Financial integration and the resulting interconnections among financial and nonfinancial institutions provide benefits and risks for countries. As articulated in earlier chapters in this book, financial integration can bring important benefits to both banks and clients, including lower funding costs, risk diversification, deeper liquid markets, increased competition, and efficiency in the financial system. At the same time, with conglomerates operating in multiple jurisdictions in the LA-7 countries (Brazil, Chile, Colombia, Mexico, Panama, Peru, and Uruguay) and across the Central American region, country-specific weaknesses in regulatory and supervisory frameworks or insufficient regional coordination may allow regulatory arbitrage. Moreover, financial integration can increase spillover risks and lead to contagion across the region in the event of a crisis.

Financial integration and the resulting interconnections among financial and nonfinancial institutions provide benefits and risks for countries. As articulated in earlier chapters in this book, financial integration can bring important benefits to both banks and clients, including lower funding costs, risk diversification, deeper liquid markets, increased competition, and efficiency in the financial system. At the same time, with conglomerates operating in multiple jurisdictions in the LA-7 countries (Brazil, Chile, Colombia, Mexico, Panama, Peru, and Uruguay) and across the Central American region, country-specific weaknesses in regulatory and supervisory frameworks or insufficient regional coordination may allow regulatory arbitrage. Moreover, financial integration can increase spillover risks and lead to contagion across the region in the event of a crisis.

The focus in this chapter is on the risks related to financial integration and the prudential tools required to mitigate them. The chapter uses market-based tools to assess current levels of systemic risk in LA-7 countries and the potential for contagion. It provides an overview of the importance of network analysis in mapping financial interconnections and how they add to systemic risk and the potential for contagion. The authors then undertake an overview of the importance of robust regulatory and supervisory frameworks and of consistency of prudential standards in the LA-7 to address the risks inherent in financial integration. Mapping the financial networks that have been created by conglomerates across the LA-7 countries illustrates the challenges faced by regulatory and supervisory frameworks and the importance of consolidated and conglomerate supervision. The authors suggest prudential tools (both micro and macro) that can be used to address the risks of interconnectedness and the possibility of contagion that are part of financial integration. The last section includes recommendations for pursuing financial integration within a robust framework that address its risks.

Current State of Affairs

Systemic (or spillover) risk arises when the failure or weakness of one or more financial institutions or infrastructures disrupts financial services and imposes costs on the economy as a whole. The failure or weakness of multiple financial institutions may arise through a variety of direct and indirect mechanisms (Table 9.1).1 Direct bilateral exposures across institutions are the most direct transmission mechanisms of shocks within a financial network. However, indirect linkages may arise from exposure to common risk factors, such as the adoption of similar business models, common accounting practices across financial institutions, the market perception of financial institutions’ coincidence of fortunes, fire sales, and informational contagion. These risk factors can be as important as direct exposures.2

Table 9.1

Examples of Direct and Indirect Financial Interconnectedness

article image
Source: Depository Trust and Clearing Corporation (2015).

Contagion spillover risks in the LA-7 countries are currently at a low level. Although growing cross-border activity and financial integration increase the potential for contagion spillovers in a crisis, some quantitative analysis suggests that spillover risks among Latin American financial systems are currently contained (see the next section). The conglomerates operating in LA-7 jurisdictions and Latin America in general can act as pathways for increased regional banking and financial sector connectivity through their network of subsidiaries and intergroup and other counterparty exposures. Country authorities have begun to focus on these potential risks, in some cases limiting cross-border activities by imposing formal or informal restrictions. The current low levels of spillover risk leave room for further financial integration while regional supervisory and regulatory oversight is strengthened and any country-specific weaknesses in supervisory frameworks are addressed to avoid regulatory arbitrage. Contagion risks will naturally rise with greater integration or under adverse conditions. This potential requires that cross-border activity and exposures be monitored in a regionally coordinated manner.

Quantifying Market-Implied Spillover Risks

Market-based spillover analysis suggests that contagion risks among large financial institutions in Latin America are contained. This quantitative analysis looks at the market-based interlinkages of large financial institutions in six Latin American countries. In particular, the analysis quantifies potential spillovers across institutions through financial markets using data on traded securities. Overall, the relatively limited market-implied spillovers are in line with the current relatively low levels of cross-border balance sheet exposures within the region.

Financial linkages among financial or banking institutions can be broadly split into two categories: direct and indirect. Direct financial linkages denote explicit balance sheet positions from one financial institution onto another; essentially, assets or liabilities of financial institutions vis-à-vis each other. Indirect linkages arise when there are no explicit direct linkages among financial institutions, but their market indicators (for instance, stock prices or credit default swap spreads) tend to exhibit some degree of comovement or synchronicity. These indirect linkages could be the consequence of having similar business models or common exposures to related economic sectors, or simply of being perceived by the markets as being vulnerable to the same type of shocks (for example, a change in legislation affecting most banks in one country).

The aim of the market-implied interlinkage analysis is to quantify both direct and indirect linkages among financial institutions in Latin America. The sample includes the largest listed banks from Argentina, Brazil, Chile, Colombia, Mexico, and Peru; it covers the period 2005–16 and relies on publicly available daily time series of financial variables (Table 9.2).

Table 9.2

Latin American Banks Included in the Spillover Analysis

article image
Source: IMF staff compilation.

