Why Citigroup's Stress Test Failure Shouldn't Be So Surprising

Flash back to your days in school and the stressful day that report cards were handed out and you will get a sense of what top bank executives were going through on March 26.

On that day, the Federal Reserve announced the results of its annual stress tests, which assess the financial health of the nation's largest banks. Just like on report card day, recipients of passing grades were left smiling, while other were left to contemplate the implications of a failing grade. Among the banks that failed the test are Citigroup, HSBC North America Holdings, RBS Citizens Financial Group, and Santander Holdings. The Fed also rejected the capital plan from Zions Bancorp because it did not meet minimum post-stress capital requirements.

The Dodd-Frank Act requires banks with consolidated assets of more than $10 billion to conduct annual stress tests that assess their ability to withstand adverse economic conditions, such as spiking interest rates and high unemployment. The Fed also considers the overall strength of the bank's capital planning process. The Fed examines banks with more than $50 billion in assets through its Comprehensive Capital Analysis and Review. If the Fed objects to a capital plan, the bank may only make capital distributions—including dividend payments to shareholders and stock buy-backs—with written approval.

For five of the nation's biggest banks angry shareholders are asking how, with all their risk-management and compliance resources, they came up short. The problem—and why failures likely came as a shock to many—is that the tests are as subjective as they are quantitative. Asking whether the bank's balance sheets could survive an economic calamity without collapsing or running to the government for a bailout? That is straightforward math. How the Fed links those numbers to internal controls, audits, and governance is more subjective, like grading an essay where the test administrator and the test taker can have differing views on how well they did.  

“It's the most stressful week on the street among the big banks,” says Dan Ryan, leader of PwC's financial services advisory practice. “I was meeting with one of the banks that passed swimmingly, but they were still very worried. The fear of failing in the public eye is a fate no executive wants to suffer. The biggest tool the Fed has is to embarrass you by putting your bank on the front page of the Wall Street Journal. They are using this to achieve their supervisory objectives.”

HSBC and RBS Citizens were singled out for inadequate governance and weak internal controls around their capital planning processes. Deficiencies were found in RBS Citizens' practices for estimating revenue and losses under a stress scenario across all business lines. HSBC had deficiencies in its practices for estimating revenue and losses for material aspects of its operations. Santander was cited for “widespread and significant deficiencies” across its capital planning processes including governance, internal controls, risk identification and risk management, and its management information system.  The Fed objected to Zions' capital plan because its post-stress ratio of high-quality, liquid assets measured against debt and riskier assets fell below the 5 percent minimum, to 3.6 percent.

Citibank, the largest on the list, has failed twice in the last three years, despite ongoing negotiations with the Fed and a to-do list of specific grievances regulators wanted to see fixed. While it made “considerable progress” improving risk-management and control practices over the past several years, its capital plan, according to the Fed, still had numerous deficiencies, including its ability to project revenue and losses throughout its global operations and internal testing that failed to reflect its full range of business activities and exposures.

A Citigroup Surprise

The failure reportedly caught Citigroup management off-guard. Should it have?

Citigroup's magnitude, complexity, and multitude of business lines likely played a role in its troubles, suggests John Reed, former chairman and CEO of Citigroup. “Citigroup is immensely large and geographically dispersed,” he told attendees at a financial conference in Boston last week. “That complexity makes for a very difficult management structure. With hundreds of thousands of employees, spreading compliance and risk management discipline throughout the entire organization can be an incredibly difficult task.”

“You can imagine why it would be very difficult for them to respond to the requests they received from the Fed,” he added. “The fact they have had four CEOs over the last 10 years can't have helped.”

Not everyone is blaming the bank. Some say the subjective nature of the test was too difficult for banks to navigate. Also, the Fed's expectations and banks' assumptions didn't necessarily mix well. “Most of the banks that failed are befuddled,” Ryan says. “The qualitative test is obviously a much more subjective process and you can understand the frustration of bank executives.”

“The biggest tool the Fed has is to embarrass you by putting your bank on the front page of the Wall Street Journal. They are using this to achieve their supervisory objectives.”

—Dan Ryan,

Leader, Financial Services Advisory Practice,

PwC

More than one aspect of the stress tests caught banks off-guard, he says. This year, regulators assumed bank balance sheets would grow in the future, while most banks predicted they would remain flat or go down due to the heightened capital demands of Basel III.

