Some Perspectives on Regulatory Reform Proposals
Thank you for that kind introduction and for making me feel welcome in Hong Kong. Though I've been here only a short time, I find it easy to see why this unique city has become so important to the financial services industry throughout the world. I'd also like to thank Christopher Rogers and the Institute of Regulation & Risk for their kind invitation. It was that invitation that prompted my thinking in putting together this two-week visit to China. I know other Federal Reserve presidents have participated in this event in the past, and I am pleased to add my name to that list and the list of distinguished panelists who have joined us here tonight.
We are slowly emerging from the worst financial crisis since the 1930s. The hardships created by these exceptional circumstances for households and businesses are well known. Governments and regulators around the world have responded to the crisis with a variety of aggressive and innovative policy actions, including giving special assistance to specific institutions.
Now, as we slowly emerge from the crisis, we are engaged in a vigorous debate on how best to address the major weaknesses in our financial regulatory framework that were revealed by the crisis. Our goal, clearly, is to avoid another crisis of this magnitude. Financial reform will not be easy. We face complex problems that will require a comprehensive, multi-pronged approach. But reform is critical for ensuring our long-term economic stability.
Today, I would like to offer my thoughts on some of the reform proposals that are being discussed. I should note that my remarks reflect my own views and are not those of the Federal Open Market Committee or the Federal Reserve System.
To highlight some of the changes that are being considered, there are proposals that would assign monetary policy a more active role in fighting asset bubbles; proposals that would strengthen current microprudential regulations; proposals that would introduce a systemic regulator and macroprudential regulations; and proposals that would create resolution authority – particularly for systemically important financial institutions.
Time does not permit me to discuss the specifics of each of these proposals today. Instead, I would like to offer my thoughts on some of the challenges we are likely to face in implementing even the most well-thought-out policies.
Let me be clear. I don’t bring up these potential challenges as roadblocks to the healthy debate that is underway. Rather, I offer them as issues we need to consider as we build a better financial infrastructure.
One preemptive action that is being debated concerns the role of monetary policy in combating asset bubbles. Given the rapid rise in some asset prices prior to the recent crisis, there are increasing calls for central banks to be more proactive in responding to signs that an asset bubble may be emerging and to raise their target rates in order to lower asset prices that, by historical standards, seem unusually high. In previous forums, I have discussed why I view these proposals with skepticism.1 I won’t cover the same ground here again. Instead, let me just note that I am skeptical about our ability to easily and definitively sort out in real time whether a rapid increase in asset prices is associated with overvaluation. That is, how confidently can we state that we are in the midst of a bubble? I also think that monetary policy is too blunt a tool for pricking bubbles: It can’t be targeted precisely and it will affect other financial and macroeconomic variables in addition to the suspected bubble asset. In addition, the typical changes in interest rates that a central bank might contemplate are likely to be too small to produce big changes in asset prices.
Fortunately, monetary policy is not the only tool that policymakers have to deal with financial exuberance. In my view, redesigning regulations and improving market infrastructure offer more promising paths. Regulation may or may not be sufficient to avoid all of the events that create crises, but it should go a long way toward doing so. Better supervision and a sound regulatory infrastructure can also increase the resiliency of markets and institutions and their ability to withstand adverse shocks that do occur.
How do we promote such resiliency?
Within the existing structure, regulators have the ability to promote better, more resilient financial markets, either through rule-making or by serving as a coordinator of private initiatives. They can also encourage more and better disclosure of information—a key element of effective risk management. A number of initiatives along these lines have been taken and additional ones are being considered.2
We can use existing regulatory tools more effectively, but we also need to address the shortcomings of current regulations. The ongoing work of the Basel Committee on Banking Supervision regarding the possible introduction of liquidity standards and adjustments to the existing capital requirements are examples of such efforts.
While such enhancements to micro-prudential regulations are necessary, I would argue that they are not close to being sufficient to address the complex issues we faced during the recent crisis. Success in preventing and controlling potential risks requires very early and courageous action by policymakers. Typically, risks and problems in the financial system build over a number of years. There is an awful amount of uncertainty as to whether risks are developing; how they will be perpetuated; and when to take action. Microprudential regulations alone are not likely to resolve these issues.
