Abstract
This Selected Issues paper for Chile describes the postcrisis recovery experience. The recovery from the 2008–2009 global crisis has been markedly different both among advanced and emerging economies. The steady improvement in the labor wedge-distortions related to the consumption leisure decision helped support the recovery. In Chile, the growth generated by this improvement, was sufficient to overcome the relatively weak performance of efficiency (TFP). Chile’s recovery has been characterized by strong investment growth, 0.8 percentage points higher than the precrisis trend. The establishment of the Financial Stability Council in 2011 is an important step to ensure close coordination among the institutions involved in Chile’s financial prudential framework.
Systemic Risk Assessment and Mitigation in Chile1
Systemic risk assessment and mitigation tools are an integral part of a macroprudential policy framework. This chapter provides a quantitative framework for systemic risk monitoring in Chile and discusses the Chilean macroprudential toolkit.
A. Systemic Risk Assessment
1. This section provides a framework for systemic risk monitoring for Chile. It focuses on quantitative approaches to systemic risk. Two challenges should be acknowledged. First, in practice, these quantitative tools need to be complemented with qualitative assessments. Second, progress in quantitative risk assessments depends on data availability.
2. In monitoring systemic risk, it is important to consider both the time and the cross-sectional dimensions.
Time dimension. Risk is built up over the macroeconomic cycle with a procyclical bias, as financial institutions tend to take on excessive risks in the upswing of an economic cycle only to become overly risk-averse in a down-swing. This characteristic amplifies the boom and bust cycle.
Cross-sectional dimension. The growing complexity of the financial system is raising interconnectedness and common exposures conducive to rapid contagion risk when crises occur. Shocks are amplified and transmitted rapidly between financial institutions. Moreover, the failure of one systemically important financial institution can threaten the system as a whole.
3. This chapter focuses on the time dimension of systemic risk. The analysis focuses first on the behavior of credit aggregates, and is then complemented with other macroeconomic and financial variables. For an analysis of the cross-sectional dimension of systemic risk in Chile, see Chan-Lau 2009 and 2010 that focus on price-based and balance-sheet network analysis.
Single indicator (credit)
4. Economic activity and credit fluctuations are closely linked through wealth effects and the financial accelerator mechanism. In an upturn, better growth prospects improve borrower creditworthiness and collateral values. Lenders respond with an increased supply of credit and, sometimes, looser credit standards. More abundant credit allows for greater investment and consumption and further increases collateral values. In a downturn, the process is reversed. Theory has identified several channels that may lead to excessive risk taking during episodes of rapid credit growth.2 Such channels can explain why “the worst loans are made at the top of the business cycle” and justify policy intervention to prevent excessive risk taking during the boom. Also, the rapid growth of the loan base may mask an underlying deterioration of loan quality.
5. A large and growing literature identifies credit growth as a powerful predictor of financial crises. The empirical literature identifies credit booms as “significant” deviations above trend using different methodologies to compute the trend and different thresholds that determine a boom. Nonetheless, the finding that credit significantly above trend is a good predictor of financial crises is pretty robust across methodologies and thresholds. This chapter considers a variety of methodologies to measure credit conditions.
GFSR (2011) finds that increases in the credit-to-GDP ratio of more than 3 percentage points, year-on-year, is a good early warning signal one to two years before a financial crisis.
Mendoza and Terrones (2008) identify a credit boom when the deviation from the long–run trend in the logarithm of real credit per capita exceeds 1.75 times the standard deviation of the cyclical component. The long-run trend is calculated using the Hodrick-Prescott (HP) filter with the smoothing parameter set at 100, as is typical for annual data.
Borio-Lowe-Drehman (2002, 2009) conclude that among several variables, the “credit gap”, based on the credit-to-GDP ratio, is the most powerful indicator for banking crises. They estimate such a gap by extracting the trend (interpreted as the equilibrium credit-to GDP ratio) from the ratio using the HP filter with relatively high smoothing parameters (lambda equal to 1600 instead of 100 for annual data).
Dell’Ariccia et al (2012) define a credit gap measure as the percentage deviation of credit-to-GDP from a backward looking, rolling, cubic trend estimated over the period between t-10 and t. A credit boom occurs when the deviation from trend is greater than 1.5 times its standard deviation and the annual growth rate of the credit-to-GDP ratio exceeds 10 percent or when the annual growth rate of the credit-to-GDP ratio exceeds 20 percent.3
6. Bank credit growth in Chile has been strong in the last two years. The Chilean economy recovered rapidly from the global financial crisis and the February 2010 earthquake. From mid-2010 to June 2011, the central bank raised the policy rate from 0.5 percent to 5.25 percent, helping bring inflation expectations closer to the target. Credit growth strengthened during this period, in line with income. Nominal and real credit growth have been in double digits for most of the last two years.
Monetary policy and growth
(in percent)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: IFSMonetary policy and growth
(in percent)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: IFSMonetary policy and growth
(in percent)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: IFS7. However, bank credit growth has been below most thresholds of credit booms. Credit growth has not been “excessive” as defined by the Borio-Lowe-Drehmann, the Mendoza-Terrones, or the Dell’Ariccia et al. methodologies (Figure 1). However, since September 2011, the increase in the credit-to-GDP ratio has exceeded (or nearly exceeded) the 3 percentage point threshold suggested in IMF (2011b). The GFSR threshold is more conservative than other methodologies in that it flags risks more often, predicting more crises that fail to materialize, but missing fewer crises than other methodologies.4 Credit growth has been moderating but authorities should continue to monitor the evolution of credit to GDP ratio and other credit gap measures.
