II. Capital Requirements, Leverage Ratios, Countercyclical Capital Buffers and Dynamic Provisions
In the aftermath of the financial crisis the regulatory discussion has centered on strengthening the solvency of individual banks and reducing procyclicality. Among other measures, the new Basel III regulatory framework recommends the use of higher and better quality capital and the introduction of leverage ratios to strengthen the resilience of banking institutions; and the adoption of countercyclical capital requirements to build up buffers which can be drawn down during periods of distress. In addition, regulators are exploring the merits of dynamic (statistical) provisions. This note describes these measures concisely and assesses their implications for Latin America.
B. The Tools and Their Objectives: Solvency and Leaning Against the Wind
Provisions and capital buffer a bank from credit losses. Provisions can be either general, to account for expected losses in the portfolio that have yet to be identified since they have not realized yet; or specific, to account for losses from specific impaired loans and write-offs. Because ex-ante loss estimates may differ from realized losses a bank holds another buffer, capital, to be able to cover unexpected losses, or losses beyond the mean ex-ante estimate (Figure 1). Clearly, the adequacy of provisions and capital to withstand losses depends on how reliable the estimated loss distribution is.
Figure 1.Capital and Provisions
Capital comprises Tier-1 capital and Tier-2 capital, or “high” quality capital and “lower” quality capital. Tier-1 capital, which includes common equity, allows a bank to absorb losses while remaining a going concern. Tier-2 capital, which includes subordinated debt, provides loss absorption on a gone-concern basis.2 Regulatory capital requirements specify that the ratio of capital to risk-weighted assets should be above a pre-specified minimum level. Because of the substantial capital losses incurred by banks during the financial crisis, Basel III now requires banks to hold common equity and total capital in excess of 7 and 10½ percent respectively (Table 1).).3 The shift towards higher levels of common equity, or high quality capital, has been guided by the fact that non-core Tier-2 capital could be very volatile during periods of distress, as observed during the financial crisis.
|Common equity||Tier 1 capital||Total capital||Countercyclical|
|Basel III||4.5||2.5||7.0||6.0||8.5||8.0||10.5||0 - 2.5|
Minimum leverage ratios work against excessive leverage and risk-taking incentives in banks. By requiring a bank to hold a minimum level of Tier-1 capital relative to total assets, regulators reduce leverage and its associated risks. Because leverage ratios are measured relative to total assets, they complement capital requirements: a bank holding little capital can meet the regulatory capital requirement by holding safer assets and reducing the amount of risk-weighted assets. In the absence of a leverage ratio, a bank could build up excessive leverage even if it complies fully with capital requirements. Basel III proposes a minimum leverage ratio of 3 percent during a trial period from January 2013 to January 2017.
Countercyclical buffers attempt to reduce the buildup of risks during economic booms and financial in the ensuing downturn. Two such measures have featured more prominently in the policy discussion. The first measure, a countercyclical capital buffer, requires banks to build an extra layer of common equity during the upswing of the cycle. The buffer aims to ensure that, in addition to safeguard individual bank solvency, the banking sector in aggregate can help to maintain the flow of credit in the economy during an economic downturn. Basel III proposes a countercyclical capital buffer in the range of 0 to 2½ percent which would be triggered by changes in an aggregate credit indicator. Therefore countercyclical capital buffers would apply system-wide. The second measure, dynamic provisions, requires banks to build up provisions during an economic expansion that would later offset loan losses when the economy slows down or contracts. In contrast to countercyclical capital buffers, dynamic provisions are generally bank-specific and calibrated according to the bank’s lending activity. Both measures could help to dampen excess credit growth during an expansion. The countercyclical capital buffer raises the cost of credit reducing its demand.4 Dynamic provisions, by requiring banks to hold higher provisions, reduce the resources available for funding loans and help restrain credit growth (Box 1).
Box 2.1.Accounting Treatment of Provisions and Capital
Provisions can be either general, to account for expected losses in the portfolio that have yet to be identified since they have not realized yet; or specific, to account for losses from specific impaired loans and write-offs. General provisions are considered appropriations of retained earnings so their increase reduces the capital of the bank. Specific provisions are considered a current expense and can be deducted from taxes. The differential tax treatment provides banks with incentives to minimize general provisions and end under-provisioned relative to expected future losses. Under Basel II, the incentive was partly offset by the allowance to count general provisions towards Tier II capital up to a maximum of 1.25 percent of risk-weighted assets.References: Sunley (2003) and Ryan (2007).