Let me illustrate the sort of problems a microprudential regulator faces with a specific example. As you know, the problems with residential mortgages, particularly with subprime mortgages, were one of the key areas that precipitated the current crisis. Currently, the U.S. financial system faces problems with commercial real estate (CRE) loans. At the end of 2009, depository institutions in the U.S. held over $1.5 trillion in commercial real estate and construction loans on their books. In addition, there are currently nearly $800 billion in commercial mortgage-backed securities (CMBS) outstanding. Over the past two years, delinquencies on these loans and securities have been rising at an uncomfortably rapid rate. Figure. 1 Of the over $1 trillion in commercial real estate loans held by depository institutions today, nearly 4 percent are noncurrent.3 This ratio was about 0.5 percent before the crisis (June 2007). The picture is even worse for the riskier construction and land development loans. While these loans total less than $0.5 trillion, the noncurrent portion had risen from 1.5 percent at the end of June 2007 to nearly 16 percent by the end of last quarter. For CRE loans packaged into securities, serious delinquencies represent 4 percent of all CMBS currently outstanding, up from nearly zero before the crisis.
Does such a fast rise in CRE and CMBS delinquencies mean that bank examiners missed clear signs of forthcoming problems and failed to take action? Commercial real estate loans are a key problem area for the banks in my District. I went to my supervisory staff and said: “I know you are struggling with commercial real estate loan portfolios. What are the difficulties? What do you think we needed to have done in the past in order to avoid the current problems?”
I have to admit that their response made me pause. They said “You know, Charlie, if we wanted to avoid the current situation, we needed to act very, very early – probably in 2004 or 2005.” That is a full two to three years before the onset of problems in the sector. Clearly, we needed to act very early. But at that time, it would have been difficult to argue convincingly in favor of reigning in this lending. The economy was coming out of the jobless recovery and just beginning to gain traction. And the banking industry had proven it could maintain profits through a recession, it had reduced problem loans back to historically low levels, and it appeared to have more than sufficient capital to cushion against potential losses.4
Given previous problems with commercial real estate loans, my supervisors understood the potential risks. Here is a typical situation they faced. Imagine you are an examiner and you go out to review a large financial institution in 2005. The institution is warehousing commercial real estate loans prior to securitization during a period when CMBS issuance is just taking off and, for every $1,000 in CRE loans, only $6 are noncurrent. Nonetheless, as an examiner, you have a discussion with the bank managers and you learn about their lending practices, and you kind of wonder, “How well-controlled is all of this?”
The loan officers will give you some very good arguments about what their business is and how the risks are being controlled. First, they are not really storing the loans on their books. They are underwriting the loans with the full intention of packaging them into securities. They have to build up a critical mass before securitization, but they are not going to keep the loans.
From the banks’ perspective, they are not in the storage business. They are in the transportation business. It is rather short term – 60, 90 days. Presumably, the risk is only proportional to how long they are holding on to it – which is not very long. Furthermore, during this period, real estate prices are going up, delinquencies are negligible, and banks have a variety of hedges in place. They look at commercial real estate prices and think, if needed, they can get out of their portfolio at little cost. And even if some losses materialize, they have adequate capital.
I have some interesting people on my staff who can push back in a pretty challenging fashion. But at the end of the day, after carefully considering the banks’ arguments, they think: “All right, I guess the loan officers are looking at this pretty reasonably and are protecting their institution.” And the risk to the deposit insurance fund from all this activity seems pretty small. So, you end up being convinced that the activity is probably OK.
Today, with hindsight, we know that while most of the micro risks appeared very small at the time, their sum was far less than the macro risk that was silently building up. That’s the key thing: A collection of negligible micro risks can add up to a far greater macro risk. Focusing on individual institutions and controlling risks on a firm-by-firm basis are not enough for detecting and controlling systemwide stress points.
That is why we need macroprudential supervision and regulation.
Suppose for a particular class of assets, values decline on an economy-wide basis. This means losses are going to be taken at the macro level. Perhaps managers at a few individual banks can be smart, foresee the price declines, and liquidate their positions in time to avoid large losses at their institution. But the macro economy has to take these losses, and that’s where we get stuck. Not everyone can get through the exit door at once; someone has to end up bearing the macro losses.
This is why macroprudential regulations that aim to assess and control systemwide risks should play a critical role in our regulatory structure.
For instance, dynamic capital requirements and loan loss provisions that vary over the cycle can temper some of the boom–bust trends we have seen in the past. History shows that during boom times, when financial institutions are perhaps in an exuberant state, they may not price risks fully in their underwriting and risk-management decisions. During downturns, faced with eroding capital cushions, increased uncertainty, and binding capital constraints, some institutions may become overcautious and excessively tighten lending standards. Both behaviors tend to amplify the business cycle. Varying required capital loan loss provisions over the cycle could serve to offset some of this volatility.