The methodology relies on the computation of empirical distributions characterizing the joint and conditional probabilities of distress among financial institutions. This is largely based on Segoviano’s (2006) Consistent Information Multivariate Density Optimizing Methodology.3 In particular, two synthetic measures are used:

  1. The vulnerability index (VI), which measures the susceptibility of a particular institution to fall in distress given distress in other financial institutions (loosely speaking, it measures an institution’s vulnerability to contagion from other financial institutions). Algebraically, the VI of a given financial institution “i” is given by:


    where the weight αj=1NP(Aj), N denotes the number of financial institutions in the sample, and P(Aj) is the probability that institution j falls in distress.4

  2. The contribution to systemic risk, which measures the contribution of a given institution to changes in the vulnerability to contagion of other institutions (that is, its role as a source of contagion). In other words, it is the percent share that a given financial institution represents in the changes in the vulnerability index of all other institutions in the sample.

To analyze the dynamics of these two market-based measures, both are computed for a sample of 15 banks from six Latin American countries.5 Figure 9.1 shows the evolution of the vulnerability index for the 15 banks (panels 1, 2, and 3). The figure also shows the percentage contribution of each of these banks to the rest of the system’s change in vulnerability during three selected periods (panel 4). Argentinean banks and, to a lesser extent, Banorte (Mexico) appear to have been the most vulnerable to contagion during the global financial crisis (Period I). However, the relatively high market-implied interlinkages during that period appear to be important mainly among themselves (Table 9.3).

Figure 9.1
Figure 9.1

Market-Based Interlinkages in Selected Latin American Banks

Sources: Moody’s KMV; Thompson Reuters Datastream; and IMF staff calculations.Note: See Table 9.2 for bank abbreviations used in this figure.
Table 9.3

Contribution to Systemic Risk during the Global Financial Crisis

article image
Sources: Moody’s KMV; Thompson Reuters Datastream; and IMF staff calculations.Note: Data are for the period October 31, 2007, to March 2, 2009. See Table 9.2 for bank abbreviations used in this table.

In the most recent period (Period III), Brazilian banks appear to be important drivers of the market-implied contagion (Figure 9.1, panel 4). However, in terms of absolute magnitude, the actual spillovers and VI outside of Brazil appear to be rather small compared with the VI levels observed during the global financial crisis. In other words, large domestic public banks in Brazil might be very important for the domestic market in Brazil, but they are not so important for the rest of the region. These market-based measures seem to be in line with the limited actual cross-border balance sheet exposures of the banking sectors in Latin America.

Network Analysis and Conglomerate Connectivity

Financial integration brings the prospect of increased financial network connectivity that is already established across the LA-7 and the Central America region through the growth of financial and mixed conglomerates. Banks, particularly from Brazil and Colombia, are becoming regional institutions, with the whole area as their home base. Some of this expansion was linked to the exit of some global banks from the region. Cross-border participation in stock exchanges is also apparent, with the Brazilian stock exchange purchasing 8 percent of the Chilean exchange. Nonfinancial corporations—particularly retail institutions from Chile and conglomerates from Brazil and Mexico—are expanding across the region. Since 2011, the presidents of Chile, Colombia, Mexico, and Peru have been meeting regularly to advance the Pacific Alliance (PA; see Chapter 8). Mercosur has brought six Latin American economies together with the objective of integration; although the Mercosur process has recently lost momentum, conditions may be favorable for a revitalization of the financial integration process. Regional financial integration moves have helped many systemic banks—as part of even more systemic conglomerates—create increased networks through headquarters–subsidiary linkages as well as regional bank linkages to U.S. global banks.

Network Structures: Interconnectedness and Contagion Risks

Analysis of financial networks in Latin America can be a useful tool to capture interconnectedness and contagion (spillover) risks from financial integration. The expansion of systemic banks and financial and mixed conglomerates across Latin and Central America has led to increased direct and indirect financial interconnectedness, which could add to the propagation of shocks through interconnections in the LA-7 financial network. Risks within the LA-7 financial network also reflect the complex organizational structures and the risks inherent in the business models of the financial and mixed conglomerates in the network. A detailed mapping, monitoring, and analysis of the LA-7 financial network would allow supervisors and risk managers to determine: (1) at what point these financial networks become fragile; (2) how that fragility propagates through the network—through specific connections and through systemic conglomerates and other connected entities; (3) what type of financial network structures (topologies) are more fragile than others; and (4) how systemic risks in the network vary over time as a result of interconnectedness and contagion risks from greater regional integration in the LA-7. Network analysis using mathematical graph theoretical frameworks on data from various direct and indirect mechanisms of interconnections (see Table 9.1) can be an extremely useful complement to the analysis of systemic risks from market-based distress dependence models discussed earlier. Specifically, mapping, monitoring, and analysis of the LA-7 financial network provides a more detailed view of systemic risks from interconnectedness and contagion than the view from market-based models on their own.6

Financial and mixed conglomerate networks across the LA-7 countries are multilayered and complex, both domestically and across borders. Conglomerate financial networks in the LA-7 countries involve interconnections between systemic and nonsystemic financial and nonfinancial entities. Some of these interconnections are intraconglomerate—between parents (holding companies), subsidiaries, or branches—across multiple types of financial and nonfinancial institutions (multilayered). These networks not only extend across types of institutions but can also extend geographically on a cross-border basis, linked together through a multitude of direct and indirect connections (see Table 9.1). Using simple constructed examples, the authors show that network complexity can extend across multiple dimensions, moving from simpler domestic financial and mixed conglomerate networks to more complex multilayered cross-border networks (Figures 9.2 and 9.3). Domestic conglomerate networks (represented by the gray circle in Figure 9.2) are complex and challenging to supervise, but networks become much more complex and difficult to supervise when interconnections occur across borders (colored circles). This implies that systemic risk would be better estimated by mapping complex financial and mixed conglomerate networks across the LA-7 than by mapping simpler networks. As depicted in Figure 9.3, complex conglomerate networks have cross-border and multilayered dimensions that evolve dynamically over time. Thus, mapping, monitoring, and analysis of real-world complex financial and mixed conglomerate networks across LA-7 countries raises important supervisory, data, and coordination challenges for authorities, central banks, and supervisory agencies in the region.