C. Implications for Latin America
Latin American banks hold higher capital levels and are less leveraged than banks in advanced industrialized countries. The new capital requirements and the minimum leverage ratio proposed by Basel III were designed to address shortcomings of banks in industrialized countries. But Latin American banks appear to exhibit characteristics quite different from their peers in advanced countries. Data for Brazil, Chile, Colombia, Mexico, Peru, and Uruguay show that banks are well capitalized. In general, the common equity to risk-weighted assets ratio in most banks is above the minimum required in Basel III, and in many instances, the ratios also satisfy the required conservation and countercyclical buffers (Figure 2). Similarly, banks have leverage ratios well above the minimum 3 percent proposed in Basel III (Figure 3).5 These findings reflect the prudent stance of the supervisory authorities and reforms in regulation and supervision (Caruana, 2010).
Figure 2.Common Equity to Risk-Weighted Assets Ratio Distribution in Selected Latin American Countries1
Sources: Central banks and/or national banking superviosry agencies, and author’s calculations.
Note: Basel III proposes a minimum common equity to risk-weighted asset ratio of 4.5percent, augmented to 7 percent to include conservation buffer, and to 9.5 percent after adding a capital buffers and counter cyclical capital.
1 Data as of June 2010, except for Uruguay(Janaury2011).
Figure 3.Leverage Ratio Distribution in Selected Latin American Countries
Sources: Central banks and/or national banking superviosry agencies, and author’s calculations.
Note: Basel III proposes a minimum leverage ratio, calculated as Tier-1 capital to total assets, of 3 percent.
1 Data as of June 2010, except for Uruguay(Janaury 2011).
Because Latin America is not immune to boom and bust cycles, dampening procyclicality remains a major policy challenge. In some instances, substantial capital inflows have contributed to excess credit growth in the region and the buildup of credit risk. Because banks appear well capitalized and leverage is low, it could be argued that the main rationale for adopting countercyclical buffers is to “lean against the wind” and reduce procyclicality rather than to enhance bank solvency.
Countercyclical capital buffers, as long as they have a non-negligible impact on the cost of credit, could help to smooth the flow of credit. During the expansionary stage of the cycle, higher capital buffers would increase the cost of credit, i.e. lending spreads, and slowdown the credit expansion. On the other hand, during an economic downturn, the release of the capital buffer, i.e. allowing banks to reduce their capital ratios by the amount of the countercyclical capital buffer, could help to ensure that bank lending is not curtailed drastically.6Figure 4 shows the sensitivity of lending spreads to increases in the capital ratio, in percentage points, of a representative bank in LA 5 countries.7
Figure 4.Capital Requirements and Lending Spreads, LA5 Countries
Source: Bankscope, BCBS, Haver Analytics, central banks and banking supervisory agencies, author’s calculations
Increasing capital requirement during the upswing of the cycle could be an effective countercyclical tool in Brazil and Chile. In these two countries, a 1 percent increase in the capital ratio causes lending spreads to widen by 5 to 6 percentage points. Interest income is substantial for banks in these countries, so keeping the rate of return on equity unchanged following an increase in capital requirements requires charging higher spreads to clients. In contrast, lending spreads in Colombia, and especially Peru, do not react much to higher capital requirements. Because capital ratios in the region exceed 10 percent, countercyclical capital requirement may require raising them to levels well above those recommended in Basel III.
A countercyclical leverage ratio could serve as an alternative to the countercyclical capital buffer. Increasing the leverage ratio implies increasing the share of capital relative to total liabilities which raises the cost of capital to the bank, and in turn to borrowers. A countercyclical leverage ratio, hence, could serve the same purpose as a countercyclical capital buffer. Figure 5 shows the sensitivity of lending spreads to increases in the leverage ratio, in percentage points. Given that the observed leverage ratio of a representative bank in Latin America is high, achieving a moderation of domestic credit growth may require setting leverage ratios as high as 10 percent during the upswing of the cycle.8
Figure 5.Leverage Ratios and Lending Spreads, LA5 Countries
Source: Bankscope, BCBS, Haver Analytics, central banks and banking supervisory agencies, author’s calculations
Standard provision practices contribute to credit procyclicality and increased insolvency risk of individual banks during downturns.9 There are incentives for banks to under-provision during good times, including differential tax treatment and compensation schemes directly related to lending volumes, profits, and earnings. Under-reporting provisions is common in advanced and emerging market economies (Bikker and Metzemakers, 2005) and contributes to procyclicality (Brunnermeier et al, 2009, Burroni et al, 2009). The under-reporting of provisions frees resources that lead to increased lending. When the cycle turns, a credit crunch ensues as large capital losses due to nonperforming loans lead to a drastic reduction of credit.
