|Bank of America received an "incomplete" on the latest stress test by the Federal Reserve|
For the top 50 or so banks, this month's issue of the moment is the formidable "stress test." The Federal Reserve presides over a bank test to measure whether large financial institutions can survive worst-case scenarios without threatening the financial system and jeopardizing required payouts to depositors, trading counterparties, and long-term investors.
The tests are hypothetical, but regulators treat them as if the scenarios will occur, in the way we observed worst-case scenarios of all kinds during last decade's financial crisis. The tests are required by Dodd-Frank regulation and formally fall under the banner "Comprehensive Capital Analysis and Review," or "CCAR."
After the crisis, regulators around the globe concluded bank capital at sufficent amounts is the primary safeguard to protect the system when the next crisis occurs. In finance, investors often proclaim "cash is king." In bank regulation, regulators (and risk managers, too) exclaim "capital (as in capital cushion) is king."
It's all about capital, capital as a balance-sheet cushion to ensure that a bank enduring hard times or struggling through a volatile market will have enough of a capital base to permit the bank to meet all obligations to depositors, creditors, and other counterparties.
Capital as a Cushion
The principle of having "sufficient or adequate capital" is that balance-sheet or operating losses will be absorbed by shareholders. Banks with significant amounts of capital can tolerate a certain amount of losses. In worst-case scenarios, losses might be inevitable, but the best-capitalized institutions will survive and continue as ongoing institutions. Stress-test losses include projected losses from misbehaving capital markets, from mismanged operations, and from unforeseen events (even losses attributed to "tail events"). Most of all, test-givers want the bank to absorb these losses without having to resort to or rely on (or count on) support from regulators or the government.
Let's present a simple example. A bank, say, tallies its balance sheet and capital and computes them to be $15 billion and $1 billion, respectively. Is the $1 billion an adequate amount of capital for both ongoing operations and for a worst-case, stressed scenario? It assesses the risk of each item on the balance sheet and evaluates its overall operating activity. All activities (on and off the balance sheet) are "risk-adjusted." (In "risk adjustments," $1 million in cash or U.S. Treasury securities has much less risk than $1 million in loans to a BB-rated oil-drilling company.)
All activities (on and off the balance sheet) are then "stressed," based on market and economic scenarios presented by regulators: Unforeseen events occur, interest rates move sharply, equity and commodity markets become volatile, bank borrowers default, depositors flee, and banks have no access to new funding or new capital. When quantified precisely, what is the maximum of losses in this scenario? And what impact do they have on the capital base and the bank's ability to survive without difficulty?
In our example, suppose the bank (with regulators looking over the bank's shoulders) computes the maximum worst-case loss total to be $300 million. That leaves $700 million in post-stressed capital. The tests help decide whether that amount is still a sufficient to anchor a risk-adjusted balance sheet. Regulators determine:
(a) whether this remnant $700 million capital is satisfactory (and "passes the test"),
(b) whether it's satistfactory, but too borderline, or
(c) whether it's not satisfactory and the bank has failed the test.
If regulators determine it's (b) or (c), then remedial action will be recommended or must be taken. That usually means the bank must reduce dividends or shareholder payouts (to ensure capital won't decline) or it must raise new equity over a defined timeframe.
Recent Stress Tests
The stress tests are exercises, but regulators take them seriously and publish the results widely. Therefore, stock markets, investors, and banks collectively take this seriously. Markets react, investors might express concern by selling shares without thought, and banks pour resouces by the hundreds of millions to ensure they pass these tests and appear at the top of industry lists of the best-capitalized banks and the banks best prepared for a crisis. (Citi's CEO Michael Corbat last year even promised to resign if Citi failed a stress test in 2015. It passed; he won't have to resign, and it was reported Citi's trading floors applauded wildly.)
In the most recent test (Mar., 2015), results show 28 of 31 banks passed the test, but not all in a blaze of glory. The U.S. operating entities of Deustche Bank and Santander failed. Bank of America received an "incomplete" grade, partly because regulators want more time to review the bank's plans to reduce capital with dividend payouts and share repurchases.
On most industry scorecards or league lists, Goldman Sachs roams near the top (the top of the most profitable banks, near the top of mergers-acquisition and equity-underwriting lists, etc.). For this round of stress tests, it appears in the bottom half (although with a passing assessment), despite its prowess in just about all banking activities it engages in. But its lukewarm stress-test outcome was enough to push down its stock price a notch or two in the past week.