Figure 9.2
Figure 9.2

Simplified Examples of Domestic and Cross-border Conglomerate Networks

Source: IMF staff.
Figure 9.3
Figure 9.3

Complex, Multilayered Cross-border Networks

Source: IMF staff.

Some Recent Network Approaches for Latin America

Systemic risk in the Mexican banking system between 2007 and 2013 is underestimated by 90 percent when the focus is on a single exposure network rather than multilayered exposure networks. Recent work by Poledna and others (2015) for the Mexican banking system has shown that analyzing systemic risk contributions from four exposure layers of the interbank network (derivatives, securities cross-holdings, foreign exchange, and interbank deposits and loans) underestimates systemic risk by 90 percent if only a single interbank deposits and loans network is considered. Looking at financial market measures and market-based models using the VIX (the Chicago Board Options Exchange Volatility Index), credit default swap (CDS) spreads, or other market prices might also underestimate systemic risk ex ante.

Colombian multilayered financial institutions and financial market infrastructure networks (FMIs) are better able to capture systemic risks and insulate against them. Work by León, Berndsen, and Renneboog (2014) showed that Colombian transactions and exposures among financial institutions that are registered, settled, cleared, and safeguarded by FMIs can be expressed as multi-layered networks. The analysis makes it clear that systemic risks would be much higher than those of each network in isolation. This result is due to consideration of cross-system risks. The study also suggests that integrating Colombian financial institutions’ networks of transactions and exposures with FMI networks adds resilience to the multilayered network, as FMIs are able to isolate feedback and cascades. However, this result holds only as long as the FMI itself is functioning well and does not fail. The evidence presented makes clear the advantages of mapping, monitoring, and analyzing the real-world multilayered financial networks.

Network analysis has been applied to other regions of the Western Hemisphere outside of LA7 countries. For instance, network analysis of the Caribbean region, in the context of Caribbean Regional Financial Project (CRFP), showed that the financial system of the Caribbean is highly interconnected. However, the system was found to be resilient to a range of moderate to severe propagating shocks (Annex 9.1).

Regulatory Oversight

A key precondition for substantial cross-border financial integration is having a robust and forward-looking best practice regulatory and supervisory framework in place. The global financial crisis highlighted the crucial need to enhance both domestic and international regulatory standards and to strengthen prudential requirements and develop macroprudential tools to reduce risks in the financial system, including cross-border risks. In Latin America, financial activity risks can be mitigated by having a suitable entry, operating, and resolution framework for cross-border institutions;7 sound prudential standards and national regulatory frameworks that reflect appropriate timelines and banking system complexity (Basel III); a complete picture of the financial institution (needed for cross-border consolidated and conglomerate supervision); and macroprudential policies that protect the national and regional financial systems from systemic macro-financial stability risks.

Cross-Border Establishments

Legal and regulatory frameworks for subsidiaries and branches of foreign firms have been enhanced but could be further strengthened. The legal frameworks of the surveyed countries generally require that branches and subsidiaries follow all regulations and practices of the host countries. All countries require endowment capital for branches, and most authorities have the power to restrict subsidiaries’ issuance of dividends and capital, including cross-border. In addition, the pricing of centralized functions such as IT and treasury is subject to oversight, thereby avoiding circumvention of the restrictions on dividend and capital transfers. However, there is room for improvement. Local asset maintenance requirements for branches and limits on intragroup exposures for subsidiaries could usefully be reviewed. In the context of a broader update of bank resolution frameworks, powers to deal with cross-border coordination should be strengthened, including by removing the automaticity and discriminatory features of ring-fencing mechanisms.

Basel III: Capital, Liquidity, and Leverage Requirements

The LA-7 countries have made progress in adopting the Basel standards (Figures 9.4, 9.5, and 9.6). The Basel Committee’s Eighth Progress Report on the adoption of the Basel regulatory agenda shows rapid recent progress in many emerging markets. Brazil and Mexico are fully compliant with the Basel III capital standard, the liquidity standard, and the leverage ratio. They also are compliant with regard to Basel II and Basel 2.5. Other LA-7 countries are implementing at a pace they consider in line with the complexity and development of their banking systems. With financial integration, many large banking groups and conglomerates face the prospect of varying regulatory capital regimes across the LA-7 countries in which they operate. This creates the potential for adverse incentives to grow in LA-7 countries where capital standards are weaker. A move to the Basel III capital definition across the LA-7 would remove the potential for regulatory arbitrage and—by coalescing on a consistent, high-quality capital measure—would enhance macro-financial stability by increasing loss absorbency against potential spillover risks (see Annex 9.2).

