Just like an overweight victim recovering from a severe heart attack, the financial system must change its ways. After working tirelessly – and in the end successfully – to stabilise the patient, the world’s central bankers and supervisors are developing a rigorous diet and exercise programme to help avoid a relapse. Yet, now that the immediate danger has passed, a natural scepticism has set in. Does the financial system really need to change its ways? Why bother with all this unpleasant exercise? Are the benefits of more stringent regulation and supervision really worth the cost?
Two papers released this week by the Financial Stability Board and the Basel Committee on Banking Supervision give a clear answer. Stronger capital and liquidity requirements bring substantial benefits – and only with modest costs. The first study (Basel Committee on Banking Supervision 2010) looks at the long-term impact of stronger capital and liquidity requirements, while the other (Macroeconomic Assessment Group 2010) examines the transitional economic impact as the new standards are phased in. Taken together, these two reports show that the benefits will be significant, while the costs are likely to be very modest.
The Basel Committee on Banking Supervision and The Financial Stability Board are well advanced in the preparation of detailed internationally agreed financial reforms. Improved capital and liquidity standards form the core of these reforms. They outline that banks should:
- increase and improve their capital base;
- set a simple leverage ratio as a supplementary backstop to the risk-based capital framework;
- adopt capital conservation measures to generate buffers that can be drawn down in periods of stress;
- and – in contrast to their behaviour before the financial crisis – manage their maturity and currency mismatches in a prudent manner.
The motive for these specific reforms – as well as capital regulation in general – is that if banks are left to their own devices, they will hold too little capital and liquidity. While a lower level of capital may result in higher returns to equity holders, it will also mean a smaller buffer to weather loan defaults and investment losses. Less liquidity meanwhile, implies not only a higher fraction of long-term assets funded with short-term debt that raises interest rate margins and profits, it also makes banks more exposed to sudden withdrawals and difficulties in rolling over debt. As we have learned at no small cost, the upside of these risks belong to banks’ shareholders and managers, while the size of capital and liquidity cushions determines how much of the downside risk is borne by all of us.
Serious financial crises occur every 20 to 25 years; the average annual probability of a crisis is of the order of 4-5%. When they do occur, banking crises are often quite costly, bringing with them deep recessions and a permanently lower GDP. Estimates of the cumulative (discounted) output losses range from a minimum of 20% to well in excess of 100% of pre-crisis output, depending primarily on how long-lasting the effects are estimated to be. A study by the Basel Committee on Banking Supervision (2010) using the median estimate of the costs of crises across all comparable studies – around 60% – finds that each percentage point reduction in the annual probability of a crisis yields an expected benefit per year equal to 0.6% of output. While individual country experiences obviously vary, on balance the frequency of crises does not differ much between industrial and emerging economies and, if anything, costs appear somewhat higher in industrial economies.
The ups and downs of a better diet
Although there is considerable uncertainty about the exact magnitude of the effect, the evidence indicates that higher capital and liquidity requirements can significantly reduce the probability of banking crises, though with declining marginal benefits. Furthermore, the stronger the banking system the less severe a banking crisis is likely to be, should it come. Higher aggregate levels of capital and liquidity should help insulate stronger banks from the strains faced by weaker ones. It is with the aim of reducing the frequency, severity, and public costs of financial crises that the G20 leaders are committed to significantly increasing the levels of capital and liquidity in their national banking systems.
While a better diet and more exercise will increase life expectancy, what about the hunger and soreness the patient experiences as he starts the new programme? Even if the long-term benefits are recognised, what about the short-term costs of implementing a strong regulatory system?
This is the question addressed by the Macroeconomic Assessment Group, a group that draws together the modelling expertise of two dozen national authorities and international organisations. The report concludes that, for each percentage point increase in required capital implemented over a four year horizon, the level of GDP relative to the baseline path declines by a maximum of about 0.19%. This maximum GDP loss occurs four and a half years after the start of implementation; after which GDP recovers towards its baseline level. The 0.19% figure is the sum of 0.16%, the median GDP decline estimated for specific countries by national authorities, and 0.03%, which is the potential impact of international spillovers (reflecting exchange rates, commodity prices and shifts in global demand) as estimated by the IMF.
Putting this into a more commonly used terminology, we can translate the decline of 0.19% and the subsequent recovery into implied changes in the average annual growth rate during the transition. First, note that with a four year implementation period, the maximum deviation from the baseline occurs after 4½ years. In order to reach a cumulative 0.19% fall in the level of output in this time period, growth would have to be roughly 0.04% below trend. Following this, the economy slowly recovers with growth 0.02% above trend.
No pain no gain?
These results are for every one percentage point increase in target capital ratios, whether the source of this increase be higher regulatory minima, required buffers, changes in the definition of capital, the application of a leverage ratio, or some other change in standards. And, importantly, the numbers are scalable. So, the report’s results imply that a two percentage point increase in target capital ratios would bring with it a fall in growth of 0.08% at an annual rate relative to trend for 4½ years, followed by annual growth of 0.04% above trend as the recovery ensues.
As with any policy simulation or forecast, the conclusions of the analysis depend on a variety of assumptions about behaviour. For example, most of the effects reflect the presumption that higher capital levels will increase bank funding costs, and these will be passed on to borrowers in the form of higher loan rates. But, instead of (or in addition to) raising loan rates, banks may choose to shrink their balance sheets, cutting back on lending – rationing credit – in ways that wouldn’t be reflected in lending rates. When some modellers in the Macroeconomic Assessment Group studied this possibility, their work suggested more substantial output effects.