Chile: Assessing Credit Growth
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: IFS, WDI and Fund staff calculations.Chile: Assessing Credit Growth
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: IFS, WDI and Fund staff calculations.Chile: Assessing Credit Growth
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: IFS, WDI and Fund staff calculations.8. It is also important to look beyond credit extended by banks. A systemic financial risk assessment should look at all institutions that perform critical financial market functions, including credit intermediation, maturity transformation, the provision of savings vehicles, and the support of primary and secondary funding markets. In Chile, banks account for only about half of the financial system as measured by assets. The other half consists mainly of pension funds, insurance companies, and nonbank consumer outfits. The claims on the private sector by pension funds and insurance companies represent roughly 40 percent of the claims on the private sector by banks. Figure 2 applies the GFSR methodology to the broad credit-to-GDP ratio a broad measure of credit5 to the private sector, comprising banks, insurance companies and pension funds. Broad credit-to-GDP appears growing in line with bank credit. In particular, it has also been increasing above the 3 percentage point threshold suggested by the IMF (2011b) methodology to flag risks.6 The flip side of increases in credit is given by the increased leverage by borrowers. Household leverage (at about 36 percent of GDP) has been stable while corporate indebtedness (90 percent of GDP) has risen; both appear in line with Chile’s level of economic development from a cross-country perspective.7
Chile: Broad Credit Growth
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: IFS and Banco Central de Chile.1/ Narrow credit is defined as Claims on the Private Sector by Other Depository Corporations (ODC). Broad Credit is defined as Claims on the Private Sector by Other Financial Corporations (OFC). In the case of Chile, ODC include banks and OFC include pension funds and insurance companies.Chile: Broad Credit Growth
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: IFS and Banco Central de Chile.1/ Narrow credit is defined as Claims on the Private Sector by Other Depository Corporations (ODC). Broad Credit is defined as Claims on the Private Sector by Other Financial Corporations (OFC). In the case of Chile, ODC include banks and OFC include pension funds and insurance companies.Chile: Broad Credit Growth
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: IFS and Banco Central de Chile.1/ Narrow credit is defined as Claims on the Private Sector by Other Depository Corporations (ODC). Broad Credit is defined as Claims on the Private Sector by Other Financial Corporations (OFC). In the case of Chile, ODC include banks and OFC include pension funds and insurance companies.Multiple indicators
9. Analyzing the joint behavior of credit and other macroeconomic and financial indicators generally provides a better signal than just looking at credit.
Borio and Lowe (2002), Borio and Drehmann (2009) and IMF (2011b) show that combinations of credit and asset price deviations from long-term trends are the best leading indicator of banking distress. Reinhart and Rogoff (2009) and Barrel et al (2010) provide further evidence on the ability of housing prices to predict financial crises. GFSR 2011 shows that, in emerging economies, the real effective exchange rate (REER) tends to appreciate rapidly in the run-up to a crisis.
Kaminsky and Reinhart (1999) and Barrel et al (2010) find evidence that current account deficits can predict banking crises. However, when accounting for both current account deficits and credit growth, Jorda et al (2011) show that credit growth emerges as the single best predictor of financial instability. In recent decades, the correlation between lending booms and current account imbalances has grown much tighter.
10. In Chile, additional macroeconomic and financial indicators do not indicate an elevated systemic risk, but suggest some channels to require continued monitoring by authorities. We follow Borio-Lowe-Drehmann trying to detect the symptoms of the build-up of financial imbalances in monitoring unusually rapid and sustained growth in credit and in asset prices. For some small open economies, the cumulative appreciation of the real exchange rate might also be helpful. It could capture the pressure associated with capital inflows as well as the potential build-up of concomitant foreign exchange mismatches. Because time series for Chile are relatively short8, this note centers the analysis on growth rates instead of deviations from trend, in line with Fitch Ratings methodology to compute a Macro Prudential Index (MPI).9 Figure 3 shows the growth rates together with the indicative thresholds to flag risks suggested by Fitch ratings.
Chile: Credit Growth and Asset Prices
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Source: IFS and Banco Central de Chile.Chile: Credit Growth and Asset Prices
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Source: IFS and Banco Central de Chile.Chile: Credit Growth and Asset Prices
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Source: IFS and Banco Central de Chile.Stock prices. The behavior in the Santiago Stock Exchange IPSA index10 does not show any sign of asset boom and does not appear to be an increasing vulnerability that could potentially feed back into the real economy and financial sector.
Real effective exchange rate (REER). The peso is about 10 percent above its 1996-2012 average. Since REER time series are available back to 1980, the gap analysis in Borio-Lowe-Drehmann can be conducted. Over the last years, the REER gap exceeds the thresholds suggested in Borio-Lowe for all countries (7 percentage points) and for emerging economies (5 percentage points). However, it should be noted that the gap measures when using shorter data series is heavily influenced by the developments in the 80s. Staff analysis (see Chile 2013 IMF Article IV Consultation staff report) finds that the peso is on the strong side though not clearly overvalued.
Current account deficit. The current account has shifted to a deficit of 3.5 percent in 2012 from a surplus of 1.5 percent of GDP in 2010. Strong domestic demand has sustained strong imports (though moderating in 2012), while exports have suffered from some weakening in the still historically high copper prices and the sluggish demand in key partner countries. In staff’s view, while the current account deficit should be watched, it is not large enough to present immediate stability or sustainability risks. On the external financing side, foreign loans and deposit liabilities of the private sector (banks and non-banks) have been shown to accelerate rapidly before a crisis (GFSR 2011).11 In the case of Chile, private sector foreign loan and deposit liabilities are not accelerating (see below further discussion on the evolution of bank foreign liabilities).
