If you are a market-watcher or an industry insider in the financial markets, you might think that the banks are regulated more than enough and forced to hold more than enough capital in reserves. If you are a member of the public who doesn’t ever want to see another bank-led recession, you are apt to feel quite the opposite.
Monday brought three speeches from Federal Reserve presidents. One came from Fed Governor Daniel Tarullo at the Yale University School of Management Leaders Forum in New Haven, Connecticut.
Tarullo said that the Federal Reserve has substantially refined its supervisory stress test since the recession. Its stress tests have been integrated into the Fed’s Comprehensive Capital Analysis and Review (CCAR) aimed at approving capital distributions and safety of bank balance sheets.
24/7 Wall St. was surprised at the number of times that the term “buffer” was used in this speech. In fact, the formal use of “buffer” was used 28 times just in the prepared remarks for the speech. That did not even include the side-bar commentary.
Tarullo’s official commentary said:
As DFAST has evolved, it has become an increasingly valuable tool for evaluating whether the largest financial firms are holding sufficient capital to continue providing credit in the event of significant macroeconomic and financial stress. However, in itself DFAST does not set any capital ratios or limit any capital actions by the firms. Those functions are implemented through the CCAR program, which was created through the regulatory process by the Federal Reserve.3 In CCAR, the Federal Reserve assesses the overall capital adequacy of the firms–including evaluations of whether each firm’s capital provides an adequate buffer for the losses that would be incurred during the stress scenarios, whether its risk management and capital planning processes are appropriately well-developed and governed, and how its plans to distribute capital through dividends or share repurchases could affect its ability to remain a viable financial intermediary in the hypothesized scenarios. Given how central these considerations are to effective risk management and the soundness of these firms, we are progressively integrating the qualitative elements of CCAR into our year-round supervisory program for the largest banking organizations.
More specific commentary sure leans on the “buffer” term a lot, and it had to act to make investors and economists worry that even higher capital requirements may be proposed ahead. Tarullo’s “buffer” commentary said in a better alignment of CCAR with the new regulatory capital rules:
To better align CCAR with the regulatory capital rules as they now stand, we will be considering adoption of a “stress capital buffer” (SCB) approach to setting post-stress capital requirements that would further integrate our capital rules with our stress test and CCAR frameworks. The simplest way to describe it is that the SCB would replace the existing 2.5 percent CCB as a component in each CCAR firm’s point-in-time capital requirements. The SCB would be risk-sensitive and vary across firms, based on the results of the annual stress test. Specifically, it would be set equal to the maximum decline in a firm’s common equity tier 1 capital ratio under the severely adverse scenario of the supervisory stress test before the inclusion of the firm’s planned capital distributions, a simplification from the current practice of calculating the capital positions of the firm separately for each quarter in the scenario horizon.
As is now the case with the CCB and other applicable capital buffers, a firm would be subject to restrictions on its capital distributions whenever its capital levels fall below the combined regulatory minima and buffers. Although it is likely that the stress test losses will continue to exceed 2.5 percent of risk-weighted assets for most GSIBs, the SCB would have a specified floor at this current CCB level to avoid any reduction in the stringency of the regulatory capital rules. A firm’s buffer requirement would be recalculated after each year’s stress test, and its capital plan would not be approved in CCAR if the plan indicated that the firm would fall into the buffer under the stress test’s baseline projections.
Let me use a hypothetical example to illustrate how the SCB would work. Assuming that a firm’s common equity tier 1 capital ratio declines in CCAR’s severely adverse scenario from 13 percent to 8 percent, that firm’s SCB would be the greater of 2.5 percent or the 5 percent decline–thus, 5 percent. Assuming further that this firm is a GSIB with a surcharge of 3 percent and that the countercyclical buffer is not in effect, the firm would be constrained in making capital distributions that would bring its common equity tier 1 capital ratio under 12.5 percent. The 12.5 percent figure is the sum of the 4.5 percent minimum common equity tier 1 risk-weighted capital requirement, the 3 percent surcharge, and the 5 percent stress loss as calculated in annual stress test.
Because the GSIB capital surcharge already exists as an additional buffer requirement in the regulatory capital rules, the stress capital buffer approach would effectively add the GSIB capital surcharge to our estimates of the amount of capital needed under stress. This would generally result in a significant increase in capital requirements applicable to the GSIBs, for most of which CCAR has been their binding capital constraint. But having both the stress loss buffer and the capital surcharge included in our capital requirements is wholly consistent with the reasons for having a capital surcharge in the first place. Indeed, not having the two together has actually been an anomaly that arose from the sequencing of the various capital strengthening measures we have undertaken.
Whether or not this means more regulation is really coming to banks remains to be seen. Either way, bank investors and the analysts and portfolio managers who are targeting big upside for the low valuations generally do not want to consider even more layers of capital being set aside and more layers of regulation.