Do SIFIs Have a Future?
Thomas M. Hoenig
President and Chief Executive Officer
Federal Reserve Bank of Kansas City
June 27, 2011
We are approaching the one-year anniversary of the Dodd-Frank Act. With so much of the Act’s implementation work unfinished, it is not an anniversary that we can celebrate; rather it is an opportunity to take stock. In that vein, I congratulate the organizers of this conference for bringing together such an excellent group of individuals to do just that. In particular, I want to recognize the New York University Stern School of Business and economics faculty for the outstanding research they have done on critical financial reform issues. Their collaborative efforts have produced a series of must-read books that have a combination of rigorous economic analysis and practical policy prescriptions that is rarely seen these days.
Much of the Dodd-Frank discussion revolves around the economic distortions and disruptions caused by the largest and most complex financial companies, the so-called systemically important financial institutions, or SIFIs.
As we consider the topic of SIFIs, let me ask the following questions: How can one firm of relatively small global significance merit a government bailout? How can a single investment bank on Wall Street bring the world to the brink of financial collapse? How can a single insurance company require billions of dollars of public funds to stay solvent and yet continue to operate as a private institution? How can a relatively small country such as Greece hold Europe financially hostage? These are the questions for which I have found no satisfactory answers. That’s because there are none. It is not acceptable to say that these events occurred because they involved systemically important financial institutions.
Because there are no satisfactory answers to these questions, I suggest that the problem with SIFIs is they are fundamentally inconsistent with capitalism. They are inherently destabilizing to global markets and detrimental to world growth. So long as the concept of a SIFI exists, and there are institutions so powerful and considered so important that they require special support and different rules, the future of capitalism is at risk and our market economy is in peril.
To more fundamentally address this issue, we must go beyond today’s Dodd-Frank. We must confine the use of the safety net to its original intent. We must reduce the artificial complexity of existing financial structures. The rewards of success must be balanced against the credible consequences of failure. In achieving such goals, we will enhance the stability of the fundamental mechanism through which monetary policy is conducted and the economy depends.
The decline in competition and accountability in banking The U.S. economy is the most successful in the history of the world. It achieved this success because it is based on the rules of capitalism, in which private ownership dominates markets and individuals reap the rewards of their success. However, for capitalism to work, businesses, including financial firms, must be allowed, or compelled, to compete freely and openly and must be held accountable for their failures. Only under these conditions do markets objectively allocate credit to those businesses that provide the highest value. For most of our history, the United States held fast to these rules of capitalism. It maintained a relatively open banking and financial system with thousands of banks from small community banks to large global players that allocated credit under this system. As late as 1980, the U.S. banking industry was relatively unconcentrated, with 14,000 commercial banks and the assets of the five largest amounting to 29 percent of total banking organization assets and 14 percent of GDP.
Today, we have a far more concentrated and less competitive banking system. There are fewer banks operating across the country, and the five largest institutions control more than half of the industry’s assets, which is equal to almost 60 percent of GDP. The largest 20 institutions control 80 percent of the industry’s assets, which amounts to about 86 percent of GDP.
Here’s the irony: This marked increase in concentrated power, and therefore, more concentrated risk, reflects past efforts to assure greater economic stability. This might best be described as “good intentions/bad outcomes” syndrome. For example, the Federal Reserve was founded following the 1907 Banking Panic and was charged with providing liquidity support to solvent banks that were experiencing funding problems. After the Great Depression, the Federal Deposit Insurance Corp. was created to provide limited deposit insurance to protect small depositors and to further increase the resiliency of the financial system.
Then, over the past 30 years, this safety net has expanded far beyond its original intent. More recently, Glass-Steagall was repealed, giving high-risk firms almost unlimited access to funds generated through their new access to the safety net. Finally, following a series of crises during the late 1980s and 1990s, the government confirmed that because of systemic impact, some institutions were just too big to fail—the largest institutions could put money in nearly any asset regardless of risk, and their creditors would not be held accountable for the risk taken. Predictably, the industry’s risk profile increased dramatically. The SIFI was born.
Is it any wonder then that in the fall of 2008 we experienced the greatest financial crisis since the Great Depression? Financial institutions had again become irresponsible in their lending practices. They had increased their leverage ratios to unprecedented levels. They became “dry kindle” for a financial fire and, with the end of the housing boom, the match was struck.
Now, with their bailout costs amounting to billions of taxpayer dollars, SIFIs are larger than ever. Strikingly, they are arguing that they should not be held to stronger capital standards if the United States hopes to remain globally competitive. That assertion is nonsense. The remainder of my remarks today will describe how the United States can achieve a stronger, more stable financial system in order to secure its future as a global economic leader.
