The rules of the road making the financial system safe for everyone

Speech by Mr Stephen G Cecchetti, Economic Adviser and Head of Monetary and Economic Department of the BIS at the CFA Institute, Paris, France, 03 November 2011.

[1]Thank you for inviting me to speak to you today. It is an honour and a privilege to be a guest of the CFA Institute this morning.

Today I would like to take you on a road trip. Try to recall the worst road trip you have ever experienced. Maybe it was leaving Paris for the South on 1 August. Or, possibly you were driving on the périphérique when there was an accident. Some of you will remember the scene with an endless line of cars in the classic 1967 Godard film Week-end. For me, it is the childhood experience 40 plus years ago of driving around Italy with my family. We would be cruising along the autostrada, when suddenly everyone would completely stop. Emergency vehicles rushed by, sirens blaring, on the shoulder of the road as we watched. I found this entertaining. My parents did not. And when, an hour or two later, traffic started moving again, we would eventually drive by some wreckage - a burned out hulk of a car or truck - that had been pushed off to the side of the road.

The technology of both highways and cars has improved since I was a boy. And the capacity of roads to carry traffic at high speeds has gone up with it. But, even so, when things go wrong, they can still go very badly wrong.

Imagine a road designed for high speed completely filled with cars and trucks. Suddenly, a truck carrying 30,000 litres of gasoline goes out of control, crashes and explodes. Everything nearby goes up in the ensuing flames, and the closely packed cars behind the crash pile one into another as the drivers are powerless to stop.

This is the global financial crisis: by far the largest financial conflagration in living memory. But, unlike the highway accident, where the driver of the gasoline truck would almost certainly perish for his carelessness, many managers of the financial institutions were miraculously transported to safety under the protection of their governments. And those who were forced to resign deployed their golden parachutes to get out.

We are rewriting the rules of the road to prevent such an accident from recurring. Our new rules take into account the fact that, while every vehicle potentially poses risks to other drivers, some pose more risks than others. In particular, trucks, especially the heavy ones carrying the most dangerous loads, create externalities. That is, their actions impose costs on others that they themselves do not face. And, neither the caution of the truck drivers nor the safety features in the cars travelling around them are enough to protect people from these behemoths on the road.

To address the problem on the road, we could simply ban trucks. But we like the services they provide; transporting large quantities of goods cheaply for long distances. Instead, what we do is impose tougher safety standards on the trucks and their drivers. They face more frequent inspections, more stringent equipment requirements, tougher emission standards and higher tolls. In some places they must abide by lower speed limits and they are not even allowed on certain roads. Truck drivers have stricter licensing requirements. And, depending on the cargo being transported, there are additional restrictions. These rules are in everyone's interest. They reduce the frequency and severity of accidents, forcing the trucks' owners and drivers to pay the costs of the externalities they create.

In the same way that regulating vehicle safety is in the interest of drivers of both cars and trucks, new financial regulation is in everyone's interest. Having a safe financial system not only benefits the owners, managers and customers of the banks; it makes it possible for pension funds, insurance companies, hedge funds, private equity firms and individual investors to go about their business. And, in the same way that we pay especially close attention to large trucks carrying hazardous cargo, we need to be particularly careful about how we treat systemically important banks.

Now, the drivers of some of the big trucks are arguing that the new rules that will apply to them are too tough - that they are anti-truck. Not only will they hurt us, they say, but they will hurt our owners, and most importantly, they will hurt our customers. But these drivers are forgetting that they are not on the hook for all the damage that their mistakes and recklessness can cause everyone else on the road. The drivers of the trucks forget that they do not pay for evacuation and care of the injured, or the repair and replacement of damaged vehicles. Instead, the public, through the taxes it pays to governments, foots the bill.

While the financial pile-up had many causes, we must not overlook the critical role played by a number of very large global banks with insufficient capital, inadequate liquidity, and poor risk management practices. If these banks had had enough capital to absorb the losses coming from their unwise investment decisions in collateralised debt obligation tranches, and enough liquidity to meet their unwise commitments to commercial paper vehicles, just to give two examples, then things would have turned out very differently. Instead, the result was a severe global recession and a clean-up that we are all going to be funding for at least a generation.

