credit growth.

While public banks are found in many countries of the region, only in some countries—including Argentina, Brazil, Chile, Costa Rica, Dominican Republic, Uruguay, and Venezuela—do they account for more than 10 percent of system assets.

These negotiable instruments include certificates of deposit issued by depository corporations (see http://www.imf.org/external/pubs/ft/mfs/manual/index.htm)

This partly reflects a large stock of central bank debt that needs to be held by some party, arising from a history of losses in central bank operations (including on sterilized foreign exchange operations) and from bank rescue operations associated with past banking crises.

Such scarcity may reflect deficiencies in laws and frameworks for resolving disputes that make it hard to enforce credit contracts or seize collateral. It may also reflect infrastructure shortcomings that make it difficult to assess borrower creditworthiness, for example, where credit reporting and bureaus are in their infancy.

For an earlier discussion, see IADB (2005).

When the adverse shock has a fiscal origin, public banks may not have space to play countercyclical role.

For an overview of securitization in Latin America before the crisis, see Scatigna and Tovar (2007).

Nonlinear bets on the volatility of the exchange rate by a number of companies led to large losses when the exchange rate depreciated in October 2008. The existence of these positions had been generally unknown to creditors and shareholders, and perhaps even to some senior company management. (See May 2009 Regional Economic Outlook, Box 2.2, and Jara and others, 2009.)

Topics

Countries

Series

About

Resources

Abstract

An Overview of the Financial System

Learning from the Last Financial Cycle

How Big Was the Recent Cycle?

Shifts in Funding and Asset Structure

Differences in Bank Behavior: Foreign and Public Banks

Asset Quality

A Complementary Financing Role of Bond Markets

Policy Implications for the Ongoing Cycle

Using Macroprudential Regulation to Smooth out Financial Cycles

Keys to Financial Deepening in a Safe Manner

Increasing Crowding in, where Possible

Foreign and Public Banks

Strengthening Other Financial Markets

Final Remarks

Within Same Series

Other IMF Content

Other Publishers Content

Asian Development Bank

Inter-American Development Bank

International Labour Organization

Nordic Council of Ministers

The World Bank

Share Link

Table of Contents

Sections

Figures

References