Another conflict was that banks estimate future revenues and potential losses based on their own historical results. The Fed, however, echoes the boilerplate language of an investment prospectus: “Past results are not indicative of future performance.” Also, this year, despite big banks growing their capital base, their actual ability to pay dividends went down in the eyes of the Fed because more loans come with a greater need to absorb risks. Weaknesses uncovered through the normal examination process are also factored in.

The Fed uses the stress tests as a means to assess whether banks link the way they manage risk with capital actions. “What the Fed is looking for can't be done on a spreadsheet in the treasurers' or chief risk officer's office,” Ryan says. “To the extent that you have models and scenarios, they want it all to be clearly linked and tie back to the basic risk framework of a company.”

Banks also have to develop their own worst-case scenarios to supplement those presented by the Fed. “The nightmare scenario for your bank might not be reflected in the canned scenarios they gave you,” says David O'Connell, senior analyst with Aite Group, a research and advisory firm for the banking industry. “They want to know what iceberg-under-the-water scenario will get your bank into trouble and what would happen.”

To craft those scenarios, banks need to have an incredibly accurate view of all of their risks, O'Connell says. “It is actually pretty hard for banks to do that. If it sounds like I'm cutting the banks some slack, I am,” he says.

BEYOND SPREADSHEETS

The following is an explanation of the “qualitative” criteria used by the Federal Reserve for its 2013 stress testing of big banks.

The qualitative assessments carried out by the Federal Reserve are a critical component of the CCAR review. Bank holding companies that were part of Comprehensive Capital Analysis and Review this year differ significantly in their size, complexity, geographic footprint, and business models. Reflecting these differences, each firm is expected to focus on the idiosyncratic risks it faces when conducting its internal stress tests and capital planning. Such a focus on individual risks is not something that can be fully captured when running a standardized stress test.

Therefore, even if the supervisory stress test for a given BHC results in a post-stress tier 1 common ratio exceeding 5 percent and post-stress regulatory capital ratios above the minimum requirements, the Federal Reserve could object to that BHC's capital plan based on a qualitative assessment of the practices supporting its capital planning.6 In the CCAR qualitative assessment, the Federal Reserve evaluated:

The extent to which the analysis underlying each BHC's capital plan captured and appropriately addressed potential risks stemming from all activities across the consolidated institution under baseline and stressed operating conditions.

The robustness of the BHC's capital planning process, including supporting risk-identification, risk measurement, and risk-management practices; the reasonableness of the assumptions and analysis underlying the capital plan.

Corporate governance and internal controls over the capital-planning process, including the BHC's capital policies as approved by its board of directors.

The Federal Reserve's qualitative assessment of BHCs' capital plans in CCAR 2014 reflects differing expectations across the various aspects of BHCs' capital planning processes for BHCs of different sizes, scopes of operations, activities, and systemic importance. For example, the Federal Reserve has significantly heightened supervisory expectations for the largest and most complex BHCs—in all aspects of capital planning—and expects these BHCs to have the most sophisticated, comprehensive, and robust capital planning practices.

Source: Federal Reserve.

The subjective nature of the stress tests, while challenging for banks, is an understandable reaction to the Financial Crisis, says Mark Williams a lecturer at Boston University, risk management practitioner, and former examiner for the Federal Reserve Bank.

For many years, banks have been judged by CAMELS ratings system (capital adequacy, assets, management capability, earnings, liquidity—also called asset liability management, and sensitivity to market risk). “What the Fed is doing is setting up the opportunity to put more qualitative attributes within the process,” he says. “Banks that have more capital are less likely to fail. But we also learned during the Financial Crisis that banks with excessive risk taking and poor corporate governance were more likely to fail. It was the qualitative attributes that really came into play as reasons banks got into trouble and capital ratios alone didn't capture those risks.”

Nevertheless, the Fed's current approach is deeply flawed, says Sim Segal, fellow of the Society of Actuaries, an enterprise risk management expert, and president of SimErgy Consulting. “The Fed is not asking the right questions or looking in the right places,” he says. “It is looking at one set of economic shocks at a time, not combinations of things.”

The Fed also “paints with a broad brush,” using the same stress tests for each bank, even if they will react differently to the same event, he says. “It is a one-size-fits-all approach. The frightening thing is they are not asking for a range of risks; they are looking for Armageddon. These stress tests are not realistic.”

Executives at banks that passed this year's stress test shouldn't breathe too easy, Ryan warns. “The way they may look at this is, ‘phew, thankfully it wasn't us.' But the message from regulators is that the bar is going to keep getting raised and banks need to continually up their game. Just because you passed this year, doesn't mean you are going to pass next year,” he says.