These costs are the consequence of weak regulatory rules, lax supervisory oversight, poor managerial decisions, and a lack of sound resolution procedures. We know that the quality of regulation and supervision mattered, since the banking systems of some countries proved to be more resilient to the crisis than others. We know that some banks made better decisions than others, since those that went into the crisis with higher levels of capital, more liquidity and better risk management processes survived the ordeal of the last few years without having to call on public support. And we know that weaknesses in resolution frameworks played a role since public authorities in some instances were forced to choose between letting institutions fail, further disrupting the financial system, or rescuing them with public funds, socialising the losses and aggravating moral hazard.

The new framework for regulating and supervising systemically important financial institutions (SIFIs) currently being developed by the Financial Stability Board (FSB) and Basel Committee on Banking Supervision (aka the Basel Committee) is informed by this painful history. It comprises three complementary components: greater loss absorbency, more intense supervision and stronger resolution. We know that the distress or failure of certain institutions has a greater impact on the system than the distress of others. Addressing this externality means insisting that certain institutions both have more capital to absorb losses and be subject to more intense supervisory scrutiny. Further, we need to make it possible to close or restructure such institutions, even those with significant cross-border activities, without massive disruption to the rest of the financial system. Doing all of this allows us to pledge, credibly, that no institution is too-big-to-fail. This, in turn, should help make the system as a whole more efficient, levelling the playing field between large and small financial institutions.

I should note that, while most of the recent discussions have related to systemically important banks, there are other financial institutions that are potentially systemic, such as insurance companies, asset managers and providers of market infrastructure. It is essential that regulation and supervision be strengthened for all of these. And here, the principle is the same as for banks: the owners and managers must be forced to face the costs of the systemic risks that they create. A combination of higher loss absorbency, streamlined resolution and more intense supervision must reduce both the likelihood and impact of failure for both bank and non-bank financial institutions. But this will only work if market participants take their responsibility seriously, doing the monitoring that is required. The market discipline arising from such vigilance was sorely lacking in the build-up to and the early stages of the crisis.

The measures agreed by the Group of Governors and Heads of Supervision, the oversight body for the Basel Committee, represent an important step forward for this agenda. The methodology developed by the Committee to identify global systemically important banks (G-SIBs) is based on size, interconnectedness, substitutability, global activity and complexity. G-SIBs are then allocated into buckets according to their relative systemic importance. The proposal is to allocate banks to four buckets and require additional loss absorbency comprising common Tier 1 equity ranging from 1% to 2.5% of risk-weighted assets. In addition, as a disincentive to a G-SIB becoming even more systemically important, an initially empty bucket with a capital surcharge of 3.5% sits at the top.

To understand the implications of this, take the example of a bank that incurs a 2% surcharge. Such a bank would face a 4½% minimum, plus a 2½% conservation buffer, plus the 2% surcharge, making a total common equity requirement of 9%. National supervisors can use their discretion to add an additional countercyclical buffer of up to 2.5% under conditions when excess credit growth seems to be promoting a build-up of systemic risk. Taking into account Basel III's tougher definition of capital, the result is a substantial and necessary increase in minimum capital requirements.

Returning briefly to my highway metaphor, all the traffic safety engineers have said that, while minimum standards should be raised for all trucks, the safety requirements for any specific one depend on how big it is and how dangerous its load. The bigger the truck and the more hazardous its cargo, the safer it has to be. And, some trucks simply pose too big a risk for others, so they need to be taken off the road.

Where did these figures come from? Let me very briefly outline two methods to answer this question. One starts with the estimate of the real economic impact of a SIFI failure as between three and five times that of a non-SIFI failure. Next, note that reducing the probability of failure to the point where the expected loss - the probability times the impact - is the same, would require additional capital in the range from 2% to 8% of risk-weighted assets.[2] The agreed minimum standard is at the low end of this range.

The second approach to determining the surcharge builds on the long-term economic impact study[3] that examined the benefits and costs associated with the Basel III capital and liquidity requirements. That study concluded that, since financial crises are likely to have a permanent impact on growth, the balance of the costs and benefits of higher capital ratios (using the new definition of capital), peaks at a level of between 10% and 11% - 3 to 4 percentage points above the 7% Basel III minimum.

Some jurisdictions have announced their intention to impose requirements above this minimum. This is entirely appropriate, as some national banking systems are very large relative to the rest of their country's economy. That is, in some places, banks are not only too-big-to-fail, they are also too-big-to-bail.