Real estate prices. The Banco Central de Chile (BCCh) has started to assemble a database on residential real estate prices (lacking a few years ago). Data are not available on commercial real estate. Residential prices have been rising faster than CPI inflation. Average price dynamics have moderated but remain strong in Greater Santiago. Following IMF (2011b) and Lund-Jensen (2012), a panel logit model is used to estimate crisis-probability 2-years ahead using credit growth and asset price growth. While equity price growth was used in IMF (2011b), real house price growth is used in this chapter, together with changes in the credit-to-GDP ratio to estimate probability of banking crisis (see Annex I). Due to data limitations, Chile is not included in the estimation of the model, but the coefficients are used to track the predicted probability to the Chilean case. Based on the model, the probability of a financial crisis, while still low, appears to have risen in the last two years.
Credit and exchange rates (I)
(in percent)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: IFS and Fund staff calculations.Credit and exchange rates (I)
(in percent)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: IFS and Fund staff calculations.Credit and exchange rates (I)
(in percent)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: IFS and Fund staff calculations.
Credit and exchange rate (II)
(in percent)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Credit and exchange rate (II)
(in percent)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Credit and exchange rate (II)
(in percent)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Current Account Components
(in percent of GDP)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: IFS and Banco Central de Chile.Current Account Components
(in percent of GDP)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: IFS and Banco Central de Chile.Current Account Components
(in percent of GDP)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: IFS and Banco Central de Chile.
Private sector foreign liabilities to GDP
(yoy change; in percentage points)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Private sector foreign liabilities to GDP
(yoy change; in percentage points)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Private sector foreign liabilities to GDP
(yoy change; in percentage points)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Probability of a financial crisis in two years
(in percent)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: BCCh and Fund staff calculations.Probability of a financial crisis in two years
(in percent)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: BCCh and Fund staff calculations.Probability of a financial crisis in two years
(in percent)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: BCCh and Fund staff calculations.11. In addition to the indicators above, cross-country studies show that the stability of funding is an important potential vulnerability. We focus on two aspects: the role of deposit funding versus wholesale funding and the role of external versus domestic funding (Figure 4). In Chile, deposits comprise two thirds of total bank liabilities, of which two thirds are from retail and corporate customers. Institutional sources such as pension funds, mutual funds, and asset managers, together provide about 14 percent, and the rest are public sector and other deposits. Chilean banks’ loan-to-deposit ratios are above the 120-percent threshold identified in IMF 2011 to flag risks (based on cross-country data). However, the ratio ignores the stability of the liabilities other than deposits. In Chile, long term debt is the main source of financing for mortgage loans (accounting for roughly a quarter of bank loans). Additionally, external credit lines (historically very stable even during the 2008-09 crisis) are the main source of financing for foreign trade loans (near 10 percent of bank loans). Indeed, when adjusting the ratio by those two factors, the ratio drops below 100 percent. With respect to foreign funding, in Chile, foreign liabilities account for less than 15 percent of total deposits, way below the 30 to 40 percent threshold identified in IMF 2011 prior to crisis events.
Chile: Bank Funding Risks
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: SBIF and Haver Analytics.Chile: Bank Funding Risks
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: SBIF and Haver Analytics.Chile: Bank Funding Risks
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: SBIF and Haver Analytics.Developments in Chile’s Housing Sector
The Chilean real estate sector is important in terms of its contribution to GDP and its share in the portfolio of different agents. Mortgage debt represents around 60 percent of households’ debt and equal about 20 percent of GDP. Banks provide about 90 percent of mortgages and mortgages represent about 25 percent of banks’ loan portfolio. In addition, 8 percent of bank loans are to real estate developers and construction companies. The GDP share of the construction sector stands at 8 percent. Life insurance companies are exposed through endorsable mortgages loans, mortgage bills, and real estate (between 20 and 30 percent of total assets).
Real housing prices
(2004Q4=100)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Real housing prices
(2004Q4=100)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Real housing prices
(2004Q4=100)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Real housing prices
(2004Q4=100)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Real housing prices
(2004Q4=100)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Real housing prices
(2004Q4=100)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Residential mortgage debt to GDP /1
(in percent)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: OECD, European Mortgage Federation Hypostat 2011, SBIF and Banco Central de Chile.1/ Value for Chile is Household mortgage debt 2012.Residential mortgage debt to GDP /1
(in percent)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: OECD, European Mortgage Federation Hypostat 2011, SBIF and Banco Central de Chile.1/ Value for Chile is Household mortgage debt 2012.Residential mortgage debt to GDP /1
(in percent)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Sources: OECD, European Mortgage Federation Hypostat 2011, SBIF and Banco Central de Chile.1/ Value for Chile is Household mortgage debt 2012.
Chile - Mortgage loan-to-value evolution
(corresponding to new operations; in percent)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Chile - Mortgage loan-to-value evolution
(corresponding to new operations; in percent)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
Chile - Mortgage loan-to-value evolution
(corresponding to new operations; in percent)
Citation: IMF Staff Country Reports 2013, 199; 10.5089/9781484316481.002.A002
The housing market has been active over the last year. Sales have been strong, in line with the strong domestic demand and GDP growth, the labor market dynamics and the relatively low levels for mortgage interest rates. In terms of prices, aggregate price indexes vary widely depending on the methodology used (stratified, hedonic or repeated sales). Overall, the growth in real housing prices has been moderate but certain residential areas have seen sharp price increase. Loan-to-value ratios average has increased to around 85 percent (relatively high from cross country comparison). Moreover, most of the existing bank mortgages (85 percent) are non-endorsable mortgage loans, which have no regulatory loan-to-value limit.
