Too Big to Fail: The Way Forward | The Region | December 2008 | Publications & Papers| Federal Reserve Bank of Minneapolis
Editor's note: This column is based on remarks presented at Winona State University, Winona, Minn., Nov. 13, 2008.
The too-big-to-fail (TBTF) problem now rests at the very top of the ills elected officials, policymakers and bank supervisors must address.1 This ranking is sound given the expansion of the safety net over the last year, an expansion essential to quell recent market turmoil. And when thinking about what policymakers should do to address TBTF going forward, we have argued that the recommendations we have made over the last several years offer a promising approach.2
In explaining the merits of our specific recommendations, we have noted that the analytical framework used in developing those recommendations seems widely accepted at this point. But agreement on a general policy framework may, quite reasonably, not strike observers as a compelling reason to adopt our recommendations; a general framework may be consistent with a large number of specific reforms. Seemingly more compelling would be an assessment indicating that these reforms would have been of appreciable benefit if implemented prior to the recent turmoil. Indeed, we are frequently asked, "Would your reforms have made a difference?" suggesting real and, we think, understandable doubt on the part of observers.
I say "understandable" because we cannot truly know what might have ameliorated the many spillovers, particularly from one financial institution to another, that followed the collapse of the subprime lending market. In any case, policymakers have had to face developments in real time, not aspirations for what might have occurred, and as I indicated, I think the response was fully appropriate. That said, we would not have put forward recommendations if we did not think they would have provided benefits in the here and now. In particular, I will argue that these recommendations would have better prepared policymakers for the fallout that accompanied the weakening of systemically important financial firms. Such preparation may not have prevented the need for safety net expansion, but would have raised the odds that more narrow measures would have sufficed.
In the rest of these remarks, I first briefly describe recent Federal Reserve actions. I then provide some examples of recommended steps that
would have led to better preparation in advance of the crisis. Finally, I offer some specific proposals for near-term adoption.
Recent Federal Reserve actions
The Federal Reserve has taken a wide range of extraordinary actions to respond to conditions in the financial markets over the last year or so.
Given the tools available to the Federal Reserve and our mission, we have largely focused our efforts on increasing the availability of liquidity to financial institutions. Without trying to be comprehensive, I would note the following:
We have eased the terms of our discount window lending to traditional users, including reducing prices and lengthening maturities, for example.
We have also rolled out new ways to provide this credit, including auctioning it off. More dramatically, we have allowed certain securities firms, the so-called primary dealers, to access our credit facilities. Finally, in two cases, we have used our lending powers to try to facilitate the orderly resolution of financial firms whose failure otherwise posed systemic risk.
I could point to other actions, such as increasing our coordinated lending of dollars with other central banks, but suffice it to say the Federal Reserve has responded to unprecedented times with equally unprecedented actions. And, of course, we have lowered the federal funds rate target from 5¼ percent in September 2007 to 1 percent today. Such actions were appropriate given the challenges we faced, although I will comment soon about the downside associated with these policies.
We have seen some important progress in recent weeks in funding markets, due to these policy responses and due to related actions taken by other governmental institutions. That said, significant strains continue in some markets and among financial institutions. It is critical that the steps we have taken succeed in restoring stability. But as I noted, these actions have had the undesirable side effect of exacerbating the TBTF problem. Once immediate fires have been doused, policymakers will have to turn to reining in TBTF because, left unchecked, the TBTF embers remaining from our emergency response will likely contribute to future financial conflagrations. I now discuss some reforms to address TBTF that I think policymakers ought to consider seriously at that point.
Policies to address TBTF
I have long recommended that policymakers evaluate policies to address TBTF against their ability to appropriately reduce the likelihood that government will provide support to nominally uninsured creditors of large financial institutions. I believe that policymakers provide such support in order to limit the fallout, or spillovers, that arise when a large financial institution gets into trouble. So, effective TBTF policies are those that allow policymakers to better manage the spillovers from the collapse or failure of a large financial firm. Based on recent public statements from a range of officials, I see a consensus emerging on this policy framework.
This framework, however, does not provide sufficient detail to really guide policy. In prior work, we have provided a fairly extensive list of specific recommendations; more recently, we have offered a near-term plan with three specific reforms, which I will discuss later. These recommendations flow directly from the framework that policymakers seem comfortable with and thus are a good place to start.
However, both implicit and explicit feedback we have received suggest some underlying doubt about the reforms recommended and the justification for them. Put simply, we have been asked the rhetorical equivalent of the following two questions:
If our reforms were so on-target, why were they not adopted in the first place?
Would these reforms have actually made a difference to recent events?
Let me try to respond to these questions.
