New In-house Estimates of the Impact of the Basel Capital Framework Andrew Cornford

The state of play in the effort to achieve a final revised version of the Basel III capital framework for banks (now widely and understandably renamed Basel IV) is currently far from clear. The word from the Basel institutions themselves is that the goal is close to attainment. In mid-September the Group of Central Bank Governors and Heads of Supervision (GHOS), the oversight body of the Basel Committee on Banking Supervision (BCBS), issued a statement endorsing the direction of Basel III. This was accompanied by a statement of the chairman of the BCBS that the long process of post-GFC reforms could be finalized by the end of 2016. Nevertheless, doubts persist. A recent meeting of negotiators in Chile failed to progress beyond what has been described as an outline deal. Finalization by the GHOS is now scheduled for January 2017.

Positions articulated by regulators elsewhere in member countries of the BCBS indicate obstacles to the achievement of final consensus. In some EU countries the framework is under attack on the grounds that Basel III is likely to raise capital requirements to levels which will lower economic activity, that its restrictions on the use by banks of internal models to calculate these requirements will have similar effects, and that the measure of exposures in the denominator of the minimum leverage ratio, which is intended to reinforce the protection provided by capital requirements based on risk-weighted exposures, fails to allow for the risk-free or low-risk character of some of banks’ on- and off-balance-sheet positions.

Criticisms with the opposite thrust have been coming from the United States, where there is significant sentiment amongst some regulators and academic commentators that, so far from softening, the requirements of the Basel framework need strengthening. Such views have already had an impact on regulatory capital requirements for large banks. Moreover in the United States there are currents of opinion in the ruling Republican party which favour dismantling substantial parts of the Dodd-Frank Wall Street Reform and Consumer Protection Act of 2010 and replacing them with substantially higher capital requirements for banks.

Into these controversies have stepped two BIS staff members with estimates of the framework’s macroeconomic impact (Fender and Lewrick, 2016). Their paper is a sequel to and revision of earlier estimates of the BCBS in 2010 (BCBS, 2010).(reviewed in G24 Policy Brief No. 65, 26 November  2010: Andrew Cornford, “Macroeconomic effects of Basel III: estimates of the FSB and BCBS”).

The new study makes use of analytical building blocks similar to, but going beyond, those of the 2010 long-term economic impact (LEI) study. The expected benefits of the capital framework result from the reduction of expected post-reform losses of output due to systemic banking crises. The estimates of these benefits result from multiplying the probability of such crises by their macroeconomic cost, should they occur. The probabilities are estimated on the basis of econometric models yielding a functional relationship between crisis probabilities and capital levels. The two alternative benchmark measures of the net-present-value impact of crises on GDP in the new study – 63 per cent and 100 percent of GDP –are derived from a wide range of alternatives suggested by previous studies.

Expected net benefits of increased capital requirements are estimated through a comparison of expected benefits with the estimated costs of increased regulation under the capital requirements of Basel III. Estimated costs follow from the assumption that banks counterbalance declines in their returns on equity due to regulation by increases in the spreads incorporated in their lending rates of interest. To estimate the consequent loss of output – the macroeconomic impact -the increased spreads are then fed into several different macroeconomic models used by central banks participating in the exercise.

Since the 2010 study of the BCBS there have been a number of changes in the proposed regulatory framework of Basel III. One of these changes, which has contributed to the increased compass of the new study, is the enhanced focus on the risks associated with systemically important financial institutions. For the purpose of policy measures aimed to address these risks in November 2011 the Financial Stability (FSB) identified a group of Globally Important Systemically Important Banks (G-SIBs) in accordance with a methodology developed by the BCBS. This group, periodically reviewed and now consisting of 30 banks, is allocated to buckets corresponding to required levels of supplementary capital buffers.

