|Home > News and Events > Speeches|
Reducing the Systemic Risk in Shadow Maturity Transformation
|March 8, 2011||
Sandra C. Krieger, Executive Vice President
Remarks at the Global Association of Risk Professionals 12th Annual Risk Management Convention, New York City
Thank you for inviting me to speak here today. The views expressed are mine and do not necessarily reflect those of the Federal Reserve Bank of New York or the Federal Reserve System. That said, I want to acknowledge my many colleagues at the Bank who provided essential input.1
My objective today is to talk about the systemic risk that can be created by financial intermediaries that do not have direct and explicit access to official liquidity—the so-called shadow banks—and how these risks might be reduced. My focus is on maturity transformation activities—that is, the use of short-term funding to finance longer term, risky assets. These activities exploded during the credit bubble. And they popped along with the bubble, killing or nearly killing the sponsoring institutions with the toxic (and nontoxic) assets therein.
We were reminded during the financial crisis of how banks are special—they have access to direct and explicit official credit and liquidity backstops. That is, banks have access to Federal Reserve credit and insured depositors don’t need to be short-distance runners. It is a different story for financial intermediaries without this type of backstop. Their liquidity support is less assured and their funding can be quick to flee.
Large banks were the bankers to the shadow banks and activities that lacked official support during the first stages of the financial crisis. But, as we saw, this was insufficient to prevent damaging run dynamics from emerging. The banks did not have the capital to bring all of their off-balance-sheet liabilities onto their balance sheets nor was there always enough "there there" in the shadow banks to permit bank lending to satisfy their obligations. The selling that ensued to try to square the circle in individual cases just made the aggregate imbalances worse.
In response to the dramatic erosion in market liquidity conditions, central banks and governments stepped in and lent freely, to traditional banks against traditional and nontraditional assets and—when that lending proved insufficient to stabilize the situation, they also lent also to nontraditional banks—to the shadow banks. Thus, the defining characteristic of the shadow institutions and their obligations—the absence of direct and explicit access to official credit and liquidity—was violated.
The undertaking of central bank lending to nonbanks that were not subject to the regulation and oversight of traditional banks occurred out of necessity to stabilize a situation that threatened to lead to a much broader and more sustained collapse than was already underway. But these actions were certainly not without their costs; they risked establishing the wrong incentives—incentives to take on too much risk; incentives not to know the collateral that underlies short- term deposits; incentives to continue to ignore the maturity transformation that is taking place.
This topic is one of many systemic risk issues of importance to the Federal Reserve. I lead the Credit and Payments Risk Group at the Federal Reserve Bank of New York—a group that we built out during the financial crisis to manage the risks associated with the Bank’s sharply changed balance sheet. The balance sheet that, as I just noted, took on the kind of assets to which I have been referring—assets that embed the maturity transformation activities of shadow banking entities, for financial stability reasons. Staff in the Credit and Payments Risk Group work with others in the Bank—in Bank Supervision, Research, Markets, Legal and the Executive Office—and with colleagues at other Federal Reserve Banks, the Board of Governors, and other regulators and central banks—to influence and implement policies to reduce systemic risk and strengthen financial stability.
What Is Shadow Maturity Transformation?
Credit intermediation and shadow maturity transformation
For banks, credit intermediation is enhanced by credit and liquidity "put" options provided through deposit insurance and access to central bank liquidity, respectively. These types of "official" enhancement are direct and explicit.
Official enhancements to credit intermediation activities have four levels of "strength" and can be classified as either direct or indirect, and either explicit or implicit. A liability with direct official enhancement must reside on a financial institution’s balance sheet, while any off-balance-sheet liabilities of financial institutions are indirectly enhanced by the public sector. For example, an insured deposit is directly enhanced by official credit and liquidity puts, while a bank’s off-balance-sheet asset-backed commercial paper conduit (or "ABCP conduit") is indirectly enhanced by a backup line of credit. Insurance is explicit, while the securities lending activities of a commercial bank have implicit liquidity puts. An unfunded commitment is something else still—unenhanced.
Shadow maturity transformation includes maturity transformation which is (1) implicitly enhanced, (2) indirectly enhanced or (3) unenhanced by official guarantees—everything that is not directly and explicitly enhanced by the official sector.
The role of banks in shadow maturity transformation and the stability of the parallel banking system
It is this maturity transformation that renders financial intermediaries intrinsically fragile since, by definition, an intermediary engaging in maturity transformation cannot honor a sudden request for full withdrawals.