Dynamic provisions aim at building up provision buffers ahead of realized losses which helps reducing lending procyclicality. Different regulatory dynamic provisions specifications build on the principle that provisions should always be set in line with estimates of long-run, or through-the-cycle expected losses (Mann and Michael, 2002). Regulatory dynamic provisions are usually based on variations of the simple formula below (Burroni et al, 2009):
where specific provisions correspond to realized losses. The formula shows that during good times dynamic provisions are positive and contribute to the increase in loss provisions as realized losses are below the through-the-cycle loss ratio. During bad times, the opposite takes place and negative dynamic provisions deplete the loss provision buffer. Smoother profits work against procyclicality, and the build-up of provisions consistent with through-the-cycle estimates reduce the probability of failure of banks during a downturn.10 Besides Spain, a number of Latin American countries have already adopted regulatory dynamic provisions (Box 2).
Box 2.2.Dynamic Provisions in Spain and Latin America
The Spanish dynamic provisions formula builds general provisions that accounts for expected losses in new loans extended in a given period and the expected losses on the outstanding stock of loans at the end of that period after netting off specific provisions incurred during the period. If new loans of an homogeneous category k are denoted by Δ
where the different loan categories, and the choice of parameters in the formula above are determined by the banking regulatory agency. There are six different loan groups or categories in ascending order of risk: negligible risk, low risk, medium-low risk, medium-risk, medium-high risk, and high risk. The general provision parameters, or alpha-parameters, corresponding to these groups are 0, 0.6, 1.5, 1.8, 2, and 2.5 percent respectively; and the specific provision parameters, or beta-parameters, are 0, 0.11, 0.44, 0.65, 1.1, and 1.64 percent respectively. The system also specifies that cumulative provisions should not exceed 125 percent of the inherent losses of the loan portfolio,
Reference: Saurina (2009).
Banks are required to maintain a dynamic provision in the range of 1½ to 5½ percent of total loans, depending on the type of loan: 1½ percent for mortgage loans, 1.6 percent for microfinance loans; 2.3 percent for consumer loans and prime corporate loans; and between 3.2 percent and 5.5 percent for subprime corporate loans. Banks can access the provision stock to offset up to half of the additional specific provisions required in a given month provided that the loan quality has deteriorated for six consecutive months and the dynamic provision has been phased in fully). Reference: Wezel (2010).
Colombia adopted dynamic provisions for commercial and consumer loans in 2007. Banks can measure the credit risk of the loans using either the regulatory reference model or approved proprietary models. The regulatory model establishes three types of tax-deductible provisions: individual, countercyclical, and general provisions. General provisions should exceed 1 percent of the total loan portfolio, and can be used to meet countercyclical provisions. Countercyclical provisions cover credit risk from changes in the borrower’s creditworthiness due to changes in the economic cycle. Both individual and countercyclical provisions are accounted under the same item. In the reference model, individual provisions are calculated based on expected losses under a regulatory baseline scenario. During periods of rapid growth, countercyclical provisions are calculated as the difference between expected losses in a more adverse scenario and the baseline scenario. During periods of slow growth, countercyclical provisions are not required. Finally, banks can use countercyclical provisions at the discretion of the regulator to compensate for increases in individual provisions during an economic downturn. Reference: Fernandez de Lis and Garcia Herrero (2010).
The countercyclical provisioning rule requires Peruvian banks to build up additional minimum provisions whenever the rule is activated by one of the conditions below:
- a) the annualized average percent change of GDP during the past 30 months reaches or exceeds 5 percent from below;
- b) the annualized average percent change of GDP during the past 30 months is above 5 percent and the average annualized percent change of GDP during the past 12 months exceeds by 2 percentage points its value one year before;
- c) the annualized average percent change of GDP during the past 30 months is above 5 percent and 18 months have elapsed since the rule was deactivated by second deactivation condition.
Countercyclical provisions are deactivated by one of the two conditions below:
- a) the annualized average percent change of GDP during the last 30 months falls to or below 5 percent;
- b) the annualized average percent change of GDP during the last 12 months is lower by at least 4 percentage points than its value one year before.
The minimum countercyclical provision is 0.4 percent for corporate clients, 0.45 percent for large enterprises; 0.3 percent for medium-sized corporates; 0.50 percent for small and micro-corporates; 1.5 percent and 1 percent for revolving and nonrevolving consumer loans respectively, and 0.4 percent for mortgage loans. Reference: SBS, 2008, Resolutión S.B.S. No. 1356, November 19.