The items eligible for inclusion in the numerator of the required capital ratio are now based on more restrictive definitions of what constitutes eligible capital, while the risk-weighted exposures in the denominator incorporate more stringent requirements designed to reduce recourse by banks to lowering their exposures through inadequate accounting for risks. For G-SIBs there is to be a higher loss absorbency capital surcharge (HLAC).This surcharge in the form of equity and other instruments with a similar capacity for loss absorption, so-called CET1 capital, is to be in a range where the figure for an individual G-SIB is determined by the bucket to which it is assigned by the list of G-SIBs described above. For the purpose of the estimates of the effects of the revised framework for the G-SIBs included in its sample of banks the new study applies an average surcharge of 1 per cent of risk-weighted exposures.

Under the changed regulatory framework banks must now maintain a minimum leverage ratio (LR) of 3 per cent. The numerator of the leverage ratio is equity (Tier 1) capital, and the denominator is the sum of the following: on-balance-sheet exposures (such as loans), derivative exposures at replacement cost, exposures associated with the financing of securities transactions, and off-balance-sheet exposures (such as letters of credit). The option of additional LR requirements for G-SIBs, leading to an LRS, is also under discussion. Since there is not yet agreement on the level of the G-SIB surcharge, for its estimates the new study deploys two alternative possibilities, 0.5 per cent and 2 per cent.

For the purpose of estimating the additional costs to banks resulting from the LR the effects are translated into the additional capital required for meeting the LR (and the LRS for G-SIBs). The minimum LR of 3 per cent translates into additional capital of 0.7 of risk-weighted exposures, and the LRS into a capital surcharge of 0.1 per cent for a LRS of 0.5 per cent and 0.8 per cent for a LRS of 2 per cent.

Total increases in capital due to Basel III are then obtained by adding to these figures the basic minimum ratio of 7 per cent for CET1 capital in relation to risk-weighted assets and, when applicable in the case of G-SIBs, the HLAC of 1 per cent.

The higher levels of capital required feed into a reduction of the probability of crises. Since a range of higher capital levels is indicated in the new study, there is also a range for the reduced probability of crises. Here too the reduced probability varies with the classification of the bank by source of potential systemic risk: when the applicable figures are the minimum required capital ratio of 7 percent and a LR of 3 per cent, the reduction is 1.96per cent; when a 1-per-cent LR is added, the reduction rises to 2.15 per cent; and when a LRS is included, the reduction is 2.19 for a LRS of 0.5 per cent and 2.41 per cent for a LRS of 2 per cent.

For the benchmark assumption of 63-per-cent net present value of GDP for the cost of a systemic banking crisis the expected benefits of the increased capital requirements vary between 1.24 per cent and 1.52 per cent of GDP according to the bank’s classification by systemic riskiness. For the alternative assumption of the cost of 100-per-centnet present value of GDP for the cost of systemic riskiness the expected benefits vary between 1.96 and 2.41.

The expected costs of higher capital levels also vary between 0.26 and 0.41 per cent of GDP according to the bank’s classification. The resulting net benefits thus vary between 0.97and 1.11 per cent of GDP for the assumed cost of a systemic banking crisis of 63-per-cent net present value of GDP, and between 1.7 and 2 per cent of GDP for the assumed cost of a systemic banking crisis of 100-per-cent net present value of GDP.

The new study then qualifies these figures for net benefits to take account of various additional factors: the impact of additional capital due to Total Loss-Absorbing Capacity (TLAC) (for G-SIBs); the exaggeration of banks’ costs implied by the assumed static balance-sheet approach used to estimate the additional capital banks would need to meet the LR (for all banks) and the LRS (for G-SIBs); and adjustment of banks’ estimated funding costs to take account of their reduction in response to increases in capital implied by the changes in the capital framework.

The TLAC requirements reflect the FSB’s scepticism that the supplementary capital requirements prescribed in Basel III for G-SIBs would be sufficient to guarantee for banks facing insolvency a resolution strategy which mitigated risks to financial stability, ensured continuity of critical banking functions (such as payments), and minimised taxpayers exposures to losses. The instruments required for TLAC have to be available for resolutions – i.e. with usability not restricted by their terms. They overlap with – but do not include all of – the capital instruments which count towards the fulfilment of the minima of Basel III. TLAC requirements are expected to reduce the likelihood and costs of systemic banking crises, and thus also the benefits associated the reforms of Basel III. The net benefits of the reforms, which are assumed to vary with the cost of systemic crises and the level of LRS, are somewhat reduced and amount to figures of between 0.37 and 0.87 of GDP.

