Following banking crises — large (think
One such debate currently engaging a lot of industry attention is the proper role of risk-weighted capital ratios in the context of both the recent crises and the
Unfortunately, by misidentifying the problem the authors have also misidentified the solution. With regard to
But this proposed leverage ratio doesn't fully correct the flaw that proved fatal for Silicon Valley Bank and First Republic Bank: Not all unrealized securities losses reduce capital — those associated with held-to-maturity securities still qualify. So, tangible equity ratios leave a loophole for banks to game risks associated with securities holdings. Using one catchall ratio doesn't prevent "gaming the system" for certain risks, and it would require the entire banking industry to hold significantly more equity to protect against an easily quantified and directly attributable risk, a highly inefficient solution.
The capital regime in the U.S. sets floors for several different ratios for a reason: Each ratio constrains a particular family of risky behaviors without creating capital burdens for banks that don't engage in those risky behaviors. The common equity tier 1 (CET1) ratio (calculated by dividing CET1 capital — a bank's highest-quality capital — by risk-weighted assets) focuses on credit risk, market risk and operational risk since risk-weighted assets reflect the riskiness of a bank's assets. This ratio constrains banks with high-risk assets (e.g., speculative loans or volatile investments). If their balance sheet is heavily weighted to U.S. Treasuries, then this capital requirement will not be of much interest to the bank.
The revised edition of the Basel III endgame capital rules has been stuck in limbo at the Federal Deposit Insurance Corp. board of directors. The issue there seems to be about the process of reproposal — but the process dictates the substance of the rule.
The leverage ratio (calculated by dividing tier 1 capital by total exposures including on- and off-balance-sheet assets without risk adjustments) primarily addresses leverage risk, focusing on the total size of a bank's exposure rather than risk sensitivity. It constrains banks with low-risk assets but high leverage (e.g., large investment banks with many low-risk securities).
We learned one obvious capital policy lesson in the most recent crisis: Unrealized securities losses aren't equity. The failure to embrace this truth in bank capital policy is a deficiency that has long needed correction. But neither the leverage ratio nor the CET1 ratio addresses what we learned from the failures of SVB and FRB: Only a ratio that incorporates unrealized losses from all securities holdings, regardless of their accounting designation, would have required these banks to restructure their securities holdings early in the rate cycle or hold enough equity to survive the rate hikes.
Instead of raising capital requirements for non-complex banks for ratios like CET1, or, as Hoenig and Miller argue, imposing a high, risk indiscriminate leverage ratio, a better solution is for one of the capital ratios in the capital requirement matrix to be modified to adjust for all unrealized securities losses. Doing so in the regulatory leverage ratio would fit the nature of the other risks constrained by this ratio. Modifying the numerator for this ratio in this way, then setting an appropriate floor would constrain asset leverage and interest rate risk.
Capital policy shouldn't merely chase the elusive objective of finding the "right" level of capital for banks to hold. It should also strive for the arguably easier goal of discouraging risky behavior before it builds. That's not what is happening when regulators reverse unrealized securities losses in regulatory capital measurements. In fact, it encourages undercompensated risk building.
Another problem with the system as it exists is its inability to adapt to emergent threats, and to quickly recalibrate following a crisis. The natural policy response to the GFC was to raise the risk-weighting on the mortgage-related assets that caused the losses. Yet, as clear as this prescription was, it took over a decade to implement fully. This isn't a problem with the risk-based system: It's an administrative problem. Going back to a leverage ratio and getting rid of the risk-based system sounds compelling for its simplicity, but it won't fix the issue. Furthermore, holding banks to a ratio that doesn't discriminate based on risk will only encourage risk taking that maximizes the returns on the indiscriminately required capital.
If we bankers don't start actively demanding proper risk-recognizing and risk-discouraging capital standards, we will repeatably bear the cost of the kind of moral hazard that is always created when regulators fail to address emerging risks in the banking system. It's time to change the regulatory capital treatment for unrealized losses on available-for-sale and held-to-maturity securities. If we don't embrace this capital change, we will once again be subject to indiscriminate increased capital requirements that will disproportionately penalize well-run banks and make less capital available to small businesses in the U.S.
Make no mistake, we have little hope of lower capital requirements. If we don't insist on better capital requirements, we will simply get higher capital requirements.