In Chile, the real estate sector does not appear of immediate concern but authorities should remain vigilant and ready to act. Internal estimates by the BCCh show that the aggregate price trend is consistent with the dynamics in output and interest rates in the last couple of years for the hedonic and stratified sales index. The situation is not substantially different for the repeated sales index. Moreover, the housing price to disposable income ratio has been relatively stable during the last year, as well as households’ debt to disposable income ratio. Finally, according to the BCCh’s lending conditions survey, banking financing conditions for the sector have tightened in the second half of 2012. However, data is not yet available to confirm the effect on sales and prices. In particular, a risk remains that credit generation may have shifted to non-bank financial institutions. Authorities should continue to monitor if the high price trend in residential properties persists and becomes generalized to other sectors.
B. Systemic Risk Mitigation
12. Macroprudential instruments can help address potential sources of risk. Since the world is still gaining experience with macroprudential policies and since, in any event, the appropriate set and design of macroprudential tools depend on country-characteristics, there is no one set of tools and designs that can be seen as “best practice.” That said, countries can learn from budding experience of other countries.12 Because the systemic risk assessment in the previous section does not find elevated systemic risks (although some channels and risks require continued monitoring), this section focuses on the development of a macroprudential toolkit and not on the activation of specific macroprudential tools.
13. It should be noted that macroprudential policy is not well-suited to control asset prices or exchange rates.13 Rather, macroprudential policy can seek to contain the vulnerability of the system to asset price reversals. It is important to distinguish between macroprudential measures and capital flow management measures (CFMs). CFMs are designed to limit capital flows and affect the exchange rate. Macroprudential measures are designed to limit systemic vulnerabilities. These can include vulnerabilities associated with capital inflows and exposure of the financial system to exchange rate shocks, but macroprudential measures do not seek to affect the inflow or the exchange rate per se.
14. Chile has had experience with some prudential instruments that may be used for macroprudential purposes (e.g. loan-to-value and debt-to-income requirements, liquidity requirements and exposure and currency mismatch limits). Their use for macroprudential purposes would require a more active approach, including periodic recalibration. Other instruments, particularly to address the time dimension of systemic risk (e.g. time varying capital buffers and dynamic provisioning rules), have not been used in Chile and their implementation poses more challenges (Annex II). In particular, two (now) commonly discussed instruments are not part of the Chilean macroprudential toolkit and two other instruments present some limitation that may deserve consideration:
Dynamic provisioning. Dynamic provisioning is currently limited to voluntary provisions by banks. The individual evaluation models for unimpaired portfolio calculate provisions based on parameters like the probability of default and loss given default. Both parameters are defined in advance by the Supervisor, and are set at long-term estimates (representative of at least one cycle). Chan-Lau (2012) runs a simulation exercise for Chile and finds that Spanish dynamic provisioning would improve bank’s resilience to adverse shocks but would not reduce procyclicality. To address the latter, other countercyclical measures should be considered.
Countercyclical capital buffer. A countercyclical capital buffer (CCB) regulation is not in place in Chile. The minimum capital ratio is fixed by the General Banking Act and the BCCh may modify the requirements for market risk. The effectiveness of the CCB in smoothening credit cycles and thus reduce procyclicality in credit will depend on the level of capital that banks hold in excess of what the regulator requires.14 Chilean banks typically keep capital buffers above the required minimum. However, in order to assess the potential effectiveness of the CCB in Chile, it is important to assess the quality of the capital. The Basel III conservation buffer and the countercyclical “at its maximum” are supposed to be top quality capital (common equity), bringing the Core Tier 1 minimum ratio to 9.5 percent, the Tier I minimum ratio to 11 percent and the Tier 1 plus Tier 2 minimum ratio to 13 percent.
Risk weights. Active calibration of risk weights is readily available but with limitations, as the Superintendencia de Bancos e Instituciones Financieras (SBIF) may move risk weights up to one notch (with the agreement of the BCCh) and at most once a year.
Loan-to-Value (LTV) and Debt-to-Income (DTI). Chile has four mortgages types that differ in the source of funding and the possibility of originate to distribute. The dominant and fastest growing type (so called non-endorsable mortgages, which account for 85 percent of mortgages) is the only one that has neither LTV nor DTI caps. In the last three quarters average loan-to-value ratios have risen and should be monitored closely (and regulated if necessary).
15. The establishment of the Financial Stability Council (FSC) in 2011 is an important step to ensure close coordination among the institutions involved in Chile’s financial prudential framework. Having several agencies involved (Box 2) can make identification and mitigation of systemic risk less effective and accountability harder to establish. For instance, the decision-power over existing bank prudential tools is divided between the SBIF and the BCCh (Annex II). The BCCh has ownership over capital ratios, liquidity and currency mismatches regulation, reserve requirements, and LTV and DTI for certain types of mortgage loans. The SBIF has ownership on provisioning regulation, LTV and DTI for certain types of mortgage loans, and shared ownership over risk weights (as it requires BCCh approval). The establishment of the Financial Stability Committee – a forum for discussion without decision power- is an important step towards mitigating these risks.
16. In considering macroprudential policies it is important to take a holistic view. Macroprudential policies will lead to market reactions and the effects of the policies will need to be monitored and evaluated regularly. If policies clamp down on banking activities, other institutions may pick up the slack. Two channels of leakages that could potentially be important in Chile are cross-border arbitrage through direct cross-border lending and regulatory arbitrage through the part of the domestic financial sector falling outside the banking regulatory perimeter. In Chile, both channels can be important. The first could be important given the open capital account, the large presence of foreign-owned bank subsidiaries (representing half the system), and the active borrowing by Chilean corporates in international markets. The second could be important given the large size of the non-banking financial sector (accounting for 45 percent of consumer credit). It is therefore essential that the perimeter of systemic risk monitoring (and macroprudential regulation) be defined broadly to include all institutions which perform critical functions in financial markets, including credit intermediation, maturity transformation, the provision of savings vehicles, risk management and savings payments, and the support of primary and secondary funding markets.