Dynamic provisions were introduced in 2001. The regulation specifies that banks contribute to their individual dynamic provisioning funds, DPt, the difference between the monthly statistical net losses on loans to the nonfinancial private sector (NFPS) and the realized net loan loss in that month:
The statistical losses are derived by multiplying 1/12 of the expected rate of loss for five loan categories, αk ranging from 0.1 percent for low-risk loans to 1.8 percent for credit card loans, by the respective loan volumes, Ct. The net loan loss, LLt, incurred in a given period is calculated as the cost of additional specific provisions recorded in the profit and loss statement, net of deactivations of specific provisions (i.e., reclassifications of loans toward higher categories) and recoveries of defaulted loans already written off. At the inception of dynamic provisioning, the beta parameters were reportedly distributed around the average annual loan loss during 1990-2000, which was 1 percent of loans. The dynamic provisions fund of each bank is bounded between 0 and 3 percent of total loans to be provisioned. Reference: Wezel (2010).
There is no consensus yet on whether dynamic provisions are effective tools for leaning against the wind. While it is acknowledged that dynamic provisions helped Spanish banks to withstand the financial crisis better than banks in other advanced industrialized countries (Saurina, 2009), the jury is still out on whether they reduced procyclicality. Fernandez de Lis and García-Herrero (2010) indicate that dynamic provisions did not discourage credit growth nor prevented a housing bubble. They suggest that provisions only have a marginal impact on credit growth.11 The same authors also point out that in the case of Colombia, banks offset required increases in dynamic provisions, i.e., counter-cyclical provisions, by reducing other provisions. While the introduction of dynamic provisions in 2007 induced sharp increases in the ratio of countercyclical provisions to total provisions and in percent of total loans, total provisions as percent of total loans did not change significantly (Figure 6). On the other hand, a recent study by Jimenez et al (2011) finds that dynamic provisions helped to mitigate procyclicality in Spain.
Figure 6.Colombia: Provisions, 2000–10
Sources: Superintendencia Financiera de Colombia and author’s calculations
System-wide dynamic provisions triggered by changes in aggregate economic activity could prove more effective for restraining credit growth. Although more prudent banks are penalized, system-wide provisions would force all banks to increase provisions regardless of whether they are expanding their lending activity. Such provisioning system has been in place in Peru since 2008. Dynamic, countercyclical provisions in excess of general and specific provisions are required when the growth of economic activity exceeds a regulatory threshold, usually set at the growth rate of potential output, or if the year-on-year growth accelerates rapidly. The countercyclical provisions requirement is deactivated when economic growth falls below potential or the economy slows down substantially (see Box 2).
Countercyclical capital requirements and leverage ratios could help restrain excessive credit growth in Latin America by raising the cost of capital. In the case of capital flows-driven credit growth, these measures could complement other macroprudential tools like LTV limits and reserve requirements and facilitate the conduction of monetary and fiscal policy. Nonetheless, using these countercyclical tools may require raising capital requirements and leverage ratios well above the levels recommended in Basel III. The effectiveness of these tools could be impaired if a substantial share of the flow of credit is channeled outside the banking system, or if banks are willing to accept a lower return on equity rather than passing the increased cost of capital to borrowers.
System-wide dynamic provisions triggered by changes in aggregate economic activity could reduce procyclicality. Peru implemented such system in 2008 and it is rather early to assess its performance as provision buffers have only started to build up recently. Nonetheless, the fact that system-wide dynamic provisions lean against the wind as countercyclical capital buffers allows extrapolating the findings from BCBS (2010c). The findings suggest that countercyclical buffers triggered by aggregate indicators could do a good job in reducing procyclicality.12 It should be bear in mind, however, that provisions based on aggregate indicators could impair efficiency and competition in the banking system (Fernandez de Lis and García-Herrero, 2010).
Regardless of whether aggregate or bank-specific variables trigger dynamic provisions, their success depends on reliable estimates of long-run expected loss and a balance between rules and discretion. Estimating long-run expected losses remains a formidable challenge. For instance, in Peru expected loss estimates are based on the banking crisis experienced in the 1990s, which some argue it is too conservative and may put domestic banks at disadvantage vis-à-vis other credit providers. In dynamic provision regimes, calibration relies on past historical data and it may fail to capture the dynamics of expected losses going forward and should be complemented with discretionary judgment. Reaching the adequate balance between rules and discretion remains a challenge (Ocampo, 2003, Turner Review, 2009, Griffith-Jones and Ocampo with Ortiz, 2009).