TWN Info Service on Finance and Development
Estimating the macroeconomic effects of Basel
Largely responsible for this heightened attention have been the writings (usually critical) of commentators on the reform agenda and pronouncements of the big banks themselves. For the latter, the stakes are high: capital levels are major determinants of the denominator of banks' rate of return on equity, i. e. the standard measure of their profitability; raising additional equity to meet the new capital requirements may be costly, equity being the most expensive form of financing; and stricter rules concerning the liquidity of banks' assets and liabilities are likely to exert downward pressure on their earnings.
As a result, the banks have engaged in intense lobbying to influence the final shape of the rules of Basel III.
The initial attempts to forecast the impact of Basel III summarised below have concerned macroeconomic effects. The models and data used for these forecasts are not suitable for estimating the likely impact by type of borrower or activity. Such estimates will eventually be available on the basis of Quantitative Impact Studies.
THE DEVELOPMENT OF BASEL III: In December 2009, the Basel Committee on Banking Supervision (BCBS) issued a consultative document setting out proposals for strengthening regulation of banks' capital and liquidity (Basel III) in the light of lessons learnt from recent experience, especially that of the current financial crisis, with the goal of improving the resilience of the financial system ("Strengthening the resilience of the banking sector", Bank for International Settlements, December 2009). In the case of capital, the proposals build on the framework of Basel II as set out in the 2006 draft ("International Convergence of Capital Measurement and Capital Standards: A Revised Framework Comprehensive Version", Bank for International Settlements, June 2006). But Basel II has now been extended to include rules for the management of liquidity risk.
This extension is justified by the fact that crises or serious threats to banks' solvency (and thus to the adequacy of their capital) are typically triggered by pressures on their liquidity positions in the form of difficulties over financing their portfolios of assets. Thus, the measurement and management of banks' capital is indissolubly linked to the successful management of their liquidity, a link graphically illustrated by events during the current financial crisis.
In August 2010, the Basel Committee and the Financial Stability Board (FSB) issued two reports assessing the impact of Basel III. These followed the publication of estimates of this impact by other bodies, including one by the Institute of International Finance (IIF), an industry body of international banks.
One of the two official reports (of the Macroeconomic
Assessment Group of the FSB and the BCBS and referred to as the MAG
report in what follows) is concerned with the impact during the transition
period when the new requirements for capital and liquidity were being
phased in ("Assessing the macroeconomic impact of the transition
to stronger capital and liquidity requirements", Interim Report,
Bank for International Settlements, August 2010) and focussed exclusively
on the costs of introduction of the new requirements. The other report
of a working group of the BCBS analyses the long-term economic impact
(LEI) of Basel III
The conclusions of these two reports are that the costs in terms of lost output due to the changes in capital and liquidity requirements are likely to be moderate and less than those estimated by the banking sector itself in the parallel exercise of the Institute of International Finance (IIF, "Interim Report on the Cumulative Impact on the Global Economy of Proposed Changes in the Banking Regulatory Framework", June 2010). Moreover, the LEI report concludes that there will be significant benefits from these changes due to the lower incidence of financial crises and that these benefits outweigh the costs by a significant amount.
METHODS USED TO ESTIMATE IMPACT: For the estimation of the effects on output in the MAG and LEI reports, the new requirements on capital and liquidity were first translated into higher costs of intermediation (higher lending spreads) and then the impact of these higher costs on economic activity were estimated through macroeconomic models.
The MAG report provides a simplified example intended
to demonstrate the conceptual basis of the estimation. Imagine a bank
with the following stylised balance sheet: on the liabilities side there
are deposits and debt, on which the bank pays an average cost of 5 per
cent, and equity capital, on which the target return is 15 per cent.
Two-thirds of the bank's assets are loans and one-third are securities
and cash. Now, introduce a one-per-cent increase in the ratio of capital
to assets which raises the cost of funds (the weighted average of the
cost of capital, deposits and debt) by 10 basis points. To maintain
its target return on equity capital of 15 per cent, the bank must recover
its higher cost of funding by raising the rate of return on its assets.
In the stylised case, this is most easily done by raising the rate of
interest by 15 basis points on the loans which are two-thirds of its
The translation of the new rules on liquidity into higher costs of intermediation lacks the simplicity of that for increased capital. To meet the Liquidity Coverage Ratio of Basel III (which is designed to ensure that the banks hold a stock of unencumbered high-quality liquid assets sufficient to offset the net cash outflows encountered during a period of acute short-term stress), a bank is likely to increase holdings of low-yielding assets. To meet the Net Stable Funding Ratio of Basel III (the required amount of longer-term, stable sources of funding employed by an institution in relation to the liquidity profile of the assets in its portfolio and the potential contingent calls on funding liquidity due to its off-balance-sheet commitments), a bank may have to increase the average maturity of its liabilities. In both cases, the new liquidity requirements are assumed to exercise downward pressure on profitability and thus, upward pressure on lending margins.