The Financial Stability Council (FSC) in Chile
Objective. The FSC was established in 2011, with a clear a clear mandate for financial stability and macroprudential policy. Until 2011, the BCCh was the only institution with a mandate for financial stability1, in connection with its objectives of ensuring the due operation of both internal and external payments, and preserving the stability of Chile’s currency. The objective of the FSC is to coordinate and propose initiatives to look after the integrity and robustness of the financial system, fostering the coordination mechanisms and information exchange needed to ensure the adequate management of systemic risk, and to coordinate crisis management involving the roles and powers of its constituent bodies.
Membership. The FSC is chaired by the Minister of Finance and includes the Superintendents of the SBIF, SP and SVS. The Governor of the BCCh is invited to attend meetings on a permanent basis, although he is not a formal member to preserve its constitutional autonomy.
Functions and powers. The FSC is in charge of identifying, assessing and requiring the Superintendents to supervise risks to financial stability, reporting the results back to the council. It is vested with powers to obtain information from all financial industries and their participating institutions and to play a coordinating role to secure the consistency of financial stability efforts. It may recommend the implementation of macroprudential policies to the relevant agencies but does not have decision power and is not held accountable. Crisis management powers reside with the individual institutions and the Council operates as a coordination device.
1 The Capital Market Committee and the Superintendents’ Committee are important for coordination but do not have a formal legal basis.Annex I. Predicting the Probability of a Banking Crisis
The probability of a banking crisis is estimated with a panel logit model:
Pr(yi,t = 1|xi,t−h) = Φ (αi + xi,t−hθ + β * (DUM if ΔCtGi,t−h > 2) * RHPGi,t−h)
where yi,t denotes a binary banking crisis variable; xi,t−h is a row vector of explanatory variables, ΔCtG is the change in credit-to-GDP ratio and RHPG is the real house price growth; αi denotes the random effect for country i; Φ is the cumulative distribution function of a logistic distribution; and (θ, β) is a column vector of unknown parameters to be estimated. Note that all the indicator variables are known at time t − h. This analysis considers forecast horizons at 2 years.
We adopt the Laeven and Valencia (2010) definition under which a banking crisis is systemic if two conditions are present: (1) significant signs of distress in the banking system (as indicated by significant bank runs, losses in the banking system, and bank liquidations); and (2) significant banking policy interventions in response to significant losses in the banking system.
The basic specification includes growth in real house prices and the change in the ratio of credit to GDP as explanatory variables. We include an interaction term between a dummy for high credit growth and real house price growth. This intends to capture the idea in Borio-Drehmann (2009) that imbalances manifest themselves in the coexistence of unusually rapid growth in private sector credit and asset prices. To make the exercise informative for the Chilean case, we adopt a lower threshold (2 percentage points instead of 3 percentage points) than the threshold suggested in IMF, 2011b. But in actual estimation, both thresholds yield a significant effect on the cross product, β.
The change in credit to GDP ratio has a significant positive relationship with the crisis probability irrespective of the behavior in real house prices (Table A5.1). Real house price growth, however, show a significant effect on the probability of a banking crisis only during events of high credit growth. In line with Borio and Drehmann (2009), the interaction term captures the coexistence of asset price misalignments with a limited capacity of the system to withstand the asset price reversal. The specification chosen to compute the crisis probability in the main text is given by
Pr(yi,t = 1|xi,t−2 = Φ (−3.164 + 0.0651 * ΔCtGi,t−2 + 0.119 * (DUM if ΔCtGi,t−2 > 2) * RHPGi,t−2)
As robustness checks, a variety of alternative specifications were considered: fixed effects as opposed to random effects, a different threshold to determine the high credit growth dummy (3 percentage points) and different forecasting horizons (1 and 3 years). The coefficients on the change in credit-to-GDP and the interaction between high credit growth and real house prices growth appear to be stable under different specifications.
Determinants of Systemic Banking Crisis
Determinants of Systemic Banking Crisis
| VARIABLES | Bank_Cri |
|---|---|
| Change in Credi-to-GDP (t-2) | 0.0651** |
| (0.0306) | |
| Growth rate in real house prices (avg, t-2) | -0.0921 |
| (0.0623) | |
| Growth rate in real house prices (avg, t-2) * DUM[Change in Credi-to-GDP (t-2)>2] | 0.119* |
| (0.0660) | |
| Constant | -3.164*** |
| (0.320) | |
| Observations | 377 |
| Number of id | 25 |
| Sum Coefficients on Prices = zero | |
| test p-value | 0.291 |
Determinants of Systemic Banking Crisis
| VARIABLES | Bank_Cri |
|---|---|
| Change in Credi-to-GDP (t-2) | 0.0651** |
| (0.0306) | |
| Growth rate in real house prices (avg, t-2) | -0.0921 |
| (0.0623) | |
| Growth rate in real house prices (avg, t-2) * DUM[Change in Credi-to-GDP (t-2)>2] | 0.119* |