We saw the advantages of a systemic, macroprudential approach firsthand during the implementation of the Supervisory Capital Assessment Program (SCAP) – the so-called “stress tests.” Last spring, the Federal Reserve led a coordinated examination of the largest 19 U.S. banks. We reviewed the institutions simultaneously, applying a common set of assumptions and scenarios across all of them. Such an approach provided us with a view of these banks in their totality, as well as the financial condition of individual institutions on a stand-alone basis. The horizontal view was essential in assessing how risks taken individually by each bank are correlated and how they can add up to more than the sum of individual components. The review also had a forward-looking element that assessed the likely condition of the banks under a specific set of adverse economic conditions and determined the amount of capital the banks would need under these “stress” conditions. Such procedures also enable supervisors to identify best practices in risk management and to push banks with weak controls to improve and adopt these industry best practices. Indeed, supervisors at the Federal Reserve have already begun to adopt such an approach.
However, even with such macroprudential strategies, we are going to face challenges. Let’s think about what, as a hypothetical macroprudential regulator, we would have to do. What should be the early call focus? What should we be looking at? When should we be looking at it? How confident are we that that we are actually going to be able to identify the problem? A macroprudential regulator is confronted with the same type of questions a microprudential regulator faces, but at a systemwide level.
Consider these questions within the context of commercial real estate. The facts are, today, CMBS and CRE loans have large delinquencies. Could anyone have made this call confidently in time to arrest the problems we face today?
On this slide Figure. 2, on the top panel, you see that the outstanding volume of CMBS ramps up in 2004, 2005. At the same time, commercial real estate prices (shown in the middle panel) continue to rise well into 2007. On the bottom panel, we see the performance of loans originated during this period, depending on when the loans were made. Loans originated later in the credit cycle are performing worse than older loans. For instance, loans underwritten in 2005 did not reach a 1 percent delinquency rate until about 42 months (3.5 years) after origination. In contrast, loans made in 2008 reached the 1 percent delinquency mark only six months after origination. The progressively worsening performance of loans originated later in the credit cycle is likely due to looser underwriting standards that supported the issuance boom.
With the benefit of hindsight, I can point to the inflection point in volume and say “I should have put my finger on that right then.”
At that point in the credit cycle, “shouting” would have been an important part of risk-control, as it would have emphasized potential risks to the market players.
But I can’t imagine that supervisors’ concerns would have been taken seriously in 2005 or 2006 -- even if they started going out and shouting to the heavens that there is a big, big problem and we need to do more about it. Recall that, at that time, real estate prices were ramping up and delinquencies were low. Indeed, in 2007, the Federal Reserve, along with other bank regulators, issued a supervisory guidance on concentrations in commercial real estate. 5 We also gave a number of speeches prior to the crisis about risk pricing and about market exuberance – to little avail. These warnings were largely ignored and we got a lot of push-back from banks. During boom periods when risks are silently building up, there are a lot of people with a lot of money at stake who will come out against such pronouncements. So, if policymakers do not follow words with actions, then we are not likely to make much progress. Shouting and supervision – together – are essential.
I raised some potential issues with both micro- and macro-prudential regulations.
How do we address these issues? This is where we would take full advantage of our multi-pronged approach to regulatory reform. If we are not certain that a particular approach may not be as effective as we would like, we can put more pressure on other levers to obtain a desired amount of risk control. So, if we think that macroprudential regulations may have some potential operational issues, we would need to implement more stringent capital and liquidity requirements than we would otherwise to overcome these issues.
This is why we need a multi-pronged approach to a robust regulatory structure: a structure that takes full advantage of the existing tools supervisors have ; a structure that supplements the existing one with dynamic capital requirements and a comprehensive approach to risk management ; a structure that includes a macroprudential supervisor than can monitor and assess incipient risks across institutions and markets and, when necessary, impose higher regulatory requirements on firms that pose systemic risks.
However, even with such a structure, it would be hubris on the part of policymakers to assume that we would be able to prevent financial stress at all financial institutions. Therefore, we also need to contain the disruptive spillovers that result from the failure of systemically important institutions without resorting to bailouts or ad hoc rescues. A necessary element of this is having a mechanism for resolving the failure of a systemically important institution.6 This is something we currently lack in many cases, though there are proposals now under discussion that would provide this resolution power.
Another issue that arises in the regulatory reform debate is whether the central bank should be entrusted with supervision and regulation responsibilities. There are many synergies between monetary policy and supervision and regulation that I and others have discussed in previous speeches.7 Let me point out a couple of reasons why it might be optimal for a central bank to have a key role in financial stability and regulation.
The reality is that central banks have the unique ability to act as the lender-of-last-resort during financial crises. The central bank cannot use this tool effectively if it is not knowledgeable about the financial condition of the institutions it might lend to, particularly if such loans need to be made at very short notice.
