Executive Summary

• A quarter of record-breaking and near record-breaking results for the world’s largest banks has been capped by promising results in the Fed’s annual stress testing, allowing most banks to increase their dividends and share buybacks.
• All 35 banks held sufficient capital to stave off disaster; the Fed failed only one entity, a U.S. subsidiary of Deutsche Bank, for inadequate internal controls.
• The Tax Cuts and Jobs Act includes a number of provisions that combined to diminish banks’ capital ratios incrementally in the scenario contemplated in the stress-tests.
• The Fed proposes to integrate stress testing with the Comprehensive Capital Analysis and Review assessment process, doubling down on the blunt and speculative GSIB surcharge as the primary means of regulating the world’s largest banks.

An introduction to stress testing

The 2007-2008 financial crisis, and the failure and subsequent bailout of some of the world’s largest financial services firms, indicated a need for widescale reform. In late 2011 the Financial Stability Board (FSB) published a list of global systemically important banks (GSIBs); the Basel Committee on Banking Supervision followed up on this by introducing enhanced oversight rules for GSIBs. These laws were subsequently enacted in some form or other by participating Member States, including the United States. In the United States, the Federal Reserve (the Fed) and the Office of the Comptroller of the Currency (OCC) regulate banks with more than $50 billion in assets. The resulting enhanced oversight for GSIBs includes additional capital requirements, stress testing and the creation of a “living will”.

Since the financial crisis, GSIBs have undergone annual stress testing. Also known as scenario testing, the Fed posits a hypothetical disaster or stress scenario with the intention of testing whether banks could withstand systemic shocks. The Fed and OCC did not make the results of these tests public initially, but when eventually released by the Fed, the stress test results appeared to stabilize markets, as investors could derive comfort from the results. This year is the first that the Fed has made public the results of US subsidiaries of foreign banks.

Annual stress testing is made up of two phases. The first phase, considered a lower hurdle, measures whether banks are holding sufficient capital in the event of hypothetical catastrophic losses. The second phase is considered more stringent and focuses on a bank’s capital plan, including cash the bank intends to return to shareholders. The results of stress tests are closely followed as the Fed uses them to determine which firms it allows to issue dividends and share buybacks.

The 2018 results

This year the 35 entities that made up the Fed’s test group (covering about 80 percent of all bank assets in the United States) faced their most intensive stress testing yet. Banks were subjected to a hypothetical 9.6 percent contraction in the economy, double-digit unemployment, the collapse of home prices, and presumably a plague of locusts. The Fed estimated these stressors would lead to roughly $578 billion in total losses at the banks.

All subjects passed the first phase of testing. Industry then held its breath awaiting the results of the second phase, the Fed’s analysis of banks’ capital plans. 31 of 35 entities received clean passes. The Fed failed a U.S. banking subsidiary of Deutsche Bank, Deutsche Bank AG, and gave dishonorable mentions to State Street, Goldman Sachs, and Morgan Stanley.

Analysts had predicted that Deutsche Bank AG would fail the stress test after a rocky few years for the firm worldwide. The Fed found “widespread and critical deficiencies across the firm’s capital-planning practices” in addition to the firm’s data capabilities. State Street was granted a “conditional” pass, although cautioned for inadequate internal counterparty risk recognition.

Goldman Sachs and Morgan Stanley too achieved only conditional passes, as capital at both firms dipped under minimums required in stress testing. Despite technically failing the stress test, the Fed, in an act of forbearance unprecedented since the financial crisis, passed both firms conditionally, noting the impacts of the 2017 U.S. corporate tax cut.

Implications of the 2017 corporate tax cut

The Tax Cuts and Jobs Act (TCJA) contained a number of changes to the taxation of U.S. businesses. These changes include a reduction in the corporate income tax rate and limitations to tax provisions that allowed corporations to smooth out reported income for tax purposes, known as net operating loss (NOL) carrybacks and carryforwards. The TCJA also imposed a one-time tax on the accumulated overseas earnings of U.S. multinational firms. Taken as a whole, these provisions had, on average, a negative effect of 30 basis points on firms’ post-stress capital ratios.

Two elements of the TCJA immediately reduced the amount of regulatory capital that firms held at the end of 2017. Specifically, the reduction in the corporate tax rate from 35 to 21 percent reduced the value of deferred tax assets (DTAs), such as NOLs, that many firms carried on their books. Deferred tax assets are valued on the basis by which they can reduce tax liability. Reduced tax rates thus reduce the value of those assets held by firms. Firms were also required to recognize the TCJA’s overseas earnings tax liability in 2017, though they have eight years to make the actual payments. These revaluations were one-time events however, which the Fed recognized in evaluating firms’ capital plans.

Some of the business tax changes in the TCJA uniquely affect firms that are losing money, as contemplated in the stress tests. Specifically, the TCJA eliminated the ability of firms to apply a current NOL against taxable earnings in the prior two years and receive a tax rebate – a practice known as NOL carryback. The TCJA also reduced the degree to which firms could apply past losses against future tax payments – known as NOL carryforwards. Prior to the TCJA, firms could offset as much as 100 percent of their taxable income with past losses for 20 years; the TCJA limited that percentage to 80 but over an indefinite period. In general these effects reduce after-tax income for firms with losses, as would be the case in the adverse scenario envisaged by the stress-tests.

What happens next?

Markets responded well to the news, with share prices of banks that received passes rising 1-2 percent. Even Deutsche Bank saw a 1.5 percent improvement, indicating investors are taking a long view. Deutsche AG’s failure will prevent repatriation of profits back to headquarters in Germany; the firm will also have to remediate internal controls. Goldman Sachs and Morgan Stanley will not be permitted to increase the amounts they return to shareholders this year beyond the levels of previous years, with both firms still planning to return over $6 billion to shareholders. For the shareholders of the firms that passed the test cleanly, major dividend returns are possible; Bank of America, for example, plans to increase their dividend 25 percent and buy back $20.6 billion in stock. Wells Fargo and Citigroup got permission to pay out more than 100 percent of expected profits over the next year, an indication that the Fed believes those firms can actually reduce their loss absorbing capacity.

Despite market buoyancy recent proposals in the administration of stress tests might cause some concern. Although midsize banks will be delighted that the threshold for stress testing is to change from banks with $50 billion in assets to $100 billion, the Fed has proposed integrating stress testing with the Comprehensive Capital Analysis and Review (CCAR). The CCAR is the mechanism by which the Fed determines a bank’s minimum required capital. By using stress tests to determine, rather than test, capital levels, the Fed has indicated that it will be placing more weight on blunt capital surcharges for GSIBs. GSIBs have a 1.5 – 2.5 percent capital surcharge of risk weighted assets they must maintain on top of minimum required capital. This additional requirement is criticized because it is largely untethered from measurable data and instead represents a fear-induced penalty. The Fed’s proposals expand the scope of the GSIB surcharge, requiring GSIBs simultaneously to maintain a minimum capital requirement, a stressed capital requirement, and a GSIB surcharge. The Fed’s movement toward less calibrated rather than more calibrated capital requirements is troubling.

Conclusions

Market watchers should be delighted that the world’s largest banks are meeting and exceeding the most stringent of stress testing. That Goldman Sachs and Morgan Stanley passed conditionally is understandable, given the one-time effect of the corporate tax cut on their capital; the reduced corporate tax rate will enhance their earnings in the future.

Stress testing remains the most – if not only – nuanced capital assessment tool the Fed has its disposal. Banks should have discretion to determine their individual tolerances for working capital. The stress test remains the most effective means of testing these assumptions but should not supplant them.