Get to Know the Fed's Other Stress Test

WASHINGTON — As the Federal Reserve moves forward with its implementation of the Basel Committee requirements for long-term liquidity, many are questioning why they are necessary in light of the agency's existing liquidity stress-testing regime.

The Comprehensive Liquidity Assessment and Review — not to be confused with its better-known stress test cousin, the Comprehensive Capital Analysis and Review — is an annual examination of the largest banks' liquidity positions and risk management, and has been in place since 2012. The program subjects each of the largest banks to horizontal qualitative and quantitative analyses to determine whether their liquidity plans are up to snuff.

But CLAR is not as well known and certainly not as well understood in the regulatory landscape as CCAR, and part of that is because its inner workings are unknown.

"I don't think that's ever been made public," said Karen Shaw Petrou, managing partner of Federal Financial Analytics. "They may have described it publicly, but the conditions, the terms — it's not like CCAR, where there's rule after rule after rule."

Instead, CLAR is implemented as part of the enhanced supervision by the Large Institution Supervision Coordinating Committee, or LISCC — a body within the Fed that oversees U.S.-based global systemically important banks, U.S. operations of foreign G-SIBs, and nonbanks designated by the Financial Stability Oversight Council (though to date, only the banks have been subject to CLAR, though the nonbanks are subject to other liquidity supervision).

Industry representatives are wondering why a new proposal released by the Fed in April, which would require banks to maintain a net stable funding ratio, or enough stable funding for a year, is necessary given the Fed's existing liquidity test.

"The NSFR project was begun before a lot of these other reforms were put into place," said Wayne Abernathy, head of regulatory affairs at the American Bankers Association. "Those talks [in Basel] didn't really keep up with all of the other various ways that the regulators were looking at liquidity. And the NSFR really does seem incredibly redundant … and distracting, rather than helpful."

Their argument hinges on understanding more about what CLAR actually is.

Fed Gov. Daniel Tarullo, who oversees LISCC as head of the Board's Committee on Bank Supervision, described CLAR in a November 2014 speech as a "complement" to the NSFR and the shorter-term liquidity coverage ratio by examining the LISCC banks through quantitative metrics and qualitative examinations of the banks internal liquidity models — just as CCAR includes both quantitative and qualitative aspects. The tests also give regulators a better picture into how liquidity is performing throughout the economy, Tarullo said.

"Because CLAR assesses all LISCC firms simultaneously, the Federal Reserve is also able to compare the range of practices in liquidity risk management across the portfolio," Tarullo said. "Knowledge gained through CLAR assessments also provides a macroprudential perspective on liquidity vulnerabilities and funding concentrations in the system as a whole."

The biggest and most prominent difference between CCAR and CLAR is that banks cannot fail the liquidity test — there is no minimum liquidity threshold below which banks may not fall. Banks are subject to minimum liquidity standards as part of the LCR (and, eventually, as part of the NSFR). But CLAR isn't a process for finding out whether banks meet those standards — either the banks have the required level of high-quality liquid assets on hand or they don't.

Another significant difference is that CLAR doesn't run banks through hypothetical stress scenarios the way CCAR does. Rather, it consists of two separate analyses: an examination of banks' liquidity positions according to certain metrics — concentration, for example — and a separate qualitative assessment of their internal liquidity models. And because CLAR is limited to the biggest firms, it's a considerably smaller exercise — 12 banks undergo CLAR assessment, versus 33 banks participating in CCAR in 2016.

Abernathy said that banks do not look forward to or enjoy the CLAR examinations, but they don't object to them either, calling them "an annoyance," but not an undue burden. He said much of the reason why CLAR is not very well known is because it's not a public exercise — there's no filing deadline, no public event announcing the results, and most importantly no "test" to "fail."

That's by design — CCAR and the Dodd-Frank Act stress test are public processes in part to assure consumers that banks were adequately capitalized after the crisis revealed that many were not. Dodd-Frank mandated that the Fed examine liquidity as well, but was far less explicit in how exactly it fulfilled that mandate.

Tarullo, in that same 2014 speech, said the relationship between Basel liquidity rules and CLAR is "roughly comparable" to the relationship between Basel Capital rules and CCAR, and pointed out that one benefit of CLAR is its dynamism. Instead of a hard, firm rule, CLAR can change and adapt as liquidity changes in different markets.

"The very nature of quantitative liquidity standards means the effectiveness of the rules could wane over time because of changes in funding markets, reductions in the liquidity of assets previously deemed liquid, or regulatory arbitrage," Tarullo said. "Furthermore, it is difficult for any standardized quantitative liquidity regulation to capture all relevant risks."