ESTIMATED IMPACT: The MAG report's estimates of the deviations of GDP from baseline forecasts in the eighteenth quarter after the introduction of increased capital requirements and increased holdings of liquid assets are modest. For a one-per-cent increase in the capital ratio introduced with a transition period of four years, the median estimates (based on models favoured by the MAG) is a decrease of GDP of about 0.15 per cent, and for an increase in the ratio of liquid to total assets of 25 per cent, the median benchmark estimate is a decrease of 0.08 per cent. For an increase in the capital ratio to two per cent, this combined effect translates into a decrease of GDP of 0.30 per cent (or 0.38 per cent if the impact of the increase in the ratio of liquid to total assets is also taken into account).
The LEI report provides a range of estimates of costs and benefits according to different levels of the ratio of capital to risk-weighted assets and according to whether the capital rules are or are not accompanied by additional requirements for liquidity. A one-per-cent increase in the capital ratio is associated with a 0.09 per cent annual reduction of output if not accompanied by additional liquidity requirements, and with a 0.17-per-cent reduction of output if accompanied by additional liquidity requirements.
Expected net benefits in the LEI report - the difference between expected benefits as measured by the decrease in the annual probability of a crisis multiplied by the cumulative costs of a crisis, on the one hand, and the expected costs of new requirements for capital and liquidity, on the other - vary according to assumptions as to whether or not the crisis is assumed to have permanent effects on output. Where the crisis has no permanent effects on output, the expected benefits of a one-per-cent reduction in the annual probability of crisis multiplied by the cumulative resulting output losses are 19 per cent of pre-crisis GDP. Where there are assumed to be long-lasting but moderate effects - the case corresponding to the median cost of crises reported in all the studies in the literature surveyed for the LEI report - the expected annual benefits are 63 per cent of pre-crisis GDP. Where there are assumed to be large permanent effects - the case corresponding to those in studies allowing for permanent effects, the expected annual benefits are 158 per cent of pre-crisis GDP.
The corresponding estimates of long-run annual net benefits of a one-per-cent rise in the capital ratio are 0.20 per cent of output when there are no permanent effects, 0.87 per cent of output when net effects are long-lasting but moderate, and 2.32 per cent of output when there are large permanent effects.
The estimates of the costs of tighter rules on
capital and liquidity requirements of the MAG report are lower than
those of the IIF for the
TECHNICAL AND CONCEPTUAL LIMITATIONS OF THE ESTIMATES: The estimates of the MAG report are subject to various limitations: (1) the models on which they are based reflect the still-imperfect state of conceptualisation and econometric technique in this area; (2) the benchmark estimates do not allow for governments' policy actions intended to offset unfavourable effects of more stringent capital and liquidity requirements on the cost and volume of lending; (3) the estimates also do not allow for developments in financial markets in response to these requirements; and (4) the estimates do not incorporate the impact of changes in banks' portfolio management and business models which would help to sustain lending and keep down its cost.
The MAG report summarises the state of the art of modelling as follows: "Most central banks, and many other economic agencies, have one or more large-scale, regularly updated macroeconomic models that have over time demonstrated their usefulness for forecasting and policy analysis purposes. While time-tested and well understood, these models suffer from the fact they do not directly incorporate banking sectors in a way that would allow investigation of the impact of prudential policy changes" ("Assessing the macroeconomic impact of the transition to stronger capital and liquidity requirements", Interim Report, Bank for International Settlements, August 2010, p 14). In the work used for the MAG report, statistical relationships between liquidity requirements and lending costs were often weak, and for some countries included in the exercises, it was impossible to estimate the impact of the Basel III net stable funding ratio. Precise matching of the model-based estimates in the MAG report with the changes proposed in Basel III is impossible owing to the imperfect correspondence between the measures used in the models and the indicators which are the target of Basel III.
The benchmark estimates of the MAG report abstract from possible policy responses by governments to macroeconomic contractions resulting from the new regulations, from changes in the terms of banks' financing from financial markets, and from a range of possible responses of banks themselves to the new regulations. Differences between the estimates of the MAG and IIF reports can be explained at least in part by differences regarding assumed scenarios concerning the future.
Abstracting from a monetary-policy response to unwanted effects of more rigorous rules for capital and liquidity enables separation of the impact of the increased capital and liquidity requirements from other developments. However, such abstraction may also result in overestimation of the costs as compared to a more comprehensive scenario which allows for such a policy response. For example, in models which do incorporate an endogenous change in monetary policy to dampen the contractionary effects of a one-per-cent increase in capital requirements, the median estimate of the decrease in GDP in the MAG report falls to 0.06 per cent ("Assessing the macroeconomic impact of the transition to stronger capital and liquidity requirements", Interim Report, Bank for International Settlements, August 2010, p 22).