Since financial intermediation is intrinsically fragile, what makes modern, bank-based intermediation relatively stable? The answer is the combination of the existence of explicit, official support by central authorities: that is, the conditional provision of (credible) secured funding (for example, central bank discount window access) and the conditional obligation of the official sector to assume loss, that is, the protection of intermediaries’ liabilities in the event of their default (for example, deposit insurance).
By extension, today’s nonbank based system of financial intermediation also exists and thrives because entities performing maturity transformation receive some form of both liquidity and credit support. (One may push this and say that the provision of this support is a necessary condition for the existence of shadow maturity transformation).
While various types of entities have provided this support to nonbanks, it is the banks themselves that are the central players in performing this function. The reason that banks are the central providers of support services, and therefore are the main backers of shadow banking, is that their sponsoring services are credible. And this credibility in turn emanates from the support they receive from the official sector.
These patterns suggest two basic paths to reduce the fragility of shadow banking activities. The first is to strengthen the ability of, and increase the cost to, sponsoring banks to backstop them; the kind of liquidity and capital enhancements that are embedded in Basel III. The other is to insist on the ability of the shadow institutions to provide for a robust and credible backstop that resides outside of traditional banks.
Major types of shadow maturity transformation
The Federal Reserve created seven emergency liquidity facilities to deal with the unwind of shadow credit transformation: the term auction credit facility, foreign exchange swaps with foreign central banks (not new but used in an expanded manner), a primary dealer credit facility (PDCF), a term securities lending facility (TSLF), an asset backed commercial paper money market mutual fund liquidity facility (AMLF), a commercial paper funding facility (CPFF) and a money market investor funding facility (MMIFF). While successful in achieving their unique goals, these facilities were merely a bridge to more normal markets, buying time for well-needed structural reform.
I will provide a little background on these markets, including how they looked during the credit bubble and then during the financial crisis. In every case, we should focus on the source and credibility of the credit and liquidity backstops.
Asset-backed commercial paper
For corporate users, ABCP benefits included some funding anonymity; increased commercial paper (CP) funding sources; and reduced costs relative to strict bank funding. ABCP conduits expanded from financing of short-term receivables collateral to a broad range of loans, including auto loans, credit cards, student loans and commercial mortgage loans. At the same time, as the market developed, it came to embed much more maturity mismatch through funding longer-term assets, warehoused mortgage collateral, etc. Securities arbitrage vehicles used ABCP to fund various types of securities, like collateralized debt obligations (CDOs), asset-backed securities (ABS) and corporate debt.
ABCP is traditionally enhanced with an "explicit liquidity put to a commercial bank" where the amounts of the liquidity proceeds are sufficient to pay off maturing ABCP. Exceptions in the past were structured investment vehicles (SIVs) and "SIV lites" that had limited or no liquidity commitments from a commercial bank and instead relied on a sale of the securitized assets to pay off maturing commercial paper.
The run on ABCP began in the summer of 2007 when a foreign bank was unable to value the collateral underlying the off-balance-sheet conduit. Short-term funding costs spiked. Meanwhile, the markets for the underlying instruments plummeted.
The Federal Reserve responded to resulting pressures in short-term funding markets expanding its traditional repo operations, for which U.S. Treasury, and U.S. government-sponsored entities (GSE) debt and mortgage-backed securities (MBS) are acceptable collateral. Soon after, the Federal Reserve made it clear that bank borrowing from the discount window would be viewed as acceptable.
When short-term funding costs spiked anew in the fourth quarter of 2007, the Federal Reserve stepped up again, introducing a term auctioned credit facility. Also in December 2007, the Federal Open Market Committee (FOMC) authorized swap lines with other central banks, facilitating the provision of short-term U.S. dollar funding to foreign banking organizations. As the crisis intensified, so did the sizes of these facilities. At their respective peaks, there was almost $500 billion of term auctioned credit outstanding and nearly $600 billion of foreign exchange swaps.
In the fall of 2008, the Fed introduced an explicit backstop for ABCP through the AMLF. This peaked just over $150 billion. The Federal Reserve also introduced the CPFF, which was authorized to purchase three-month unsecured and ABCP directly from eligible A-1/P-1 CP issuers. Like the AMLF, the CPFF provided greater assurance to both issuers and investors that firms would be able to meet redemptions—in this case, to roll over their maturing commercial paper. It also increased the availability of term commercial paper funding to issuers. The maximum outstanding of asset backed-CPFF loans was $125 billion; for unsecured CP it was roughly $225 billion.
During the financial crisis, ABCP outstanding fell considerably from its peak. But, of course, the underlying longer-term assets did not disappear—they were sold into distressed markets or came onto the balance sheets of the sponsoring banks.