Figure 9.4
Figure 9.4

Progress in Implementing Basel II

Sources: Data for Brazil and Mexico are from the Basel Committee on Banking Supervision (2016); for other countries, from the Bank for International Settlements (2015).Note: AIRB = advanced internal ratings-based approach; AMA = advanced measurement approaches; BIA = basic indicator approach; FIRB = foundation internal ratings-based approach; P2 = Pillar 2; P3 = Pillar 3; SA = standardized approach; TSA = alternative standardized approach.
Figure 9.5
Figure 9.5

Progress in Implementing Basel 2.5

Sources: Data for Brazil and Mexico are from the Basel Committee on Banking Supervision (2016); for other countries, from the Bank for International Settlements (2015).Note: Mkt risk = revisions to the Basel II market risk framework; Rev P1 = revisions to Pillar 1; Rev P3 = revisions to Pillar 3; Suppl P2 = supplemental Pillar 2 guidance.
Figure 9.6
Figure 9.6

Progress in Implementing Basel III

Sources: Data for Brazil and Mexico are from the Basel Committee on Banking Supervision (2016); for other countries, from the Bank for International Settlements (2015).Note: C-Cycl = countercyclical buffer; Conserv = conservation buffer; Def cap = definition of capital; D-SIBs = domestic systemically important banks; G-SIBs = global systemically important banks; LCR = liquidity coverage ratio; Liq = liquidity; LR = leverage ratio; Risk cov = risk coverage.

Importance and Benefits of Consolidated and Conglomerate Supervision

Forward-looking domestic and cross-border consolidated and conglomerate risk-based supervision is a key supervisory approach to address the complexity and potential underestimation of systemic risks inherent in conglomerate networks in the Latin and Central American region. Many supervisory agencies in the LA-7 countries have moved to a forward-looking risk-based regulatory and supervisory framework that uses supervisory tools such as macroeconomic, multiyear credit, market, and liquidity stress tests to flag solvency (capital) and liquidity vulnerabilities of financial institutions. However, this approach does not fully account for systemic risks, and even the use of price-based models would underestimate systemic risks arising from financial and nonfinancial institutions’ interconnectedness and potential contagion risks. Adopting a domestic and cross-border consolidated and conglomerate supervisory approach would help prioritize the mapping, monitoring, and analysis of financial and mixed conglomerate networks, including through network stress tests as well as onsite and offsite work. This approach would help: (1) fully capture systemic risks, (2) identify vulnerable financial and nonfinancial institutions, (3) identify fragile transactions and exposures, and (4) flag key pathways for contagion. Knowledge of institution and system fragility would enable countries to tailor prudential and macroprudential tools to mitigate risks and ensure macro-financial stability.

The evidence from recent financial stability assessments conducted by the IMF and World Bank under the Financial Sector Assessment Program (FSAP) is that the countries in the region have a ways to go in improving their supervisory frameworks in terms of consolidated and conglomerate supervision. Legal restrictions in some countries prevent full achievement of best practices, in particular in the handling of the nonfinancial components of conglomerates. Although subsidiarization, and regulatory and resolution ring-fencing, can dampen cross-border spillovers, cross-border safety requires that the institution be supervised on a consolidated basis. International best practices for consolidated supervision call for establishing robust supervisory regimes, cross-border supervisory processes, joint monitoring programs, and coordinated corrective/supervisory actions among all parts of a cross-border financial institution or conglomerate.

The structure of Latin American financial institutions makes consolidated supervision particularly important because many are parts of conglomerates. In a number of cases the nonfinancial parts of these conglomerates have already expanded cross-border to a much greater extent than the banks. Even where there are no direct financial flows between the bank and nonbank parts of a conglomerate, problems in the nonbank part can have major knock-on effects on the bank. For example, problems in the retail arm of the group with the first Chilean bank to move cross-border led to financial pressures and ultimate sale to another regional bank, although the Chilean bank itself had faced no difficulties and was making substantial profits.

There is increasing awareness of the importance of consolidated supervision. Uruguayan regulators declined to license a regional bank that was seeking to acquire a bank being sold in Uruguay on the grounds that the regional bank’s supervisor was not conducting consolidated supervision.

Conglomerate supervision complements supervision of individual sectors by adding a layer to the solo and consolidated sectoral supervision. Individual supervision faces limitations in dealing with double gearing of capital, conflicts of interest, risks of contagion, concentration, and other specific group risks that may hamper financial stability. Conglomerate supervision should detect and monitor these risks while avoiding unnecessary duplication with sectoral prudential standards.

Internationally agreed upon documents provide national authorities with a set of principles that support consistent and effective supervision of financial conglomerates. The main references are the Basel Committee on Banking Supervision’s “Core Principles for Effective Banking Supervision” and the Joint Forum’s “Principles for Supervision of Financial Conglomerates.”8 Focusing on both the cross-border and cross-sector dimensions of the process, these principles set expectations for supervisory powers and responsibilities, corporate governance, prudential requirements, and risk management. They focus on closing regulatory gaps, eliminating supervisory blind spots, and ensuring effective supervision of risks arising from unregulated financial activities and entities. Colombia is currently seeking parliamentary approval for a bill to provide supervisors with powers over the holding companies of financial conglomerates. However, the law would provide only limited reach into mixed conglomerates; much of their activity would remain outside the scope of the financial supervisory and regulatory perimeter.

The Principles are flexible and take a nonprescriptive approach to the supervision of financial conglomerates in order to cover a wide range of structures. They emphasize the importance of recognizing structural complexity and the potential risks it poses. This includes risks arising from all entities—unregulated or regulated—that affect the financial conglomerate’s overall risk profile. The flexibility of this framework is intended to enable policymakers and supervisors to appropriately regulate and supervise financial conglomerates while limiting the scope for regulatory arbitrage.