In terms of the first question, it is clear that we viewed TBTF as a greater risk and higher priority than many. I am not precisely sure why, but I think there are good reasons why others did not have the same level of concern. Some may have viewed TBTF reforms as a poor use of scarce resources. Policymakers always have a large number of initiatives under way, but they can give priority to only a select few. In this context, recall that by virtually all measures, most of the largest financial institutions were in excellent condition prior to recent turmoil. So ex ante, other issues may have reasonably seemed more important even if, ex post, TBTF is now viewed as paramount.
In other cases, I think the answer lies, at least partly, in the belief that previously enacted reforms would make it both exceedingly difficult and unnecessary for policymakers to support uninsured creditors. Observers seemed to believe these reforms put creditors at risk of loss and obviated concerns about TBTF. In particular, we heard from many that the regime created by the 1991 Federal Deposit Insurance Corporation Improvement Act (FDICIA) to limit TBTF support rendered our concerns about the scale and persistence of TBTF moot. Adherents of this view would not be expected to push efforts to fix TBTF to the top of the "to do" list.
Suffice it to say that we had a different view on this topic, one which we have been fairly vocal about for some time. In short, we did not think that FDICIA reforms would, when push came to shove, act effectively as a limit on creditor expectations or on policymaker actions, and recent events, in large part, bear this out. For example, policymakers invoked FDICIA's so-called systemic risk exception when they provided unlimited deposit insurance on noninterest-bearing business accounts at all banks.3 To the extent that these explanations provide the rationale for not enacting TBTF reforms previously, they no longer seem relevant, and thus we think that our recommendations are worthy of attention.
In terms of the second question, "Would the reforms have made a difference?" let me point to some representative examples suggesting that the reforms we recommended would have contributed to better preparation prior to the crisis.
One recommendation that would have increased preparedness for recent events concerns what we called scenario planning. We described key aspects of this reform as follows:
Policymakers could reduce the uncertainty that they face when a large bank fails by knowing the potential exposures other banks have to the failing institution in advance and practicing their response to such failures. … [Supervisors should examine] how the failure of one institution would affect the solvency of [other large banks]. … This amounts to checking out how much one bank … owes the others at a point in time-say, at the end of a business day. … [T]he government would focus on spillovers and cross-institution exposure. … Supervisors should develop detailed plans for addressing the failure of a large bank, test those procedures in simulations, and revise the procedures to account for test results. Supervisors should repeat the cycle regularly, given the rapidly changing operations of the largest banks. … [S]upervisors must identify the documents and data they will need to determine a bank's solvency and the exposures it would present to other banks at the time of failure. … Ultimately, supervisors must identify the gaps between what institutions can provide and what supervisors require. We view it as of the highest priority for supervisors to eliminate such gaps.4
This approach would have been of considerable value when determining potential responses to the illiquidity and/or insolvency of specific large financial institutions over the last year. To be sure, such preparation may not have ultimately changed the need for significant policy action, but policymakers would have likely had a better understanding of the specific "interconnectedness" of large financial firms, suggesting that responses to the outcomes could have been more timely and better focused.
In particular, if we (as policymakers) had grasped the net of connections of large financial firms in, say, 2006 instead of 2008, we might have taken steps to figure out how to contain the ability of this network to spread risk. For example, policymakers have now identified the absence of an effective resolution scheme as a major weakness in addressing the spillovers created when large nonbank financial firms get into trouble. This absence and a desire to contain these spillovers explain, in part, the extraordinary support such firms ultimately received. It is likely that the type of exploration we advocated would have raised the visibility of this problem.
In a second recommendation, we emphasized the importance of communicating and signaling to creditors their likely treatment in the resolution of institutions they might consider TBTF. We have been clear that policymakers need to "anchor" the expectations of these creditors to avoid surprising them with the eventual support that may, or may not, be forthcoming. Some observers have attributed the deterioration of credit and financial market conditions over the last several months to surprises that creditors of large institutions experienced.5
In a third example, we encouraged policymakers to consider new capital regimes that would have enhanced bank capital positions in bad times by locking in the ability to raise capital in the future.6 At the time we highlighted it, we noted that this proposal may not have been practicable, and it still might not be. But certainly many observers have concluded that a more "procyclical" capital regime would have better addressed the recent turmoil than the one currently in place.7
There are other recommendations we could mention. For example, we identified the benefits of increasing the use of centralized clearinghouses for derivative markets and stressed the importance of resolution schemes that could quickly make payments to uninsured creditors of the funds owed them by the failing institution.8
Again, I should stress that even with adoption of our recommendations, recent events might have unfolded largely as they did. Better preparation would not have changed the cause of our current financial troubles, though it almost certainly would have altered the effect, because better preparation makes for better policy. That said, we recognize that some recommendations we have made in the past have not held up. And I certainly make no claim for having foreseen how the decline in housing prices would spill over so aggressively to the financial sector and real economy. Finally, others did implement select reforms to address TBTF, which they identified with no help from us.