Tri-party repo market
The tri-party repo market was initially small and limited to highly liquid collateral such as U.S. Treasury and agency securities. Tri-party repos proved so popular with cash investors that they demanded more tri-party repo investment opportunities and became willing to accept even illiquid collateral like whole loans and non-investment grade securities—because receiving their cash back each morning provided them with the perception of liquidity. Ultimately, the tri-party repo market peaked in March 2008 at $2.8 trillion. The largest individual borrowers routinely financed more than $100 billion in securities through these transactions. At the peak of market activity, the largest dealer positions exceeded $400 billion. Securities dealers became dependent on this form of funding to fund their securities positions.
In March 2008, when Bear Stearns Co. had funding difficulties, its clearing bank became reluctant to continue to provide intraday credit needed to prevent a default. At this point it became clear that neither clearing banks, nor overnight cash investors, were well prepared to manage a dealer default. Each found it in their best interest to pull away from the troubled borrower before the other to avoid destabilization of their own firms. Furthermore, the liquidation of such large amounts of collateral under the extreme market pressures would have created fire sale conditions, large liquidity dislocations and undermined confidence in the whole market.
To avoid these adverse systemic consequences, the Federal Reserve stepped in and created a special lender of last resort-like facility to lend to dealers against their tri-party repo collateral. The facility effectively backstopped the market in the immediate circumstances surrounding Bear Stearns’s failure. When financial conditions worsened considerably further in September 2008, the facility was needed to forestall multiple failures and associated systemic consequences thereof and, as I mentioned, the fire sale of the underlying collateral and the broader impact that would have had. The Fed expanded the terms of the program so it could backstop virtually any type of tri-party repo collateral. Daily use of PDCF peaked at roughly $150 billion.
The Fed also supported disruptions in funding markets with a term securities lending program, introduced also in March 2008. This facility supported the tri-party repo market by permitting dealers to swap the less liquid securities collateral being shunned by investors for Treasuries, which they could use to obtain secured funding. The amount outstanding in this program at its peak was about $200 billion.
Money market mutual funds
While prime money market mutual funds (MMMFs) offer immediate redemptions of shares at a rounded price, which in practice essentially never deviates from one dollar, their assets have a longer term and may be costly to liquidate. In times of extreme stress in the financial sector, the risk profiles of prime money fund assets change rapidly, inconsistent with investors’ liquidity and safety requirements—full daily liquidity and a stable net asset value (NAV). As a result, the prime fund industry is vulnerable to a confidence shock that could result a rapid flight of investors. In turn, that could have broader systemic consequences through large-scale asset sales to meet large volumes of redemptions.
This fragility of MMMFs can quickly spread to other financial firms and the broader economy given the size of the money fund industry and its prominence in short-term financing markets. In particular, MMMFs are major investors in liabilities of financial firms, both domestic and foreign.
The fragility of money funds, and potential broader consequences was front and center in September 2008 when Lehman failed; all of what I just said occurred: the confidence shock, and then rapid changes in money fund risk profiles and investor risk appetite moving in opposite directions. In this environment, the Prime Reserve Fund, a well-established money market fund that had exposure to Lehman CP, "broke the buck." Money market fund investors at other funds voted with their feet regarding their discomfort with the lack of guaranteed credit and liquidity support for these activities, withdrawing large amounts from funds that invested in instruments that did not have full and direct government support or clearly sufficient parent support. Fund managers reacted by selling assets and investing at only the shortest of maturities, thereby exacerbating the funding difficulties for other instruments such as commercial paper.
The Federal Reserve and the U.S. Treasury stepped in, creating a number of emergency programs to backstop money funds. The Fed’s programs that supported money funds were the AMLF and CPFF, which also supported the short-term funding markets more broadly. There was also a special Money Market Investor Funding Facility (MMIFF), to provide liquidity to U.S. money market mutual funds and certain other money market investors although this backstop funding source was never used.
While the Federal Reserve created the liquidity puts, the U.S. Treasury provided the credit puts for money funds. It created the Money Market Fund Guarantee-Temporary Guarantee Program, which insured shareholder assets in participating money market funds.
Systemic Risk Created by Shadow Maturity Transformation
The sale of distressed assets becomes the problem at hand when investors exercise their rights to withdraw. The sales in stressed conditions reduces the value of the market as a pricing mechanism and capital is destroyed—not only of the affected institution, but also in all other institutions that hold these affected assets on a mark-to-market basis. Moreover, runs spread from unhealthy to healthy institutions, forcing liquidations that unnecessarily destroy value.