Beyond consolidated supervision, there is also a need for increased cooperation and data sharing among supervisors to tackle conglomerate and cross-border risks more broadly. Supervisory colleges have been established for major banks in the region. Colombia has gone further with regard to Central America, where its banks have established significant positions in most countries. These countries have established a multilateral memorandum of understanding, a regional council of finance ministers, a regional monetary council, and a joint council of supervisors. The analytical work on cross-border multilevel financial networks highlights the need for more granular direct and indirect data for both on- and off-balance-sheet items (see Table 9.1). However, important data confidentiality concerns must be overcome on a cross-border basis, new analytical tools are needed to handle very large data sets, and investment is required in staff who can work with financial risk modeling tools.

Prudential and Macroprudential Measures

Prudential tools allied with forward-looking, risk-based supervision can address systemic risk arising from financial and mixed conglomerate networks across Latin America. The tools described here are designed to address interconnectedness and contagion (spillover) risks.9

  • Microprudential exposure limits. Microprudential exposure limits are normally designed as non-risk-sensitive backstops to limit concentration from a microprudential perspective, but they are also relevant from a macroprudential point of view.

  • Capital charges, particularly for systemically important financial institutions (SIFIs). Capital requirements in Basel I and II did not directly account for interconnectedness (concentration) and contagion risks, but Basel III minimum capital rules were formulated with explicit consideration for concentration and systemic risk.

  • Liquidity regulation and limits on liquidity mismatches. The lack of liquidity requirements encouraged reliance on central bank and wholesale funding in the global financial crisis, increasing interconnectedness and contagion risks among financial institutions. Basel III liquidity requirements (the Liquidity Coverage Ratio and Net Stable Funding Ratio) can mitigate systemic risks by limiting liquidity contagion and reducing funding and counterparty interconnectedness.

  • Clearing of over-the-counter derivatives on a central counterparty (CCP). CCPs increase transparency regarding the amount and distribution of risk exposure, helping clarify interconnectedness risks. They also help mitigate contagion by mutualizing losses among all clearing members.

  • Structural limits on activities. Direct restrictions on the scope of business of global SIFIs, conglomerates, or large and internationally active banks can help reduce the complexity of business models, which directly reduces interconnectedness and contagion risks.

  • Resolution frameworks and insurance. Effective cross-border resolution frameworks can help insulate institutions from each other with regard to failure, which reduces the impact from interconnectedness and contagion risks.

Arregui and others (2013) note that prudential measures can interact bilaterally, either complementing each other in reducing systemic risks or conflicting with one another. Depending on the degree of complementarity or conflict, we can make a qualitative assessment of how each prudential tool is likely to address interconnectedness (concentration) and contagion risks and how difficult it can be to implement from a regulatory or bank perspective (Table 9.4). Prudential tools may be less effective for mixed conglomerates than for financial conglomerates, because they may apply only to supervised and regulated entities within the mixed conglomerate structure.

Table 9.4

Interaction of Prudential Tools

article image
Source: IMF staff compilation.Note: CCP = central counterparty; LCR = liquidity coverage ratio; NFSR = net funding stable ratio; SIFIs = systemically important financial institutions.

Fully complementary: tool mutually reinforces the intended objective of another tool (dark blue); partially complementary: tool may have a partially offsetting impact on the intended effects of another tool (light blue); partially conflicting: tool has a conflicting but not major impact on the intent of another tool (light red); not related or not contradictory (white).

Arrows indicate an increase (↑) or decrease (↓) in resulting impact.

Refers to the Volker, Vickers, and Liikanen structural measures—see Pazarbasioglu and others (2013).

Since the global financial crisis there has been considerable global progress in designing and implementing a macroprudential toolkit. If they consider that cyclical conditions warrant, macroprudential authorities have the power to impose additional capital charges beyond those imposed institution by institution. Specific instruments—such as limits on loan-to-value and debt-to-income, and sectoral risk weights or floors—can be helpful.

Such instruments are likely to be designed and implemented at the national level, given the different risk exposures of each country. However, where cross-border financial activity exists, especially in the case of financial and mixed conglomerates, there is a clear need for coordination and reciprocity to avoid arbitrage and macroprudential leakage across countries. In addition to the complementarity and conflicts among prudential tools, there are complementarities and conflicts when prudential and macroprudential tools operate together to address systemic risks. Policymakers need to address this tension when they design policy frameworks and institutional mechanisms, to ensure effective coordination between these tools.

Policy Recommendations

Latin American countries can benefit from taking a regional approach to implementing the remaining elements of the global regulatory agenda and developing their emerging prudential and macroprudential toolkits to address systemic risks. They should do the following:

  • Align timelines while reflecting international commitments and local circumstances as national authorities move forward with implementing the regulatory agenda.10

  • Invest in staff and resources to fully capture systemic risks through the use of new data analysis tools and mapping, monitoring, and analyzing financial and mixed conglomerate networks that can be integrated with solvency and liquidity stress tests. Coordination across the LA-7 in sharing such information (legally enshrined, if possible) would help the region tackle appropriate systemic risk analysis and address data confidentiality concerns as well.

  • Introduce or strengthen consolidated supervision—with technical assistance from the IMF or other sources—in line with recommendations from the IMF FSAPs and the Joint Forum’s Principles.

  • Introduce or enhance conglomerate supervision, including establishing regulatory limits for intragroup exposures within banking groups and between bank and nonbank segments of conglomerates.

  • Continue the development of macroprudential tools through regional conferences and possibly a more formal regional arrangement, so that tools can be designed and implemented on a regional basis to avoid cross-country regulatory arbitrage and coordinate dealing with spillovers.