I should also mention that Basel III introduces a leverage ratio as a backstop to the risk-based measure. The gains from this are twofold. First, it can help contain the build-up of systemic risk that can arise when leverage expands quickly. And second, it guards against the mismeasurement of risks during booms. Not to belabour the point, but it is essential to keep in mind how difficult it is to measure and control risk, even for the most technically skilled and well-intentioned bankers and supervisors. These systems are built by people, so they can never be entirely fail-safe. But having several layers of protection for something this important seems like a fairly basic precaution.

Let me now take a moment to address the industry's criticism of the improved capital and liquidity requirements head on; the complaint that imposing stricter standards on trucks is bad for commerce.

Some critics point to the possible unintended consequences of higher capital and more liquidity. My immediate reaction is to start by reminding everyone that the crisis was a result of bankers lending money they shouldn't have lent to borrowers who shouldn't have borrowed. Financial professionals sold safe assets that weren't safe. And investors bought assets without a proper appreciation of the risks. At first profits were high and salaries astronomical. But then the losses were even bigger.

Now, the bankers would have you believe that the new regulations are preventing them from conducting the normal business of banking. They complain that, if it weren't for the regulators, they would be making more loans.

I simply do not accept this characterisation. First, the new regulations have not even started to be phased in. It will be years until the process is complete. Second, as I just mentioned, excessive leverage is what got us into this in the first place. But even so, borrower demand is almost certainly a bigger problem than lender supply. And third, it is the markets that are rightly putting pressure on the banks.

We need a safer, more resilient financial system. And we need it now. We need prudent lending, prudent borrowing and proper incentives. We need safe, responsible banking. These are the intended consequences of the regulation.

Going beyond their focus on unintended consequences without acknowledging the intended consequences, the critics also claim that the new proposed regulations will precipitate some sort of economic Armageddon. The recently released report from the International Institute of Finance (IIF)[4] claims that the cumulative impact of the changes will result in an increase in lending rates of at least 300 basis points and that banks will need to raise $1.3 trillion in new capital. According to the IIF, the macroeconomic consequences will be catastrophic. GDP will fall by 3% over the next five years, resulting in a cumulative 7 million jobs lost.

Estimates of this size are not only completely outside historical experience, they are also totally at odds with conclusions that come from the Macroeconomic Assessment Group (MAG), which I chair. Our work, based on simulations by experts from 15 countries and a number of international institutions from nearly 100 models, yields estimates for both the increase in lending spreads and the impact on GDP that are roughly one tenth the size of those reported by the IIF. [5]

We all agree that the key drivers behind any macroeconomic impact are the increased need for capital and the knock-on effect of wider lending spreads. On the cost of lending, the IIF bases its forecasts on a predicted increase in lending spreads of more than 300 basis points. I find it useful to contrast these claims with the banks' current complaints about their inability to raise lending rates in an environment of tremendous competition; competition, that is, for the creditworthy borrowers who are looking for funding. Won't competition keep credit spreads down? I would think so. On the need for capital, the IIF's results assume banks will need to pay a high premium to issue new equity externally. But banks have already begun to build up their capital positions through retained earnings. And, given the lengthy transition period in Basel III, it seems apparent that many banks will be able to reach the target with less costly internal funds and little or no new equity issuance at all.

To understand why I believe that changes in the capital and liquidity regulation will have only a modest impact on credit availability, just think about who could pay for the additional costs associated with building more resilient banks. There are four groups: those receiving loans from the bank (in the form of higher lending rates), those making deposits or holding bonds issued by the bank (in the form of lower deposit rates), the bank's owners (in the form of lower return on equity), and the bank's managers and other employees (in form of lower salaries and bonuses).

Under a regime of stronger regulation, the bank's shareholders should be content with lower average returns that are associated with steadier returns that do not take a dive in an economic downturn. The same goes for the bondholders. And what about bankers' salaries? If shareholders are serious about protecting the value of their investments, they should pressure firms to end the adverse cycle where high earnings turbocharged with leverage periodically turn into painfully sub-par earnings, dividend cuts, and dilution as new shares are sold at the worst time. Thus, assuming that shareholders are ready to accept reasonable and steady returns, the macroeconomic impact of higher capital will be very small.