Proposal to reduce costs and risks to the safety net and financial system Following this financial crisis, Congress and the administration turned to the work of repair and reform. Once again, the American public got the standard remedies—more and increasingly complex regulation and supervision. The Dodd-Frank reforms have all been introduced before, but financial markets skirted them. Supervisory authority existed, but it was used lightly because of political pressure and the misperceptions that free markets, with generous public support, could self-regulate.
Dodd-Frank adds new layers of these same tools, but it fails to employ one remedy used in the past to assure a more stable financial system—simplification of our financial structure through Glass-Steagall-type boundaries. To this end, there are two principles that should guide our efforts to restore such boundaries. First, institutions that have access to the safety net should be restricted to certain core activities that the safety net was intended to protect—making loans and taking deposits—and related activities consistent with the presence of the safety net.
Second, the shadow banking system should be reformed in its use of money market funds and short-term repurchase agreements—the repo market. This step will better assure that the safety net is not ultimately called upon to bail them out in crisis.
Consistent with the first principle, banking organizations with access to the safety net should be generally confined to the following activities: commercial banking, underwriting securities and advisory services, and asset and wealth management services. Underwriting, advisory, and asset and wealth management services are mostly fee-based services that do not put much of a firm’s capital at risk. In addition, asset and wealth management services are similar to the trust services that have always been allowable for banks.
Restricting activities of banking organizations
In contrast, banking organizations should be expressly prohibited from activities that include dealing and market-making, brokerage, and proprietary trading, which expose the safety net but have little in common with core banking services.1 Within the protection of the safety net, they create expansive risks that are difficult to assess, monitor or control.
Thus, banking organizations would not be allowed to do trading, either proprietary or for customers, or make markets because such activity requires the ability to do trading. In addition, allowing customer but not proprietary trading would make it easy to game the system by “concealing” proprietary trading as part of the inventory necessary to conduct customer trading. Also, prime brokerage services not only require the ability to conduct trading activities but also essentially allow companies to finance their activities with highly unstable uninsured “deposits.” This combination of factors, as we have recently witnessed, leads to unstable markets and government bailouts.
Critics of this proposal contend that institutions grow to be large and complex because of economies of scale and scope and they need the size and related complexities to be profitable and to compete globally. They believe firms must have broad, mostly unrestricted access to all financial activities to provide one-stop shopping and compete on a global basis. Arguments also are given that large banks and securities firms are necessary to make efficient markets for securities trading essential for carrying out monetary policy.
These arguments are unconvincing and, in fact, mislead. First, yes, it would be unfortunate if restricting activities were to drive U.S. banks and jobs to other countries. However, we have 200 years of banking success in this country to refute that assertion. More recently, under Glass-Steagall, U.S. banks and investment banks were highly competitive and successful as each specialized in lending to or underwriting businesses all over the world. There is considerable evidence that under Glass-Steagall the United States was at no competitive disadvantage to Europe, with its mingled merchant banking system.
The United States led the world—because it had strong, prudently run institutions that knew how to manage money in the best interests of the client.
Second, there is no strong evidence of unlimited economies of scale and even less for wide economies of scope. Although both exist, they are captured at an asset size far less than that of SIFIs today.
Third, large corporations would have ample convenient access to commercial and traditional investment banking services inside commercial banking. They would have to go to securities dealers to purchase swaps and other derivatives for hedging purposes, something that has been done in the past without difficulty.
Finally, it also seems improbable to me that any country should be willing or able to expand its safety net or to expose its taxpayers to the undefined risks of protecting ever-larger and more complex banking organizations. Instead, what countries should be focused on now is getting back to fundamentals aimed at simplifying highly complex and unstable SIFIs. The focus should be on financial stability.
Reforming the shadow banking system
A legitimate concern of limiting the safety net is that this could worsen the risk of financial instability by pushing activities to the unregulated shadow banking system. Clearly, focusing solely on the regulated banking industry and ignoring the unregulated shadow banking system would not solve the problem and, in fact, might expand the shadow banking sector that was an integral part of the financial crisis.
Much of the instability in the shadow banking system stems from its use of short-term funding for longer-term investment. The solution to this instability problem is not to provide a safety net for the shadow banks and regulate them more but, instead, to remove exceptions in which money market instruments are treated essentially as deposits. The current exceptions encourage significant short-term
funding of longer-term assets.