| (0.0660) | |
| Constant | -3.164*** |
| (0.320) | |
| Observations | 377 |
| Number of id | 25 |
| Sum Coefficients on Prices = zero | |
| test p-value | 0.291 |
Annex II. Banking Prudential Toolkit and Governance
Banking Prudential Toolkit and Governance
Banking Prudential Toolkit and Governance
| Macroprudential Tools | Chilean Banking Sector | Ownership* | Established By | Readily Available for Active Calibration/Macroprudential Use? | ||
|---|---|---|---|---|---|---|
| A. Leverage | ||||||
| Capital ratio | According to Basel I guidelines. Minimum capital ratio of 8%. Adjusted for exposures to market risk. | BCCh/SBIF | GBA and BCCh norm (III.B.2) | Requirements for market risk can be modified by CBCh; capital ratio is fixed by GBA. | ||
| Risk weights | Basel I RWs. For mortgages exposures higher RW 60% vs 50% | SBIF | GBA | Yes. GBA allows SBIF to move RWs up to one notch with CBCh authorization. | ||
| Provisioning | Voluntary additional provisions | SBIF | SBIF norm | Yes | ||
| Profit distribution restrictions | No restrictions beyond satisfying CAR | SBIF | GBA | No | ||
| Credit growth | ||||||
| Caps on credit growth | No caps | # | No | |||
| Caps on LTV | Mortgages exposures: | |||||
| 80% LTV on endorsable mortgage loans | SBIF | SBIF norm | Yes | |||
| 75% LTV on mortgage bills | BCCh | BCCh norm | Yes | |||
| 80% LTV on mortgage loans financed with “mortgage bonds” | BCCh | BCCh norm | Yes | |||
| No LTV on non-endorsable mortgage loans | SBIF | SBIF norm | Yes. | |||
| Other exposures: not in place | # | No | ||||
| Caps on DTI | 25% Dividend-to-income cap on mortgage bills loans up to UF3000 | BCCh | BCCh norm | Yes | ||
| 25% Dividend-to-income cap on mortgage loans financed with “mortgage bon | BCCh | BCCh norm | Yes | |||
| Gross leverage ratio | Minimum core capital to total assets of 3% | SBIF | GBA | No | ||
| B. Liquidity | ||||||
| Liquidity requirements | Sum of mismatches up to 30 days cannot exceed core capital. | BCCh | BCCh norm (III.B.2) | Yes | ||
| Sum of mismatches up to 90 days cannot exceed two times core capital. | BCCh | BCCh norm (III.B.2) | Yes | |||
| Reserve requirements | Reserve requirements: sigh deposits 9%, time deposits 3.6%. | BCCh | BCCh norm (Cap. 3.1) | Yes | ||
| Technical reserve: sigh deposits that exceed 2.5 times the net worth must be maintained in cash or in a technical reserve consisting of deposits with the CB. | BCCh | GBA | No | |||
| Currency mismatch/open FX position limit | Net FX flows up to 30 days cannot exceed core capital and built in to Market | BCCh | BCCh norm (III.B.2) | Yes | ||
| C. Interconnectedness | ||||||
| Concentration limits | Limits on interbank (established in Chile) loans: may not exceed 30% of bank’s creditor capital. | SBIF | GBA | No | ||
| Limits on (short-term) obligations with banks: short term obligations with one bank cannot exceed 5% of debtor’s current assets (individual limit). The sum of short term obligations with all domestic banks cannot exceed 40% of debtor’s current assets (global limit). | BCCh | BCCh norm (III.B.2) | Yes | |||
| Total lending to related parties (including parents) are limited to 25 percent of capital (secured) or 5 percent of capital (unsecured). | SBIF | GBA | No | |||
| Concentration limits by economic sector | No limits. | # | No | |||
| Systemic capital surcharge | Capital surcharges for systemically important entities. Capital ratio up to 14 percent. Technical reserve if deposits exceed 1.5 times its core capital. Limit on interbank exposure up to 20% of capital. | SBIF | GBA and BCCh norm (III.B.2) | Only for M&A that result in significant market share. | ||
Banking Prudential Toolkit and Governance
| Macroprudential Tools | Chilean Banking Sector | Ownership* | Established By | Readily Available for Active Calibration/Macroprudential Use? | ||
|---|---|---|---|---|---|---|
| A. Leverage | ||||||
| Capital ratio | According to Basel I guidelines. Minimum capital ratio of 8%. Adjusted for exposures to market risk. | BCCh/SBIF | GBA and BCCh norm (III.B.2) | Requirements for market risk can be modified by CBCh; capital ratio is fixed by GBA. | ||
| Risk weights | Basel I RWs. For mortgages exposures higher RW 60% vs 50% | SBIF | GBA | Yes. GBA allows SBIF to move RWs up to one notch with CBCh authorization. | ||
| Provisioning | Voluntary additional provisions | SBIF | SBIF norm | Yes | ||
| Profit distribution restrictions | No restrictions beyond satisfying CAR | SBIF | GBA | No | ||
| Credit growth | ||||||
| Caps on credit growth | No caps | # | No | |||
| Caps on LTV | Mortgages exposures: | |||||
| 80% LTV on endorsable mortgage loans | SBIF | SBIF norm | Yes | |||
| 75% LTV on mortgage bills | BCCh | BCCh norm | Yes | |||
| 80% LTV on mortgage loans financed with “mortgage bonds” | BCCh | BCCh norm | Yes | |||
| No LTV on non-endorsable mortgage loans | SBIF | SBIF norm | Yes. | |||
| Other exposures: not in place | # | No | ||||
| Caps on DTI | 25% Dividend-to-income cap on mortgage bills loans up to UF3000 | BCCh | BCCh norm | Yes | ||
| 25% Dividend-to-income cap on mortgage loans financed with “mortgage bon | BCCh | BCCh norm | Yes | |||
| Gross leverage ratio | Minimum core capital to total assets of 3% | SBIF | GBA | No | ||
| B. Liquidity | ||||||
| Liquidity requirements | Sum of mismatches up to 30 days cannot exceed core capital. | BCCh | BCCh norm (III.B.2) | Yes | ||