Markets react because there are perceptions that banks (Goldman included) have more embedded risks than previously acknowledged, that banks are vulnerable and susceptible to sharp losses at any time, and that banks won't be able to pay the same attractive dividends many investors are accustomed to.
Banks, whether they are Bank of America or Citi, aren't shy with regulators and will challenge the test results, line by line, asset category by asset category. Banks don't want outsiders to meddle with their efforts to please shareholders with predictable dividend payouts. Regulators, nonetheless, usually have the last word. In the end, the test results (and trends in how banks perform on these tests) have enormous impact, not only on how markets and investors perceive banks, but in determining banks' long-term strategies and who they want and can be.
The Impact on Returns: Inside the CFO's Office
While regulators are obsessed with capital as a cushion, banks are obsessed and anxious about whether capital is allocated properly for various operations, businesses and balance sheets. Capital must be abundant and in excess, but it must be assigned to bank operations precisely and aptly--more than just enough, but not too much.
If a bank's business requires too much capital (in the eyes of regulators), then banks won't be able to achieve high returns or won't be able to justify the risks that imply more capital. Some won't be able to raise the capital they need or will be restricted from paying dividends in order to reach optimal capital levels.
Peek inside a bank CFO's office, and watch the staff hold heated, vibrant discussions of bank capital, not just capital related to regulatory requirements, but capital related to risks, operations, funding strategies, and economics. They debate, analyze and compute: Do we have enough? How do we allocate? How do we meet regulatory requirements with ample excess? There are capital requirements defined by Basel III regulation, defined by Dodd-Frank, defined by European regulators, and defined by Federal Reserve edicts applicable to the largest banks.
CFO offices, along with colleagues in Risk Management, must also resolve the following:
a) How do we perform the complex calculations regulators require to determine whether we have enough and do so before deadlines? (Many big banks, like Citi, hire experienced senior financial managers to perform calculations and to ensure capital is adequate everyday, not just quarterly or for stress tests.) Do we cut dividends (and by how much)?
b) Do we cancel stock repurchases to improve capital ratios and present better test results?
c) Because of limited capital, how do we decide which businesses deserve it?
d) Should we "de-leverage" the balance sheet, sell off risky assets and trading positions, and shut down designated business lines and branches to improve our capital showing?
e) And perhaps most important: With all these capital requirements and the challenge to allocate precisely, how do we generate the returns investors expect (over 10%?) on this capital?
Some regulators, academics and others argue that capital requirements and capital stress tests at banks aren't strict enough and that banks have much leeway and too much time to comply. They seem to agree that if there had been tough capital requirements and other restrictions in the mid-2000's, Bear Stearns and Lehman might have survived. (They would have survived because with more capital and capital rising to prudent amounts, they would have been able absorb eventual losses or, with the same levels of capital, they would have not beeen able to take risks and assume leverage for the capital they had.)
What Do These Tests Involve?
The tests are complex, because bank balance sheets are complex, piled with trading assets, loans of all kinds, exotic derivatives and financial instruments fluctuating in value on the ledgers of subsidiaries around the world . The tests involve many definitions or "tiers" of capital and have spawned financing structures where some liabilities can turn into capital under certain conditions or behave or appear to be capital-like (e.g., convertible debt, subordinated debt).
The principles of a stress test are straightforward, as shown in the example above. The Federal Reserve examines the consolidated bank, while other regulators may examine the specific regulated bank entity. The consolidated balance sheet of total assets is buttressed by a current amount capital. As described, those assets are "risk-adjusted" (based on regulatory guidelines that are influenced by historical trends and internal bank calculations and interpretations of risk).
A "stress" scenario is summarized and quantified. "Stress" implies a nightmarish market or economic scenario similar to what we observed in crises past. Tests try, also, to project distress that isn't projected, known or accounted for. "Stress," in these tests, implies losses--sudden losses and sustained losses over a time period. A stress scenario might entail a collapse in capital markets and increased probabilities of default in bank loans (including corporate loans, credit cards, and mortgages).