Annex 9.1. The Caribbean Regional Financial Project11

The Caribbean Regional Financial Project sought to map the structure of interconnectedness within the Caribbean financial system and, given that structure, to assess potential transmission of shocks through financial channels. The project entailed collecting a unique dataset on interconnectedness covering the banking systems of Barbados, Belize, the Eastern Caribbean Currency Union (ECCU), Guyana, Jamaica, Suriname, The Bahamas, and Trinidad and Tobago, as well as the insurance sectors of Barbados, Belize, Guyana, Jamaica, Suriname, and Trinidad and Tobago.12 After mapping the financial system, the analysis consisted of using the network structure to simulate how financial and economic shocks might propagate through the system. The aim was to assess the adequacy of Caribbean financial systems’ capital buffers to withstand such shocks and to understand which were the key regional financial sectors and clusters within the region.

The unique dataset collected suggests that the Caribbean financial sector is highly interconnected, both within the Caribbean and with the global financial system.13 Caribbean banks in particular, have important cross-border claims on sovereigns, banks, and insurers, representing, on average, 15 percent of GDP and 12 percent of total assets. In contrast to banks, Caribbean insurers’ cross-border claims are relatively small compared to the size of the economy, representing, on average, 5 percent of GDP, but are considerably larger as a share of their total assets than banks’ cross-border claims, suggesting a larger degree of interconnectedness in the insurance segment of the financial sector. Other key findings from the mapping exercise included the following:

  • The Caribbean financial sector’s exposure to cross-border sovereign risk appears to be concentrated primarily in the insurance sector, and within that sector, primarily to sovereigns from within the region. By contrast, banks’ cross-border claims on sovereigns represent a relatively small share of their total assets, and these claims are relatively diversified across the Caribbean and globally.

  • Caribbean banks are interconnected with other banks, particularly global banks, likely related to the significant presence of three Canadian banking groups in the Caribbean. Claims are primarily in the form of relatively low-risk deposits. Caribbean bank claims on banks domiciled in other Caribbean countries are relatively minor.

  • Insurers generally have limited interconnections with banks, with the exception of a few jurisdictions that are home to the largest regional insurers.

  • Caribbean banks’ exposures to cross-border risks from the insurance sector appear to be negligible, but Caribbean insurers appear to have important exposures to global insurers.

  • Network maps show that the most interconnected banking sectors in the Caribbean region are those in Barbados, The Bahamas, the ECCU, Guyana, Jamaica, and Trinidad and Tobago, while the key insurance sectors are in Barbados and Trinidad and Tobago (Annex Figure 9.1.1). The financial systems of Belize, Haiti, and Suriname appear to be on the periphery of the system.

Annex Figure 9.1.1
Annex Figure 9.1.1

Caribbean Financial Interconnectedness

Source: Country authorities; and IMF staff calculations.Note: Based on data as of June 30, 2013. BHS = The Bahamas; BLZ = Belize; BRB = Barbados; ECCU = Eastern Caribbean Currency Union; GUY = Guyana; HTI = Haiti; JAM = Jamaica; SUR = Suriname; TTO = Trinidad and Tobago.

The analysis of how financial and economic shocks could propagate through the Caribbean financial system suggested the system was resilient to a range of moderate to severe shocks. The simulations gauged Caribbean financial systems’ systemic importance and susceptibility to spillovers stemming from potential depletion of bank capital (solvency channel) and funding pressures (liquidity channel).14 Most banking systems were in a position to withstand substantial losses on total loans and on sectoral loans without falling below conventional capital adequacy ratio (CAR) thresholds. In general, propagation risks appear limited. Only one simulation, involving a simultaneous solvency and liquidity shock, was able to trigger a second-round (that is, indirect) spillover effect.15

The network was also exposed to simulations of both single-sector shocks (tourism, construction, real estate, households, energy sector, and the sovereign) and multisector shocks:

  • Several financial systems were exposed to sovereign risk. Simulations of simultaneous and uniform haircuts on Caribbean sovereigns pushed a number of jurisdictions’ banking systems below the Basel I/II minimum capital threshold. However, although the size of the simulated haircuts was based on historical experience with debt restructurings in the region, it should be noted that simultaneous haircuts on all sovereigns in the region would constitute a very extreme shock. All but one insurance sector were able to stay above the same Basel I/II minimum CAR threshold in such cases.16

  • Among the single-shock scenarios, shocks to real estate and household loans had the largest impact on banking systems’ capital adequacy. Substantial losses on real estate loans would have materially affected the capital adequacy of a few banking systems. However, these tests did not take into account collateralization of real estate lending, and partial recovery of such collateral would in fact mitigate the risks to bank capital.

  • Losses arising from single shocks to tourism, construction, and energy loans were relatively small as these sectors tend to be at least partially financed by foreign direct investment or by foreign banks.

  • Combinations of severe shocks to tourism, construction, and the sovereign would have significantly weakened the capital buffers of a number of national banking sectors. A moderate across-the-board haircut to sovereign debt, if combined with simultaneous losses on tourism and construction loans, did not push any banking sectors below the Basel I/II minimum capital adequacy threshold, although one banking sector came close.

  • Cross-border spillover risks among the Caribbean insurance sectors appeared largely contained owing to robust capital buffers.

  • Spillovers through solvency channels are more predominant than spillovers through liquidity channels. This appears to reflect that banks are primarily funded by deposits rather than interbank borrowing, reducing funding vulnerabilities.