First, investors in money market mutual and other investment funds that are allowed to maintain a fixed net asset value of $1 have an incentive to run if they think their fund will “break the buck.” Thus, if the fixed $1 net asset value is eliminated and the share values of such funds are required to float with their market values, shadow bank reliance on this source of short-term funding and the associated threat of disruptive runs would be greatly reduced.
Second, the potential disruptions to funding stemming from the repo financing of shadow banks should be ended. One of the sources of instability during the crisis was repo runs, particularly on repo borrowers using subprime mortgage-related assets as collateral. Essentially, these borrowers funded long-term assets of relatively low quality with very short-term liabilities.
These practices would be greatly reduced by rolling back the bankruptcy law for repo collateral to the pre-2005 rules. Prior to then, if a repo borrower defaulted, mortgage-related collateral could not be immediately taken and sold by the creditors. Returning to these rules would discourage the use of mortgage-related assets as repo collateral and reduce the potential for repo runs. Term lending through securitization would continue, probably at a smaller scale, with more closely matched term wholesale funding provided by institutional investors such as mutual funds, pension funds and life insurance companies.
These changes to the rules for money market funds and repo instruments would increase the stability of the shadow banking system because term lending outside the safety net would be less dependent on “demandable” funding and more reliant on term funding, and the pricing of risk would better reflect the actual risk incurred.
A final note: Implications for monetary policy
Finally, as a member of the Federal Open Market Committee, I realize that we must consider the potential effects of these proposals on the conduct of monetary policy. The impact could be significant because, as currently practiced, monetary policy operations are channeled through a limited number of counterparties called primary dealers. These dealers are required to participate in all auctions of U.S. government debt and represent a key element in the implementation of policy.
Currently, there are only 20 primary dealers. They are the largest financial firms operating in the United States, and most are affiliated with commercial banks. It is with this relationship that the changes I propose could affect the conduct of monetary policy. Specifically, given that primary dealers could no longer be affiliated with commercial banks, would this inhibit market-making in securities, including Treasuries, and therefore interfere with the conduct of monetary policy?
The answer is “no.” It is not necessary that primary dealers be affiliated with banks. It is only necessary that they be institutions that deal in U.S. Treasuries and participate in auctions of U.S. government debt. Prior to the 1990s merger boom among investment banks and the Gramm-Leach-Bliley Act, it was typical that half or more of the primary dealers were not affiliated with commercial banks.
Therefore, the fact that primary dealers are not commercial banks would have little effect on the Federal Reserve’s ability to conduct monetary policy. I would add that although commercial banks could not be primary dealers, they could remain a key part of the monetary policy mechanism. Recently, the Federal Reserve gained experience in using the Term Auction Facility (TAF). The TAF might very successfully be used in conjunction with primary dealer operations to conduct policy well into the future.
As you may recall, during the financial crisis, the TAF was an important component of monetary policy. For example, the TAF was introduced in December 2007 with an initial auction of $20 billion. The facility was then ramped up to almost $500 billion by March 2009—about one-fourth of the assets on the Federal Reserve’s balance sheet. The maturity of TAF loans was generally 28 days or 84 days.
By broadening the Federal Reserve’s monetary tools to include the TAF to provide term funds through the banking system in parallel with the primary dealers, we could greatly expand the number of counterparties used in the conduct of monetary policy. Thus, the TAF and primary dealers would provide deep markets for the term portion of policy; primary dealers and traditional open-market operations would continue as the means for managing day-to-day operations and for maintaining the federal funds rate close to the target. With more counterparties, we enhance competition and enable nearly all banks to play a role in the conduct of monetary policy. This would make the largest banks less “SI” and more “FI.”
The financial system has become far less competitive and far more volatile with the onset of systemically important institutions. Though large firms remain a critical part of our economic system in the United States, they should not become so dominant that they become unaccountable to our capitalistic system. We can make the necessary reforms to end the unique status of the SIFI and, in doing so, restore much of our global competitive vigor. Only from a position of financial strength can the United States remain the global economic leader.
1. This categorization of financial activities is from Matthew Richardson, Roy Smith and Ingo Walter in Chapter 7 of Regulating Wall Street: The Dodd-Frank Act and the New Architecture of Global Finance, edited by Viral V. Acharya, Thomas F. Cooley, Matthew Richardson, Ingo Walter, New York University Stern School of Business,
Pew Financial Reform Project and New York University Stern School of Business
“Dodd-Frank One Year On”
The views expressed by the author are his own and do not necessarily reflect those of the Federal Reserve System, its governors, officers or representatives.