Specific ratios and standards determine whether capital as a percentage of total risk-adjusted assets is sufficient or inadequate. If tests indicate capital is borderline, then regulators will insist the bank adopt a capital strategy devised by regulators. That strategy could prevent banks from paying dividends or repurchasing stock until capital levels reach more comfortable levels (accounting also for earnings and new injections). Regulators, now more than ever, want to ensure banks have not just for rainy days, but for storms and the unknowable lightning strike.
Dividend restrictions frustrate bank managers and boards. Historically, banks like to please shareholders with dividends and occasional share buybacks. Investors in financial institutions buy these shares after they observe multi-year practices of stable (and sometimes increasing) dividend payouts and intermittent share repurchases. And there emerges friction, when regulators present analysis that shows banks don't have enough capital, and banks respond with a financial argument to prove they are so well-capitalized they should have discretion to pay something to shareholders each quarter.
What Are the Long-term Implications?
The rules, the tests and the occasional confrontations between bankers and regulators have long-term impact:
1) Banks have become more precise and careful in computing capital requirements.
Just a decade or two ago, capital computations might have been done by a small team with files of spreadsheets. Capital computation and managing requirements now rely on technology systems, data feeds, and armies of bank professionals to monitor where a bank stands everyday. Computation is complex, too, because big banks measure assets in amounts that hover about a trillion, not in the billions of days gone by.
2) For banks, capital-ratio computation and risk allocation have evolved into a science relying on data management, data quality and technology, as they attempt to use capital wisely and prudently (and profitably).
They ask difficult questions with no easy answers: How do we allocate and invest capital in such a way that
(a) operating needs are fulfilled (Capital acts as cushion, but it still must fund assets and bank activity),
(b) risks are managed,
(c) regulators are pleased, while
(d) investors get the return they wanted?
And how do we allocate and invest in a way such that there is still some excess to keep regulators satisfied and assured the bank can thrive in the toughest of times?
Banks look for precise answers and avoid estimates. The stakes are too high in Washington and on Wall Street.
3) Banks won't hesitate to exit high-risk business lines that require too much capital, because of the detrimental impact on regulatory ratios and because of the near impossibility of achieving reasonable returns (ROE) (because of high capital requirements).
Over the past five years, banks have withdrawn from businesses partly because of new rules (e.g., Volcker rules that bar proprietary trading), but also because a business might require too much added capital for the same level of earnings. For this reason, banks have shut down trading desks, withdrawn from some activities, pared back in areas that use too much balance sheet, avoided risks that can't be offset by reasonable rewards, and canceled privileges it once offered clients (e.g., overdraft lines).
Banks will continue to evaluate business sectors on this basis. In assessing business activity, the earnings and projected revenue growth might exist. But if risk calculations imply capital requirements could be too onerous, the bank will avoid expansion or might even withdraw altogether.
4) Bank finance managers must always have a long-term capital plan.
Many will say they have had capital plans for decades. Yet now the capital plan will be scrutinized and opined by regulators. Banks will, too, always need a realistic, detailed strategy for how to increase the capital base when it needs to, when it can increase dividiends, and when it is appropriate to conduct stock repurchases.
A long-term plan guides the bank on when dividends should be decreased, when it must issue new shares, borrow more in long-term debt markets, or sell large stakes to institutional investors (or Warren Buffet, who purchased his large stake in Goldman Sachs during the crisis). In its waning days, without the benefit of a plan beforehand, Bear Stearns and Lehman rushed to find new investors to inject life and capital onto wiltering balance sheets. They looked abroad, they looked for advice, and they pleaded with their counterparties and creditors to be patient. These were the days before the U.S. Treasury decided to inject billions into banks once the crisis ballooned.
Bank CEO's and CFO's continue to have arduous tasks before them. They set objectives to promise returns on investment to shareholders in the 10-12% range. Yet the denominator in this ROE ratio is elusive and spiraling upward. Just as they implement a strategy to boost revenues and earnings (the numerator), often accompanied by new risks, they must boost the capital cushion. The ROE returns, it seems, can't seem to budge upward.
CFN: Basel III Becoming Real, 2013
CFN: Keeping Up: Basel III, 2013
CFN: Big Banks and Dreadful Downgrades, 2012
CFN: Wells Fargo: Sticking to What It Does Best, 2014
CFN: Why Did Citi's CEO Resign? 2012
CFN: This Fall's Big-Bank Thorn: FX, 2014
CFN: JPMorgan's Refined Regulatory Strategy, 2014