Annex 9.2. Capital Definitions and Capital Ratios Across Latin America


A robust bank capital framework helps ensure financial stability and sustain bank lending during economic downturns. Bank provisions and profits are an important buffer; provisions in particular can absorb expected losses. However, in the event that losses exceed earnings, capital provides banks with a cushion to absorb unexpected losses to reduce the risk of bank failure and prevent interruption of banking services and financing to the real economy. Unfortunately, loss absorbency elements such as provisions, capital definitions, and actual regulatory capital levels are not easily comparable across Latin America, even using harmonized market-based measures.

Capital Definition and Adequacy

Capital definitions differ across Latin American countries, and comparisons must be made with utmost caution. Some cross-country differences in the computation of capital include treatment of the revaluation of fixed assets, accounting of profits from current or past accounting periods, treatment of investments in capital instruments or requirements on donated capital, treatment of some deductions from capital (goodwill, intangibles, and deferred tax assets), and grandfathering of some capital (debt) components. Capital also differs depending on the degree of consolidation undertaken, whether at the individual (solo) bank level, the banking group level, or even the financial conglomerate level. In different jurisdictions, the regulatory risk weights applied to the same asset classes can be quite different. Differences in the national definition of capital, the Basel framework in use, and accounting standards across Latin America suggest that any direct comparison of total regulatory capitalization should be interpreted with caution.

Market-Based Estimates of Capital

Some systemic Colombian banks have lower levels of capital in excess of the regulatory minimum than their regional peers. Regulatory capital requirements differ across Latin American countries (Annex Figure 9.2.1), with some higher than Colombia’s 9 percent (Brazil, 11 percent; Guatemala, Peru, and Uruguay, 10 percent) and some lower (Argentina and Chile, 8 percent; Mexico, 10.5 percent). The decision to choose a given level of national minimum regulatory capital reflects multiple factors, including supervisory judgment and discretion. The four largest banks in Colombia have lower levels of capital than the large banks in some other Latin American countries.

Annex Figure 9.2.1
Annex Figure 9.2.1

Regulatory Capital Requirement and Total Capital Above Requirement


Sources: Bankscope; company filings; IMF, Article IV Consultation staff reports, and Financial Sector Stability Assessments.

Total capital ratios in excess of the regulatory minimum requirement stood at 2.9 percent, the lower end of regional peer comparisons. Rating agencies have tried to obtain a more consistent, harmonized measure of capital across Latin America, but comparisons on quantity and quality of capital vary depending on the measures used. For example, Colombian banks have lower levels of capital according to Standard and Poor’s risk-adjusted capital measure, which deducts all goodwill on the balance sheet from banks’ total adjusted capital (Annex Figure 9.2.2). This measure is important, as Colombia experienced a large number of mergers and acquisitions following the financial crisis of the late 1990s that, together with the geographic expansion of the largest banks over the past few years, created large amounts of goodwill assets. Using the Fitch Core Capital (FCC) measure, Brazil, Chile, and Colombia have much lower capital levels, owing to higher leverage of the system, sizable investments in insurance companies, and high levels of goodwill and deferred tax assets, all of which are deducted from equity to reach FCC levels (Annex Figure 9.2.3).

Annex Figure 9.2.2
Annex Figure 9.2.2

Risk-Adjusted Capital Ratios for the Largest Rated Latin American Banks


Source: Standard & Poor’s.
Annex Figure 9.2.3
Annex Figure 9.2.3

Fitch Core Capital


Source: Fitch Ratings.Note: Data for 2013. Figure shows generally adequate capital ratios in Latin America.

Additional Loss Absorbency and Basel III

Although capital ratios on market-based measures may seem low for some Latin American countries, these countries have additional loss absorbency in their banking systems. Many banks hold high levels of provisions (Brazil and Colombia), lower levels of nonperforming loans (Colombia), or more conservative risk weights (Chile and Colombia). Many Latin American countries are adopting Basel III standards, albeit at different paces. This should help address inconsistency of capital definitions, with recent work by the Basel Committee ensuring harmonization of risk weights. Notwithstanding the move to Basel III, actual implementation may still result in differentiation of capital stemming from differences in the adoption of above-minimum levels (Pillar 2 and conservation, countercyclical, and domestic systemically important bank buffers). These differences may reflect the need to address supervisory failings and the desire to tailor capital to bank risks, which may be above Basel III voluntary minimums in some Latin American countries.


The addition of more consistency to the already robust capital framework across Latin America will help ensure financial stability and sustain bank lending during economic downturns. Challenges from moderating economic growth, a low yield environment, volatility around U.S. monetary policy normalization, cross-border risks, and conglomerate expansion require that Latin American banks adopt a more conservative long-term capital planning approach. Moving to Basel III should help, and this is attainable for most Latin American banking systems, as current capital is sufficient to support transition and additional loss absorbency exists beyond capital.


  • Arregui, N., M. Norat, A. Pancorbo, and J. Scarlata. 2013. “Addressing Interconnectedness: Concepts and Prudential Tools.Working Paper 13/199, International Monetary Fund, Washington.

    • Search Google Scholar
    • Export Citation
  • Bank for International Settlements. 2015. “Financial Stability Institute Survey: Basel II, 2.5 and III Implementation.Bank for International Settlements, Basel, July.

    • Search Google Scholar
    • Export Citation
  • Basel Committee on Banking Supervision (BCBS). 2016. “Tenth Progress Report on Adoption of the Basel Regulatory Framework.BCBS, Basel, April.

    • Search Google Scholar
    • Export Citation
  • Caceres, C., V. Guzzo, and A. Segoviano. 2010. “Sovereign Spreads: Global Risk Aversion, Contagion or Fundamentals?Working Paper 10/120, International Monetary Fund, Washington.