The net estimated benefits of Basel III specified earlier may be conservative because the assumed capital shortfalls do not take account of banks’ independent adjustments (“balance-sheet optimisation”) of their capital levels to avoid or minimise the impact of LR and LRS. The new study’s estimates of the various possible differences in net marginal benefits which would result from assumed reductions in risk-weighted assets as part of such adjustments are none the less not large.

Over time higher capital ratios can be expected to reduce banks’ costs of funding owing to the decline in the required return on equity due to the lower leverage of their balance sheets (a reduction often classified under the heading of the Modigliani-Miller theorem of financial economics). The new study makes no attempt at its own estimate of this effect, instead citing other studies in its support.

The conceptual framework underlying the new study, at least implicitly, seems primarily focussed on sophisticated banks. Substantial parts of the study concern potential effects of Basel III on G-SIBs which, with some Chinese exceptions, have parent institutions in major advanced economies. It is not clear from the study how many of the model inputs used to estimate the macroeconomic effects of the new capital framework are provided by the central banks of emerging-market countries.

The 2010 LEI study was already open to the criticism that its focus was banks in advanced countries, and the same criticism applies a fortiori to the new study. The Global Financial Crisis began, and has had its most spectacular effects, in advanced economies so that the focus of the new study is not surprising as the authorities in these economies, which still dominate the membership of the BCBS, seek out ways to protect themselves from a recurrence.

Nevertheless this emphasis raises questions as to the applicability of the conclusions of the new study to emerging-market and developing economies. A standard reply to these questions from the designers of the capital framework is that it is intended for internationally active banks. However, this reply fails to acknowledge that since the introduction of Basel I the capital frameworks have served as basic models for rules on capital and risk management worldwide – although of course there have been and continue to be differences in the detailed formulation and application of these rules in different countries.

Estimating the impact of Basel III is not the same as estimating the overall impact of the current reform agenda for banking sectors. Many subjects which figure importantly in this agenda, such as caps on bank size and other possible changes in corporate form such as the compartmentalisation of activities by institution, are not covered. Approaches to some of these subjects can be expected to differ amongst countries, in response both to variations in the national conditions of financial sectors and in some cases to political sensitivities. Many critics might none the less argue that such omissions seriously handicap the reform agenda’s effectiveness as a shield against further crises.The inclusion of measures along these lines could lead to estimates of the benefits of the agenda which significantly exceed those of Basel III considered in isolation.

As noted at the beginning of this commentary, work on the finalization of Basel III is ongoing as part of an exercise which began more than 15 years ago. The still provisional state of Basel III is a major reason for the number of different estimates of its possible effects put forward in the new study. Despite the high-level assurances coming from the BCBS and senior financial officials, it is legitimate to wonder how and when existing disagreements will be resolved.

Some these disagreements focus on the prudential effectiveness and appropriateness of the rules of Basel III. These disagreements may be thorny but should eventually be resoluble through compromise.

But there is another potential threat to a near-term satisfactory conclusion of this process, without which estimates of the effects of Basel III are not final. Since agreement on Basel I in 1988 a major purpose of the reforms of the banks’ capital frameworks has been the achievement of similarity in the regulatory treatment of banks in order to equalise competitive conditions  for their operations (“to create a level playing field” to use the phrase beloved of officials and commentators). In view of national differences in economic structures, policies towards financial sectors, and other features of countries’ traditions, this objective may be a chimera. Targeting too comprehensive or too detailed a version of this objective could lead to prolongation of attempts to homogenise the framework of reforms which only political exhaustion will eventually be able to call off.

References

BCBS (2010), An assessment of the long-term economic impact of stronger capital and liquidity requirements, Basel BIS, August;

Fender I and Lewrick U (2016), “Adding it all up: macroeconomic impact of Basel III and outstanding reform issues”, BIS Working Paper No. 591, Basel, November.