An easing of monetary policy reduces the estimated output losses. When it is assumed that the central bank responds to the incipient aggregate demand fall and reduced inflationary pressures precipitated by the regulatory changes, the central estimate of the maximum output loss shrinks significantly. Such offsets are especially pronounced in models that incorporate credit supply constraints, for which the GDP loss at the four-and-a-half year point falls from 0.32% to 0.17%.
The adjustment costs of implementing higher capital requirements arise from the fact that equity investors – suppliers of the bank’s risk capital – require higher returns than depositors. So, higher capital requirements mean higher funding costs; costs that banks will try to recover by raising loan rates they charge borrowers or shrinking their balance sheets.
But banks can also meet the new requirements by some combination of retaining more earnings (reducing dividends), decreasing remuneration rates on deposits and other liabilities, or reducing other costs of doing business (including managerial compensation). It is also possible that, as banks become safer, markets will require a lower return on equity, tempering the pressure on funding costs. Any of these would reduce the degree to which lending rates would need to rise and loan volumes fall, attenuating the macroeconomic impact.
Indeed, Elliott (2009), who estimated that each percentage point increase in capital ratios would lead to a 19 basis point widening in lending spreads if other factors were kept constant, concluded that the widening of spreads could be as low as 4.5 basis points if banks’ return on equity, asset mix and other variables are allowed to change. Hanson et al. (forthcoming) arrive at a similar estimate for the likely widening of lending spreads. They suggest that the only impact of capital and liquidity requirements would be due to effects that are not encompassed in Modigliani and Miller’s (1958) idealised framework, notably the tax advantages of debt and the premium that banks would need to pay investors to hold longer-term, less-liquid debt instruments.
While acknowledging the potential importance of these factors, the Macroeconomic Assessment Group implemented a conservative approach by assuming banks could not make these adjustments. Focusing on the implications for borrowing costs, Macroeconomic Assessment Group researchers estimated that banks’ lending rates would need to rise by about 15 basis points for each percentage point increase in capital1. With companies needing to pay more to borrow, they will start fewer new investment projects, slowing the level of overall economic activity. Yet, as the financial system adjusts during and after the phase-in, these costs will slowly dissipate, returning GDP to the path it would have followed in the absence of the changes.
This analysis assumes the higher capital requirement will be phased in over a four year period. If the implementation is faster, say two years rather than four, then the estimated impact on GDP is likely to be both larger and earlier. The estimated maximum decline in GDP is 0.22% rather than 0.19% and comes two years earlier. In terms of growth rates, a two-year implementation implies that (per percentage point increase in target capital ratios) growth will be 0.09% lower for 2½ years rather than 0.04% lower for 4½ years.
By contrast, the report concludes that lengthening the implementation period from four to six year makes little difference in the maximum decline. (Of course, if a similar sized cumulative loss is spread over a longer implementation period means a smaller decline in annual growth, relatively to trend, during the transition.) In any event, most of the factors not considered by the modelling group, like limits to the market’s capacity to quickly absorb large volumes of new bank equity issuance, would argue for a longer rather than a shorter transition period.
Separate analyses looking at the impact of stronger liquidity standards (which would oblige banks to hold more liquid assets and rely less on short-term wholesale funding) show that these, too, are likely to have only mild transitional effects. This is at least in part because of the complementarity of capital and liquidity requirements. That is, they call for similar changes to banks’ asset and liability structures – so, as one rises, the required level of the other falls.
Studies published by banks come to different conclusions. For example, in June the Institute of International Finance (2010) issued a report concluding that phasing in an increase in capital requirements of as little as two percentage points will lead to a drop of GDP of 3% in the G3 economies. Why such a big difference? One reason is that the industry studies assume a much larger increase in the lending rate, largely reflecting the withdrawal of implicit government support. But as the industry studies also acknowledge, they have assumed no changes in dividends, compensation policies and operational efficiency, nor have they taken account of the benefits coming from a more resilient financial system, including the lower funding premia that safer banks need to pay.
A clear lesson of this financial crisis is that the safeguards that were place were too weak. But reinforcement is not free. Fortunately, the short-term costs are likely to be small and transitory, while the benefits of a stronger and healthier financial system will be around for years to come.
Basel Committee on Banking Supervision (2010), An assessment of the long-term economic impact of the new regulatory framework, August.
Elliott, D (2009), “Quantifying the effects on lending of increased capital requirements”, Pew Financial Reform Project Briefing Paper 7.
Institute of International Finance (2010), Interim Report on the Cumulative Impact on the Global Economy of Proposed Changes in the Banking Regulatory Framework, June.
S Hanson, A Kashyap, and J Stein (forthcoming): “A Macroprudential Approach to Financial Regulation”, Journal of Economic Perspectives.
Macroeconomic Assessment Group of the Financial Stability Board and Basel Committee on Banking Supervision (2010), Assessing the macroeconomic impact of the transition to stronger capital and liquidity requirements, Bank for International Settlements, August.
Modigliani, F and M Miller (1958), “The cost of capital, corporation finance and the theory of investment”, American Economic Review, 48:261–297.
1 To understand where a number like this may come from, imagine a stylised bank with a balance sheet (where total assets equal risk-weighted assets) that has the following composition. On the liabilities side, there are deposits and debt, for which the bank pays an average of 5%, and capital, with a return of 15%. Assets are composed of two thirds loans and one third a combination of securities and cash (reserves). Now consider an increase in the capital ratio of 1 percentage point. This raises the cost of funds (the weighted average cost of capital plus deposits and debt) by 10 basis points. To maintain return on equity at 15%, the bank must recover this cost increase by raising the return on its assets. If this is done solely by raising rates charged to borrowers, since loans are two thirds of assets, it must raise lending rates by 15 basis points – a number close to that in the MAG report and obtained by Elliott (2009).