What Risk Professionals Want

“Risk comes from not knowing what you’re doing.”
Warren Buffett (cited in Omaha World-Herald, January 2, 1994)

“If you don’t invest in risk management, it doesn’t matter what business you’re in, it’s a risky business.” Gary Cohn, then-President of Goldman Sachs, panel discussion, September 25, 2011

As memories of the 2007-09 financial crisis fade, we worry that complacency is setting in. Recent news is not good. In the name of reducing the regulatory burden on small and some medium-sized firms, the Congress and the President enacted legislation that eased the requirements on some of the largest firms. Under the current Administration, several Treasury reports travel the same road, proposing ways to ease regulatory scrutiny of large entities without changing the law (see here, here and here). And, recently, the Federal Reserve Board altered its stress test in ways that make it more likely that poorly managed firms will pass. It also voted not to raise capital requirements on systemically risky banks over the next 12 months.

A few weeks ago, one of us (Steve) had the privilege to speak at the 20th Risk Convention of the Global Association of Risk Professionals (GARP). Founded in 1996, GARP engages in the education and certification of risk professionals and has several hundred thousand members worldwide. (Disclosure: Brandeis International Business School and NYU Stern are GARP Academic Partners.) The organizers allowed us to solicit the views of the 100-plus attendees on two issues that are central to financial resilience: Are bank capital requirements high enough? And, do central counterparties (CCPs) have sufficient loss-absorbing buffers? They answered both questions with a resounding “NO.”

As regular readers of our blog know, we remain concerned about the resilience of the system in general and capital requirements in particular. To be sure, once we take account of the updated definition of capital, the Basel III requirements are roughly 10 times the level of Basel II. Furthermore, actual capital levels at the largest global banks—measured by an equal-weighted leverage ratio—are now double what they were a decade ago (see following chart). 

Leverage ratio of largest global banks (fully phased-in Basel III definitions), 2012-17

Note: Data are from a consistent sample of 84 large internationally active banks with capital in excess of €3 billion. Source: Basel Committee on Banking Supervision, Quantitative Impact Studies (various years).

Note: Data are from a consistent sample of 84 large internationally active banks with capital in excess of €3 billion. Source: Basel Committee on Banking Supervision, Quantitative Impact Studies (various years).

Yet, views vary widely on whether current requirements and capital levels are sufficient. At one extreme, narrow banking advocates continue to call for depository institutions to finance anything but riskless assets with 100% equity capital. Admati and Hellwig argue that banks should operate with equity capital closer to 25% of total exposure. The Minneapolis Plan to End Too Big to Fail and IMF researchers (see Dagher et al) conclude that 15% would be sufficient, while Cline suggests that a leverage ratio in the range of 8% strikes the right balance between growth and stability. All of these are significantly higher than current norms.

What about risk management professionals? How do they answer when asked: “Are capital requirements high enough?” The distribution of responses from the GARP Convention attendees is striking (see the chart below). Only one fourth of the respondents believe that the current Basel III requirement of 3% is sufficient. The median response is 15%, and the weighted average is 13%. That is, people who manage risk for a living—including those that work for financial institutions—believe that current levels of capital (about 6% of total exposure) are insufficient, and would prefer capital buffers roughly twice as large (and requirements that are four times higher than the Basel III minimum)!

Distribution of risk professional survey responses (percent): “Are capital requirements high enough?”

Note: The bar associated with each leverage requirement shows the share of the respondents who selected that option. Source: Courtesy of the organizers of the 20th GARP Risk Convention.

Note: The bar associated with each leverage requirement shows the share of the respondents who selected that option. Source: Courtesy of the organizers of the 20th GARP Risk Convention.

Turning to central clearing parties (CCPs), as we discuss in an earlier post, we support the G20-driven shift of derivatives trading from bilateral arrangements to central clearing. Properly managed CCPs diminish direct linkages among intermediaries, facilitate the enforcement of uniform collateral and margin standards, and (by making risk concentrations transparent) help impose commensurate risk premia.

But the shift to central clearing also creates a new source of risk: the CCPs themselves. Because their business typically benefits from economies of scale and scope, they naturally become very large. A number of them have become critical elements of the financial infrastructure for which there are no short-run substitutes.

To ensure resilience, each CCP has a self-insurance mechanism (called a “waterfall”) that—much like bank capital—is designed to absorb losses from a member default and allow the CCP to continue uninterrupted operation (see here). This waterfall includes the margin or collateral posted by parties to a transaction; clearing members’ prepaid guarantee fund; a small sliver of capital; and a loss-sharing arrangement, where a CCP’s clearing members agree to contribute after a shock exhausts all other resources.

Are these buffers sufficient? First, keep in mind just how big the biggest CCPs have become. LCH, the largest European CCP, has roughly $330 trillion in gross notional exposure! Compared to that, the resources available to LCH in the event of a clearing member default include just under $150 billion in collateral and about $22 billion in capital and guarantees. That is, collateral constitutes less than 5 basis points of the gross notional exposure; while the sum of own capital, the prepaid guarantee fund, and the potential assessment on clearing members sum to less than one basis point.

If CCPs do not have sufficient resources to weather a period of financial stress, we see three policy options to ensure their resilience. The first is to require greater disclosure from the CCPs. That could motivate the healthiest clearing members—typically global banks—to limit the risk of a crisis call on their resources by imposing greater discipline on the CCPs. A second possibility is to force each CCP to increase loss-absorbing resources—the waterfall of margin requirements, prepaid guarantee funds and the CCP’s own capital. Finally, there is the option of converting what are now private (in some cases, for-profit) CCPs into public utilities. Because the authorities cannot credibly commit to let them fail in a crisis, this conversion would make what is now an implicit CCP backstop explicit.

After explaining these options to GARP members, how do they wish to manage CCP risk? Nearly 90% of respondents believe that CCPs either need more internal resources (margin, capital and guarantee funds) or should be converted into public utilities (see chart).

Distribution of risk professional survey responses (percent): “How should we manage CCP risk?”

Note: The bar associated with each alternative shows the share of the respondents who selected that option. Source: Courtesy of the organizers of the 20th GARP Risk Convention.

Note: The bar associated with each alternative shows the share of the respondents who selected that option. Source: Courtesy of the organizers of the 20th GARP Risk Convention.

Our conclusion from all of this is two-fold. First, it makes sense to consult independent risk-management professionals about how to make the financial system more resilient. People who manage risk for a living understand when not to take risk (see Buffett quote). They also know that properly managing risk has a cost (see Cohn citation). Second, while the post-crisis regulatory reforms are a big step forward, we still have work to do to make the system resilient. In two important categories—bank capital and the resilience of CCPs—we need bigger buffers.

Acknowledgement: We thank Jeremy Davis of GARP for sharing data from the 20th GARP Risk Convention.