Amongst the developments in financial markets likely to dampen upward pressure on banks' financing costs, and thus, on the spreads on their lending, are the effects of more stringent capital and liquidity rules in reducing the required return on banks' equity via investors' improved perceptions of their soundness. Moreover, the increased demand by banks for government debt and other low-risk assets could well exert downward pressure on the rate of interest on such assets, which are the basis for banks' loan pricing (Bini Smaghi L, "Basel III and monetary policy", Speech at the International Banking Conference "Matching Stability and Performance: the Impact of New Regulations on Financial Intermediary Management", Milan, 29 September 2010, p 5).
The effects of increased capital and liquidity requirements on lending costs may also be affected by consequent adjustments in banks' portfolios and other aspects of their business models. These reactions may include a contraction of banks' trading books which leaves their lending business largely or completely untouched. For example, in the case of the two largest Swiss banks, a large part of the reduction of their balance sheets since the onset of the crisis in 2007 has been in the form of contractions of their trading books ("Assessing the macroeconomic impact of the transition to stronger capital and liquidity requirements", Interim Report, Bank for International Settlements, August 2010, p 55). Moreover, it is possible also to imagine banks holding down compensation costs, thus reducing pressure on lending margins, though banks' behaviour regarding staff remuneration since the avoidance of a global financial and macroeconomic meltdown thanks to the adoption by governments of bailouts and expansionary policies invites scepticism rather than optimism on this score.
The estimates of the LEI report are subject to qualifications similar to those of the MAG report. In its discussion of the net benefits of more rigorous capital and liquidity requirements, the LEI report emphasises that historical estimates of the costs of financial crises, especially in recent times, are influenced by large-scale government intervention to minimise the negative effects on output. In the absence of such intervention, the costs would probably have been significantly higher. Since expected net benefits are measured by the decrease in the annual probability of a crisis multiplied by the cumulative costs of a crisis less the expected costs of increased capital and liquidity requirements, it could be argued that the LEI estimates of net benefits are correspondingly underestimated ("An Assessment of the long-term economic impact of stronger capital and liquidity requirements", Bank for International Settlements, August 2010, p 31).
The higher estimates of the IIF study as compared with the MAG report reflect differences in the modelling and conceptual framework used as well as the country coverage (see below). The IIF estimates are also intended to cover the effects of a more extensive array of specific reforms including a tighter definition of what financial instruments would qualify as capital under Basel III, higher capital requirements for market (as well as credit) risk, and a supplementary counter-cyclical capital buffer (IIF, "The Net Cumulative Economic Impact of Banking Sector Regulation: Some New Perspectives", October 2010, pp 20-21).
SCOPE OF THE EXERCISES: The applicability of estimates of the costs and net benefits of increased capital and liquidity requirements should also be interpreted in the light of the scope of the exercises on which they are based. Major determinants of this scope are: (1) the relation between the models' specification and the conclusions which can be drawn from them; and (2) the sources of the data and models used for the estimates.
The estimates in the MAG and LEI reports are limited to the macroeconomic impact of increases in capital and liquidity requirements, and do not differentiate between countries and economic sectors. A better picture of the impact at the latter level will eventually be provided by the results of a new Quantitative Impact Study which is compiling information on the effects of the proposed new capital and liquidity standards for a sample of banks from a number of countries ("Assessing the macroeconomic impact of the transition to stronger capital and liquidity requirements", Interim Report, Bank for International Settlements, August 2010, p 35).
Regarding the global validity of the estimates in the MAG and LEI reports, it should be emphasised that the data and models used are from a sample of developed countries and of a minority of the more advanced emerging-market countries. A precise picture of the coverage of the data and models cannot be extracted from the reports - presumably because such a picture would have required unmanageably long technical annexes.
Model inputs to the MAG report were provided for
the following emerging-market countries:
By contrast, the IIF estimates were based on only
three models - for the
The revision of the capital and liquidity requirements of Basel III has so far reflected principally regulators' response to weaknesses in the existing framework (including weaknesses in rules of Basel II), which have been revealed by experience of the financial crisis in developed countries. Past experience suggests that as a result of the consultative process associated with the drafting of a new text for Basel III, qualifications will be inserted with the objective of adapting some of the rules to circumstances commonly found in countries with less developed economies and banking systems.
Moreover, since Basel III will not be mandatory, the authorities at national level will also be able to include adjustments to the rules to better fit them to national circumstances and needs, and to provide guidelines for supervisory implementation which include the same objective. The scope for such adjustments should help to ensure that introduction and implementation of the new rules of Basel III in developing countries is not at the expense of activity levels and of the attainment of other important developmental objectives.
(* Andrew Cornford, formerly a senior economist
at UNCTAD, is currently Counsellor at the Observatoire de la Finance,