These caveats notwithstanding, by the standard of these two direct questions, our previously articulated reforms clearly have merit and deserve a second look. So where should policymakers start?
Systemic focused supervision
Having recognized the value of establishing priorities in my previous comments, I have tried to impose the same discipline on myself. We would begin the effort to manage TBTF with an approach we call systemic focused supervision (SFS). I have detailed this plan elsewhere, so let me just hit the main points here.9 In general, SFS attempts to focus supervision and regulation efforts on reduction of spillovers, and it consists of three pillars: early identification, enhanced prompt corrective action (PCA) and stability-related communication.
Early identification. This is a process to identify and to respond, where appropriate, to the material exposures among large financial institutions and between these institutions and capital markets. This process relates closely to the scenario planning recommendation I discussed a few moments ago. The goals of the exercise I described are (1) to give policymakers a sense of which events are not likely to severely impair a large financial institution, thus permitting them to avoid providing support, and (2) to identify those exposures that might bring down the firm, and thus are deserving of closer policy scrutiny and, most importantly, an effective and timely response.
Enhanced prompt corrective action. PCA works by requiring supervisors to take specified actions against a bank as its capital falls below specified triggers. Closing banks while they still have positive capital, or at most a small loss, can reduce spillovers in a fairly direct way. If a bank's failure does not impose large losses, by definition it cannot directly threaten the viability of other depository institutions that have exposure to it. Thus, the PCA regime offers an important tool to manage systemic risk. However, this regime currently uses triggers that do not adequately account for future losses and gives too much discretion to bank management. We would augment the triggers with more forward-looking data outside the control of bank management to address these concerns.
Communication. The first two pillars of SFS seek to increase market discipline by reducing the motivation policymakers have for protecting creditors. But creditors will not know about efforts to limit spillovers, and therefore will not change their expectations of support, absent explicit communication by policymakers about these efforts.
Recent events have been unprecedented. I'm skeptical of claims that the Federal Reserve or anyone else should have foreseen the situation as it actually played out. I also strongly support the actions the Federal Reserve has taken in response to these events, even with the undesired side effect of intensifying the TBTF problem. A significant issue, though, is what reforms should policymakers introduce to address the magnified TBTF problem? One criterion is that we consider reforms that would have helped prepare policymakers for the financial fallout they have faced over the last year or so, and it is my conviction that several reforms I have previously articulated fit that bill.
1 Policymakers have made this point directly. See, for example, Kate Gibson and Greg Robb, "Bernanke Sees Long Slowdown, but Still Confident," MarketWatch, Oct. 15, 2008. Recent press accounts also highlight the importance of TBTF. For example, see Cheyenne Hopkins, "Big Policy Choices Face New President," American Banker, Nov. 5, 2008. See also Bob Davis, Jonathan Weisman and Timothy Aeppel, "New Economic Ills Will Force Winner's Hand," Wall Street Journal, Nov. 4, 2008.
2 Many of these recommendations are developed in Gary H. Stern and Ron J. Feldman, Too Big to Fail: The Hazards of Bank Bailouts (Washington, D.C.: Brookings Institution Press, 2004).
3 Policymakers invoked the so-called systemic risk exception of FDICIA in creating the FDIC's "Temporary Liquidity Guarantee Program" and in the context of the proposed acquisition of Wachovia Corp. by Citicorp Inc. See FDIC press releases 100-2008 and 88-2008, respectively. In contrast, policymakers did not invoke the exception in the failures of IndyMac Bank, F.S.B., and Washington Mutual Bank despite the large size of these two depositories. See FDIC press releases 56-2008 and 85-2008.
4 See Stern and Feldman (Too Big to Fail), pages 112 and 114. We explain our broad definition of the term "bank" and our use of it on pages 14-16.
5 For our suggestions, see Gary H. Stern and Ron J. Feldman, "Constructive Commitment: Communicating Plans to Impose Losses on Large Bank Creditors," in Douglas D. Evanoff and George G. Kaufman (eds.), Systemic Financial Crises: Resolving Large Bank Insolvencies (Hackensack,
N.J.: World Scientific Publishing, 2005).
6 See Stern and Feldman (Too Big to Fail), page 128.
7 For an example of an extension of the capital proposal we highlighted, see Anil K. Kashyap, Raghuram G. Rajan and Jeremy C. Stein, "Rethinking Capital Regulation," 2008.
8 See Stern and Feldman (Too Big to Fail), pages 137 and 122-23, respectively.
9 See Gary H. Stern, "Limiting Spillovers Through Focused Supervision," The Region, September 2008.