| Sum of mismatches up to 90 days cannot exceed two times core capital. | BCCh | BCCh norm (III.B.2) | Yes | |||
| Reserve requirements | Reserve requirements: sigh deposits 9%, time deposits 3.6%. | BCCh | BCCh norm (Cap. 3.1) | Yes | ||
| Technical reserve: sigh deposits that exceed 2.5 times the net worth must be maintained in cash or in a technical reserve consisting of deposits with the CB. | BCCh | GBA | No | |||
| Currency mismatch/open FX position limit | Net FX flows up to 30 days cannot exceed core capital and built in to Market | BCCh | BCCh norm (III.B.2) | Yes | ||
| C. Interconnectedness | ||||||
| Concentration limits | Limits on interbank (established in Chile) loans: may not exceed 30% of bank’s creditor capital. | SBIF | GBA | No | ||
| Limits on (short-term) obligations with banks: short term obligations with one bank cannot exceed 5% of debtor’s current assets (individual limit). The sum of short term obligations with all domestic banks cannot exceed 40% of debtor’s current assets (global limit). | BCCh | BCCh norm (III.B.2) | Yes | |||
| Total lending to related parties (including parents) are limited to 25 percent of capital (secured) or 5 percent of capital (unsecured). | SBIF | GBA | No | |||
| Concentration limits by economic sector | No limits. | # | No | |||
| Systemic capital surcharge | Capital surcharges for systemically important entities. Capital ratio up to 14 percent. Technical reserve if deposits exceed 1.5 times its core capital. Limit on interbank exposure up to 20% of capital. | SBIF | GBA and BCCh norm (III.B.2) | Only for M&A that result in significant market share. | ||
References
Ahuja, Ashvin and Malhar Naber, “Safeguarding Banks and Containing Property BoomsL Cross-Country Evidence on Macroprudential Policies and Lessons from Hong Kong SAR”, IMF Working Papers, WP/11/284
Almeida, Heitor, Liu, Crocker H. and Campello, Murillo, 2005, “The Financial Accelerator: Evidence from International Housing Markets.” Available at SSRN: http://ssrn.com/abstract=348740.
Arregui, Nicolas, Jaromir Benes, Ivo Krznar, Srobona Mitra, and Andre Oliveira Santos, 2013, “Evaluating the Net Benefits of Macroprudential Policy: A Cookbook”, forthcoming (Washington: International Monetary Fund).
Barrell, Ray, E Philip Davis, Dilruba Karim and Iana Liadze, 2010, “Bank regulation, property prices and early warning systems for banking crises in OECD countries,” Journal of Banking and Finance, Vol. 34, pp. 2255–2264.
Berger, Allen N., and Gregory F. Udell, 2004, “The Institutional Memory Hypothesis and the Procyclicality of Bank Lending Behavior,” Journal of Financial Intermediation, Vol. 13, pp. 458–95.
Borio, C. and P. Lowe, 2002, “Assessing the risk of banking crises,” BIS Quarterly Review, December, 43–54.
Borio, C. and M. Drehmann, 2009, “Assessing the risk of banking crises – revisited,” BIS Quarterly Review, March, 29–46.
Chan-Lau, J.A., 2009, “The Global Financial Turmoil and Its Impact on Chilean Banks,” IMF Country Reports 09/272 (Washington: International Monetary Fund).
Chan-Lau, J.A., 2010, “Too Connected to Fail Risk in the Chilean Banking System: a Balance Sheet Network Analysis,” IMF Country Reports 10/299 (Washington: International Monetary Fund).
Chan-Lau, J.A., 2012, “Do Dynamic Provisions Enhance Bank Solvency and Reduce Credit Procyclicality? A Study of the Chilean Banking System,” IMF Working Papers WP/12/124 (Washington: International Monetary Fund).
Committee on the Global Financial System, 2012, “Operationalising the Selection and Application of Macroprudential Instruments,” CGFS Papers No. 48.
Dell’Ariccia, Giovanni, Deniz Igan, Luc Laeven, and Hui Tong, 2012, “Policies for Macrofinancial Stability: Options to Deal with Credit Booms,” IMF Staff Discussion Note SDN/12/06 (Washington: International Monetary Fund).
Dell’Ariccia, Giovanni, and Robert Marquez, 2006, “Lending Booms and Lending Standards,” Journal of Finance, Vol. 61, No. 5, pp. 2511–546.
De Nicolò, Gianni, Giovanni Favara, and Lev Ratnovski, 2012, “Externalities and Macroprudential Policy,” IMF Staff Discussion Note, SND/12/05 (Washington: International Monetary Fund).
International Monetary Fund, 2008, “Financial Stress and Economic Downturns” World Economic Outlook, World Economic and Financial Surveys (Washington; October).
International Monetary Fund, 2011a, “Macroprudential Policy: An Organizing Framework” IMF Policy Papers (Washington; March).
International Monetary Fund, 2011b, “Toward Operationalizing Macroprudential Policies: When to Act?” Global Financial Stability Report, World Economic and Financial Surveys (Washington; September).
International Monetary Fund, 2011c, “Macroprudential Policy: What Instruments and How to use them? Lessons from Country Experiences,” IMF Working Papers 11/238 (Washington; September).
International Monetary Fund, 2012a, “Republic of Poland: Technical Assistance Report—Macroprudential Framework” IMF Country Reports 12/276 (Washington: International Monetary Fund).
International Monetary Fund, forthcoming, “Key Aspects of Macroprudential Policy” (Washington: International Monetary Fund).
Illing, Mark and Liu, Ying, 2006. “Measuring financial stress in a developed country: An application to Canada,” Journal of Financial Stability, Elsevier, Vol. 2(3), pages 243–65 (October).
Jorda, Oscar, Moritz Schularick and Alan Taylor, 2011. “Financial Crises, Credit Booms, and External Imbalances: 140 Years of Lessons,” IMF Economic Review, Palgrave Macmillan, Vol. 59(2), pages 340–378 (June).