It is for precisely this reason that banks object to the need for the net stable funding ratio — and to a lesser extent the LCR — at all. Abernathy said that CLAR strikes an appropriate balance by not only observing liquidity but also observing how it is being applied horizontally across institutions. By mandating or encouraging banks to hold certain assets over others in perpetuity, the Fed is potentially increasing instability by either soaking up liquid assets that are available or assuming the liquidity of assets that may be hard to sell in a given stress environment.

"Liquidity is a dynamic thing — what's liquid today may not be liquid tomorrow," Abernathy said. "And yet these rules are very static, and they assume certain qualities of liquidity that may not actually play out over time. And we're worried that because of that static nature, you disadvantage certain kinds of instruments that may actually be more liquid in whatever the stress may be in the future."

Oliver Ireland, a former Fed official and partner at Morrison & Foerster, said the main issue as it relates to LCR and NSFR is redundancy — because CLAR and other supervisory tools are already examining a bank's liquidity positions, the Basel ratios are applying a rigid standard where none is required. Redundancy can be useful, he said, but it comes at a cost.

"One theme that runs through Dodd-Frank, even if it's not expressed, is redundancy — they're going to go at the same problem a number of different ways," Ireland said. "That may add some benefit in terms of avoiding lapses, but whether it's worth the cost is another matter."

That relativistic view of liquidity is correct in principle, according to Mayra Rodriguez Valladares, managing principal at MRV Associates, but it misses the point. The LCR and NSFR together are intended to set limits on banks' ability to finance their operations with either overnight cash or repo transactions — a reliance whose shortcomings became apparent during the crisis. CLAR is important, arguably even more important than capital stress testing, but there have to be firm rules in place to get banks to fund their activities reliably, she said.

"Banks are addicted to this model that no other sector of the economy has, and we — the market, the media, the regulators, everybody — is allowing this addiction to continue," Valladares said. "We've got to cut people off — we have to teach them to have more stable sources of funding. This has become such a perverse system."

The Fed, for its part, appears to have anticipated this line of questioning from the banks. Michael Gibson, director of the Fed's division of banking supervision and regulation, said in response to a question from Tarullo during the May 3 public meeting on the NSFR proposal that the plan is a critical benchmark for supervisors to assess banks' liquidity and is not a substitute for CLAR, nor is CLAR a substitute for the Basel rules.

"These horizontal reviews allow our supervisors to benchmark across supervisory portfolios, identify outliers and help form a more complete view of liquidity vulnerabilities and funding concentrations in the system as a whole," Gibson said. "The NSFR would be an important addition to this framework by establishing a standardized quantitative minimum stable funding requirement."

The problem with CLAR, Valladares said, is not that it isn't rigorous enough, it's that there's no way to know how stringent it is at all since virtually no details about the process, its metrics, or the results are publicized. That could be because of legitimate concerns about subjecting a complicated and somewhat subjective matter like liquidity to widespread scrutiny, she said, or a consideration for how much additional cost would be entailed for both the banks and the Fed if it were a more transparent process.

"I think it would be good for the regulators to tell us, 'Hey, we're doing CLAR and we're looking for certain improvements in banks, and we found that this bank had a deficiency,' " Valladares said. "I can think of tons of things that I'd want to know that I don't know because none of it has been made public."

Abernathy said the CLAR process is not too opaque — rather it enables banks and regulators to speak frankly about issues that involve highly sensitive business information. If that were to change in the name of transparency, it would take what has been a relatively simple process and make it vastly more complex.

"There's a reason why have established bank supervision the way we have," Abernathy said. "Once it becomes a public exercise, then you bring in the attorneys and you bring in other compliance folks, and it is no longer a frank conversation."

Valladares countered that the idea that any transparency would hurt the process is probably overblown, since the Fed is unlikely to reconsider its approach to CLAR in the near future, and would not initiate the kind of public spectacle that the capital stress tests have become even if it did.

"The banks doth protest too much," Valladares said. "They're not ready to … do CLAR the way they do CCAR."

But Petrou said that CLAR's pattern of secrecy may actually exceed the limits of what the Fed is allowed to do in its supervisory capacity. The Fed has widespread authority to supervise banks, she said, but that authority is tempered by administrative procedure, which requires, among other things, that binding rules be publicized and that the public given a chance to comment. There's no way to know if CLAR exceeds those bounds, she said, because the only ones who know the details of the tests are the Fed and the LISCC banks.

"I don't question their statutory authority to do this," Petrou said. "I think the question is … do you get under the Administrative Procedure Act or other requirement by keeping it secret? Is this the equivalent of a rule?"