Beyond the likelihood that spreads and the return on equity will be held down by a combination of lower risk and competition within the banking industry, there is the observation that, if spreads were to rise by 300+ basis points, lending would simply move outside the traditional banking system. I expect that in such an environment, market-based finance - commercial paper, securitised lending and the like - would come roaring back and even overtake the levels achieved before the crisis. These financing channels - sometimes referred to as the shadow banking system - raise systemic risk issues of their own, but most of them can be mitigated as long as we make sure that banks manage their exposures properly. Not only that, but the mere threat of such competition will surely place some very strict limits on the activities of the banks themselves.

In the end, I am convinced that the benefits of the new capital and liquidity standards far outweigh the costs both during the transition and in the long run. The intended consequences of Basel III - namely a stronger and more resilient financial system - will benefit us all.

Returning to my highway metaphor, in the same way that it is in every driver's interest that trucks drive safely, it is in everyone's interest that banks behave responsibly and safely. Everyone, every asset manager, every individual investor, even every small saver with a bank account, should not only support the more stringent requirements that we are working to put in place, they should insist on them. In the same way that none of us would want to drive on a road populated by reckless truck drivers, we should all refuse to do business with banks and bankers that do not meet the highest possible standards.

Financial analysts like all of you in the audience this morning, whether sell-side or buy-side, have an important role to play in this. You must push institutions to behave more responsibly. Let me give you two examples where investors must play a role: leverage and sovereign risk.

Bank regulation outside the United States and Canada left banks able to leverage up without limit, at least in terms of the ratio of unweighted assets to equity. But that did not mean that the banks' risk managers and senior managers had to or should have done so. Just because something is legal doesn't make it wise. Should not investors in bank shares, and the analysts who guide them, have responded to rising leverage ratios? US securities firms reached ratios well into the 30s in 2007, while some banks in Europe found themselves in the 40-50 range. And I mean "found themselves". In the case of UBS, which reported in 2006 that almost CHF 2.4 trillion in assets was supported by CHF 50 billion in shareholder equity, its subsequent report to shareholders made it clear that management was not really managing the overall size of the balance sheet. [6] Should bank stock equity analysts and buy-side analysts have been equally nonchalant?

Another example is the management of sovereign risk. Under Basel II, big banks were supposed to use their own internal modelling to differentiate risks. Only banks using the standardised approach for credit risk as allowed to use external rates to assign weights to sovereign exposures. This includes zero weights for AAA- and AA-rated sovereigns. And yet the European stress test reported that most of the 90 biggest banks in Europe have been using the standardised approach to assess sovereign risk instead of their own estimates of key elements like the probability of default and the loss given default. Indeed, there was a certain amount of cherry-picking going on, with a higher fraction of banks choosing the standardised approach to sovereign risk than was the case for other classes of exposure. [7] Such behaviour was prohibited by some authorities. But even if the authorities allowed this, shouldn't management, especially the chief risk officer, have insisted on using the results of his own credit evaluation? And shouldn't sell-side and buy-side analysts alike have insisted that banks do their homework on this?

As an aside, I think it is interesting to note that the Basel Committee has embarked on an aggressive campaign to ensure that implementation of its rules at the local level meets the letter and intent of the globally-agreed standards. This review will also assess banks' risk weightings.

Let me conclude by harking back to the founder of the CFA certification, Benjamin Graham. Would he want to invest in a big bank as we have come to know them, one that uses measurable and unmeasurable leverage to crank up return on equity in boom periods only to dilute the shareholders, if not hit up the taxpayers, in bad times? We can look at the portfolio of Graham's student, Warren Buffett, for an answer. Why are big bank shares not "permanent" holdings next to major beverage makers, utilities, railroads, oil companies and the like in the Berkshire Hathaway portfolio? The answer is that, from Buffett's point of view, the industrial companies behave in a more predictable way than the financials.

The Basel III reforms in general and those for SIFIs in particular will succeed if, 20 years from now, you analysts can look back on a prolonged period of sustained, predictable, even boring profits in the financial sector. This strikes me as the responsibility of bank managements to their shareholders and an outcome that we should not only expect, but  demand.

In the end, like safe roads, a safe financial system is in the interests of everyone involved. And, in the same way that safe roads carry a higher volume of traffic - both people in cars and cargo in trucks - a safe financial system will help generate higher stable economic growth.

Thank you for your attention.