    • Search Google Scholar
    • Export Citation
  • Depository Trust and Clearing Corporation (DTCC). 2015. “Understanding Interconnectedness Risks to Build a More Resilient Financial System.White Paper, DTCC, October.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund (IMF). 2016. “Technical Assistance Report: Detailed Assessment of Observance of Key Attributes of Effective Resolution Regimes for Financial Institutions—Pilot of the Draft Assessment Methodology.Washington, April.

    • Search Google Scholar
    • Export Citation
  • León, C., R. Berndsen, and L. Renneboog. 2014. “Financial Stability and Interacting Networks of Financial Institutions and Market Infrastructures.Discussion Paper 2014-011, Tilburg, European Banking Center.

    • Search Google Scholar
    • Export Citation
  • Markose, Sheri M., and Simone Giansante. 2013. “Pigou Tax of Systemically Important Financial Intermediaries.” In Financial Networks: An Empirical Application of Systemic Risk Monitoring and Governance. Workshop: Twenty Years after Cadbury, Ten Years after Sarbanes-Oxley. Challenges of Corporate Governance June 24–25, Bath Management School, University of Bath, Bath, United Kingdom.

    • Search Google Scholar
    • Export Citation
  • Nier, E., J. Yang, T. Yorulmazer, and A. Alentorn. 2007. “Network Models and Financial Stability.Journal of Economic Dynamics and Control 31: 203360.

    • Search Google Scholar
    • Export Citation
  • Pazarbasioglu, C., J. Surti, A. Narain, M. Erbenova, and J. Chow. 2013. “Creating a Safer Financial System: Will the Volcker, Vicker, and Liikanen Structural Measures Help?Staff Discussion Note 13/4, International Monetary Fund, Washington.

    • Search Google Scholar
    • Export Citation
  • Poledna, S., J. L. Molina-Borboa, S. Martínez-Jaramillo, M. van der Leij, and S. Thurner. 2015. “The Multilayered Network Nature of Systemic Risk and Its Implications for the Costs of Financial Crises.Journal of Financial Stability 20: 7081.

    • Search Google Scholar
    • Export Citation
  • Scott, H. 2012. “Interconnectedness and Contagion.Discussion Paper, Committee on Capital Markets Regulation, Cambridge, Massachusetts, November.

    • Search Google Scholar
    • Export Citation
  • Segoviano, A., and C. Goodhart. 2009. “Banking Stability Measures.Working Paper 09/4, International Monetary Fund, Washington.

  • Segoviano, M. 2006. “Consistent Information Multivariate Density Optimizing Methodology.Discussion Paper 557, Financial Markets Group, London School of Economics, London, United Kingdom.

    • Search Google Scholar
    • Export Citation

According to Nier and others (2007), the failure or weakness of multiple financial institutions at the same time arises through four main mechanisms: (1) direct bilateral exposures between institutions, (2) correlated exposures of financial institutions to a common source of risk, (3) feedback effects from endogenous fire sale of assets by distressed institutions, and (4) informational contagion.


Scott (2012) and the Committee on Capital Markets Regulation argue that asset and liability interconnectedness were not the main drivers of the systemic risk concerns during the recent financial crisis in the United States but that contagion was front and center.


Extensions of the methodology are described in Segoviano and Goodhart (2009), and a description of the main quantitative indicators used in this analysis can be found in Caceres, Guzzo, and Segoviano (2010).


Marginal probabilities of distress for the different individual financial institutions in the sample are obtained from Moody’s KMV Expected Default Frequencies database.


The country coverage and the sample are determined by data availability and are not the same as in the rest of this study. In general, any financial institution included in the sample must be listed and actively traded on the stock market. This is also a requirement for individual domestic subsidiaries of foreign banks.


Markose and Giansante (2013) have argued that price-based models underestimate and are poorer at early warning of systemic risks compared to network balance sheet models, especially those based on granular balance-sheet and off-balance-sheet bilateral exposures.


The assessment of emerging markets in Latin America on the basis of the key attributes of effective resolution regimes for financial institutions (IMF 2016) is limited and will require further work. In late 2015 Colombia underwent a pilot assessment that will provide useful inputs to the Financial Stability Board from the emerging market perspective.


The Principles were released in 2012 by the Joint Forum’s parent committees: the Basel Committee on Banking Supervision, the International Organization of Securities Commissions, and the International Association of Insurance Supervisions.


See Arregui and others (2013) for more details.


As members of the Basel Committee and the Financial Stability Board, Brazil and Mexico are required to abide by the international timeline for implementation.


Prepared by Elie Canetti and Kimberly Beaton.


The data were provided to IMF staff at the aggregate banking and insurance system level rather than at the level of individual institutions for confidentiality reasons.


The data were collected in 2015, but in order to use audited accounts, were based on accounts as of June 30, 2013.


With aggregate data, it would be unrealistic to assume that spillovers occur only when the capital of the financial system is depleted. Aggregate capital would typically remain positive even during severe crises, but this would not preclude failures of individual institutions. Therefore, we modeled a systemic crisis as a capital shortfall (a capital adequacy ratio below 8 percent) rather than as an insolvency.


It is important to realize that these simulations can test shock propagation only through fundamental financial linkages. However, crises and spillovers can lead to purely panic-driven contagion across countries, even in the absence of fundamental linkages.


Although Basel I/II applies to banks, the same threshold was used for insurers for consistency. That said, the threshold treats insurers, in effect, somewhat more conservatively than banks because it measures capital adequacy in terms of total assets instead of risk-weighted assets.

A New Strategy for a New Normal