The lender-of-last-resort role inevitably thrusts the central bank into efforts to promote financial stability and avoid crises. If, however, central banks have no supervision and regulation tools, they are constrained to act with the only tool at their disposal – monetary policy.
I already mentioned that I am skeptical about using monetary policy to control financial exuberance. But without supervisory powers, there may be no choice. We know that time consistency issues can lead a central bank to choose inflationary outcomes in the short run, even though there is no long-run tradeoff between output growth and price stability. Ken Rogoff pointed out that one way to deal with this issue would be to appoint a conservative central banker who would be tougher than the public. This would ensure that appropriate decisions would be made and appropriate actions would be taken
Now, consider the reaction function of a central banker that has the additional responsibility for financial stability – but not the additional tools provided by a supervision and regulation role. Such a central banker might have to act against exuberance in financial markets more actively than it would otherwise. That would be entirely necessary and appropriate to preserve financial stability. However, that policy may not be the most appropriate one at that time for addressing the traditional goals of monetary policy of maximum sustainable employment and price stability. A central bank with three goals and only one lever is a recipe for producing some difficult policy dilemmas.
To sum up, it is clear that, in order to avoid a situation like the one we have faced in the past two years, we need to fortify our regulatory lines of defense. We need to change the rules of regulation to be more efficient and effective in their design and implementation. But we also need to openly acknowledge the challenges policymakers and regulators are likely to face in containing potential financial crises. Despite all the challenges, I believe that we can design a more effective regulatory structure through discussions such as the one we are having today.
1 For instance, see Evans (2009a and 2009b).
2 For instance, in recent years, regulators have actively supported the development of the Trade Information Warehouse (a central repository for trade reporting of over-the-counter credit derivatives contracts) and clearing houses for credit default swaps, such as ICE Trust.
3 Noncurrent loans are those that are 90 days or more past due plus loans in nonaccrual status.
4 At the end of 2004, return on equity at all commercial banks in the U.S. was 13.08 percent, near its historical peak of 16.23 percent in the second quarter of 1993. Return on assets were similarly high, and net charge-offs accounted for only 0.68 percent of total loans, well below 1.31 percent reached at the end of 2001.
5 See, “SR 07-1 Interagency Guidance on Concentrations in Commercial Real Estate” available online. More recently, the Federal Reserve issued a supervisory guidance on managing interest rate risk (“SR 10-1 Interagency Advice on Interest Rate Risk,” available online) and highlighted it in speeches (for instance, see Kohn (2010)). In addition, the Federal Reserve – along with the other Federal banking agencies – issued a policy statement on funding and liquidity risk management on March 17, 2010 (available online).
6 See Evans (2009c) for my views on the advantages of resolution authority and the issues it would address.
7 For instance, see Bernanke (2010), Evans (2010), Kashyap (2010), and Volcker (2010).
Bernanke, Ben S., 2010, The Federal Reserve’s Role in Bank Supervision, Testimony before the U.S. House Committee on Financial Services, Washington, DC, March 17.
Evans, Charles L., 2009a. “The International Financial Crisis: Asset Price Exuberance and Macroprudential Regulation.” Remarks given at the 2009 International Banking Conference on September 24, 2009 in Chicago, IL. Available online.
Evans, Charles L., 2009b. “Should Monetary Policy Prevent Bubbles?” Remarks given at the &Asset Price Bubbles and Monetary Policy Conference on November 13, 2009 in Paris, France. Available online.
Evans, Charles L., 2009c. “Too-Big-To-Fail: A Problem Too Big to Ignore” Remarks given at the European Economics and Financial Center on July 1, 2009 in London, England. Available online.
Evans, Charles L., 2010a. CFA Society of Chicago Distinguished Speaker Series: Luncheon Economic Forecast. Remarks delivered on March 4, 2010 in Chicago, IL. Available online.
Evans, Charles L. 2010b, CFA Society of Chicago Distinguished Speaker Series: Luncheon Economic Forecast, March 4. Available online.
Kashyap, Anil K, 2010 Examining the Link between Fed Bank Supervision and Monetary Policy, Testimony before the House Financial Services Committee, March 17. Available online.
Kohn, Donald L., 2010. “Focusing on Bank Interest Rate Risk Exposure.” Remarks delivered at the Federal Deposit Insurance Corporation’s Symposium on Interest Rate Risk Management on January 29, 2010 in Arlington, Virginia. Available online.
Volcker, Paul, 2010, Statement, Testimony before the House Financial Services Committee, March 17. Available online.