The Future of Shadow Maturity Transformation
Asset-backed commercial paper
The impact of the greater capital and liquidity requirements for bank-sponsored conduits likely will include higher-cost lines of credit to finance companies and the end of programs that exist solely for off-balance-sheet capital arbitrage. Mitigating behavior by the industry might include: a shift in conduit sponsorship from U.S. banks to non-banks or foreign banks with balance sheet capacity, or a re-structuring of conduits in order to avoid accounting consolidation (for example, through the sale of first-loss tranche to transfer control to third party).
Most of the focus on the ABCP providers has been on the "internal" shadow banking institutions, such as bank-sponsored finance companies, rather than the "external" shadow banking institutions. As investor appetite returns, there will be incentives to use highly rated, unregulated counterparties. Supervisors will need to be vigilant about supervised banks that rely on these types of companies for credit protection and capital relief. Investors too need to carefully evaluate the credit and liquidity protection provided by unregulated but highly rated entities. Rating agencies will need to evaluate the capital adequacy of rated entities, the ability of these entities to meet the likely calls for liquidity and monitor the ongoing viability of unregulated entities.
Outside of the regulation of borrowers, a private-sector task force was assembled by the New York Fed to address infrastructure design issues that obscured credit and liquidity risks in this market. The Tri-Party Repo Market Infrastructure Reform task force recommended and is implementing changes that will materially reduce reliance on clearing bank credit by adopting collateral substitution procedures and a 24-hour term for overnight repo.2 These changes will bring the market infrastructure into line with practices in Europe and other parts of the world, and will dampen an important channel for the transmission of systemic risk by making it less likely that a troubled dealer can destabilize its clearing bank and vice versa through its tri-party repo activities. Intraday credit will be limited and supported by a committed credit line.
These changes will also force cash investors to consider the credit and liquidity risks they assume because they can no longer assume that the clearing banks will provide an implicit credit and liquidity backstop. As a result, we expect some tri-party repo cash investors to strengthen their risk management by paying closer attention to their counterparties’ ability to repay their loans and by selecting more liquid, higher quality collateral. Some may exit the market altogether if they conclude that the risks inherent to this activity are not in line with their risk appetite. A smaller, more conservatively collateralized tri-party repo market may well emerge.
The tri-party repo market will also more conservatively price the credit intermediation.
Money market mutual funds
Several of these proposals entail the creation of liquidity and capital buffers. The former provide additional near-cash assets to deal with redemptions, while the latter enhances the loss absorption capacity available to deal with a credit event. Broadly speaking, two kinds of buffers can be set up: ex ante and ex post.
One type of ex ante buffer is to create a private emergency liquidity facility, capital reserve, or insurance. Regulated fixed NAV funds would benefit from an ex ante buffer but be forced to pay the cost. Another approach to an ex ante buffer is for individual funds to set aside resources in advance to absorb losses should they occur, as capital does in traditional banks. As an alternative, the Investment Company Institute has proposed a private sector "liquidity bank" which would provide a backstop but itself might benefit from access to official liquidity.
An ex post buffer does not require any resources to be set in advance, but is created by taking steps to ensure that investors absorb losses when they occur, and cannot flee leaving the losses behind. In particular, such measures are designed to forestall investors redeeming shares at a NAV of one dollar once credit event or liquidity event has begun. A variable NAV may be helpful in this regard, as such an NAV, if properly computed, could adjust rapidly in response to losses or liquidity shocks. However, this would be a fundamental change in the nature of MMMFs.
In summary, regulators have certainly made some significant improvements to the structure of the MMMF industry which may reduce the likelihood of runs and improve its resiliency. However, until more significant reforms are undertaken, a clear systemic vulnerability remains. It is important to note that there may well be no single measure that adequately addresses this issue, and some combination of measures may ultimately be the most appropriate course.
Much regulatory reform is focused on better aligning the cost and incentives for banks to provide the backstop support for these activities, with the intent of inducing more socially efficient levels of these activities. Other reforms are focused on reducing reliance by shadow institutions on traditional banks, by having the shadow banking entities themselves provide for the necessary credit and liquidity backstops, and shadow investors bearing the full ex ante economic cost of maturity transformation. Reforms of these types are necessary to ensure that liquidity is provided in a risk-sensitive manner and that full and credible resolution does not depend on official liquidity support.
So, I leave you with three thoughts:
We have walked through key aspects of the development of the shadow banking system. We—the financial industry and regulators together—understand systemic risk in ways we previously did not. We—the financial industry and regulators together—are beginning to address vexing problems. We—the financial industry and regulators together—must build a solid foundation for the future to avoid systemic risk arising from shadow maturity transformation. Thank you for your time and attention to these important issues.