Kaminsky, Graciela, and Carmen Reinhart, 1999, “The Twin Crises: The Causes of Banking and Balance-of-Payments Problems,” American Economic Review, Vol. 89, pp. 473 500.
Kuttner, K and I Shim (2012): “Taming the real estate beast”, mimeo.
Laeven, Luc, and Fabian Valencia, 2010, “Resolution of Banking Crises: The Good, the Bad, and the Ugly,” IMF Working Papers 10/146 (Washington: International Monetary Fund).
Lim, Cheng-Hoon, F. Columba, A. Costa, P. Kongsamut, A. Otani, M. Saiyid, T. Wezel, and X. Wu, 2011, “Macroprudential Policy: What Instruments and How to Use Them, Lessons from Country Experiences,” IMF Working Papers 11/238 (Washington: International Monetary Fund).
Lund-Jensen, Kasper, 2012, “Monitoring Systemic Risk Using Dynamic Thresholds,” IMF Working Papers 12/159, (Washington: International Monetary Fund).
Mendoza, Enrique G., and Marco E. Terrones, 2008, “An Anatomy Of Credit Booms: Evidence From Macro Aggregates And Micro Data,” NBER Working Papers 14049, National Bureau of Economic Research, Inc.
Rajan, Raghuram G., 1994, “Why Bank Credit Policies Fluctuate: A Theory and Some Evidence,” Quarterly Journal of Economics, Vol. 109, No. 2, pp. 399–441.
Rancière, Romain, Aaron Tornell, and Frank Westermann, 2008, “Systemic Crises and Growth,” Quarterly Journal of Economics, Vol. 123, No. 1, pp. 359–406.
Reinhart, C., and K. Rogoff, 2009, “This Time is Different: Eight Centuries of Financial Folly,” (Princeton, New Jersey: Princeton University Press).
Ruckes, Martin, 2004, “Bank Competition and Credit Standards,” Review of Financial Studies, Vol. 17, No. 4, pp. 1073–102.
Shin, Hyun Song and Shin, Kwanho, 2011, “Procyclicality and Monetary Aggregates,” No 16836, NBER Working Papers, National Bureau of Economic Research, Inc.
Vandenbussche, J., U. Vogel, and E. Detragiache, 2012. “Macroprudential Policies and Housing Prices-A New Database and Empirical Evidence for Central, Eastern, and Southeastern Europe,” IMF Working Papers 12/303, International Monetary Fund.
Wong, T. C., Fong, Tom, Li, Ka-Fai and Choi, Henry, 2011, “Loan-to-Value Ratio as a Macroprudential Tool - Hong Kong’s Experience and Cross-Country Evidence.” Available at SSRN: http://ssrn.com/abstract=1768546.
Prepared by Nicolas Arregui. Alejandro Jara and other seminar participants at the Central Bank of Chile provided useful comments.
Contributing to looser lending standards and greater credit cyclicality may be managerial reputational concerns (Rajan, 1994), improved borrowers’ income prospects (Ruckes, 2004), loss of institutional memory of previous crises (Berger and Udell, 2004), expectations of government bailouts (Rancière, Tornell, and Westermann, 2008), and a decline in adverse selection costs due to improved information symmetry across banks (Dell’Ariccia and Marquez, 2006). In addition, externalities driven by strategic complementarities (such as cycles in collateral values) may lead banks to take excessive or correlated risks during the upswing of a financial cycle (De Nicolò, Favara, and Ratnovski, 2012).
When applying their methodology to a sample of 170 countries from 1960 to 2010, Dell’Ariccia et al. (2012) find that one in three booms ends in a banking crisis (as defined in Laeven and Valencia, 2010) within three years.
See IMF (2011b) for comparison with the Borio-Lowe-Drehmann methodology and Arregui and others (forthcoming) for a comparison with the Dell’Ariccia methodology.
The series for credit is “Claims on Private Sector” in the IMF International Financial Statistics (IFS) database.
IFS data for Chile does not include investment funds, mutual funds, general funds, housing funds, foreign capital investment funds, factoring societies, leasing companies, and financial auxiliaries.
Currency mismatches in corporate and household balance sheet are limited and have been broadly stable.
Equity price index is available only since 1990 and housing prices are available only since 2002.
Fitch Ratings computes a Macro Prudential Index (MPI) that identifies the build-up of potential stress in banking systems due to rapid credit growth associated with bubbles in housing markets, equity markets or real exchange rates. High vulnerability to potential systemic stress is defined as: (i) real private sector credit growth exceeding an average 15 percent a year over two years, and; (ii) real property price growth of more than five percent a year in the same period, or; (iii) real effective exchange rate appreciation of more than four percent a year in the same period, or; real equity price growth of more than 17 percent a year (in the preceding two years).
The IPSA Index is a Total Return Index and is composed of the 40 stocks with the highest average annual trading volume in the Santiago Stock Exchange (Bolsa de Comercio de Santiago).
Foreign liabilities of the private sector refer only to loan and deposit liabilities and are taken from the balance of payment statistics (changes in international investment position for banks and non-banks under “other investment, liabilities.”
Lim et al. (2011) find that several macroprudential tools can reduce credit growth procyclicality. Arregui et al. (forthcoming) find that a variety of macroprudential tools has direct impact on banking aggregates such as credit growth. Almeida, Campello and Liu (2005), Wong et al (2011), Ahuja and Nabar (2011), IMF (2011d) and Kuttner and Shim (2012) study the effectiveness of LTV limits. Vandenbussche, Vogel and Detragiache (2012) look at the impact of capital requirements and liquidity measures on house prices.
See IMF (forthcoming).
With low excess capital, the introduction of the CCB will generate cost for banks provided that issuing new equity is relatively costly in comparison to other sources of funding.
