There are no riskless assets anymore — and maybe there never were

Bloomberg News

WASHINGTON — There's a recurring question that comes up throughout HBO's 2019 miniseries Chernobyl: How can an RBMK reactor explode?

I won't even attempt to explain why that's a real question, much less what the answer to that question is — Jared Harris did it quite well, so I don't have to. But the point is that the presumption of safety — even the seemingly well-founded presumption of safety — can lead to complacency, which is an essential ingredient for catastrophe.

There is an analogous phenomenon in banking, which is the risk weighting of assets to determine the minimum capital a bank needs to hold to be considered safe. The logic of risk weighting is pretty straightforward: Some assets (that is, loans) are riskier than others, and therefore capital should be higher for riskier portfolios and lower for less-risky portfolios. This approach stands in contrast to the supplemental leverage capital ratio, which is based on a much more straightforward calculation of total assets versus total liabilities. More on that in a moment.

Risk weighting has a somewhat checkered past. Prior to the Great Financial Crisis, the Basel Committee denoted sovereign debt as a zero-risk asset — that is, one that requires no capital retention — and mortgage-backed securities were weighted at 1.6%. We all know how low-risk MBS really turned out to be, and we may be reminded of how risky sovereign debt really was in the European crisis that erupted a few years later.

These questions of risk weighting have become increasingly relevant in the wake of the string of bank failures we recently experienced because the "toxic" assets that were weighing down the balance sheets of both the failed and not-yet-failed banks were Treasury securities — the safest assets of all. And in liquidity terms, those Treasuries worked just as they were supposed to — there is and remains a deep and liquid market for Treasuries. But there is a limited appetite for assets bearing 2% or 3% in a 5% interest rate world, so while these assets are not valueless, their value has declined enough to cause at least some banks to get distressed. 

The logical answer would be to empower the supplemental leverage ratio as the binding constraint for banks — keep it simple, smarty-pants. JPMorgan Chase CEO Jamie Dimon riffed on that idea during his investor day call with analysts yesterday, saying that "constant capital confusion is a bad idea," as embodied by the stress capital buffer, which determines a bank's minimum capital requirement based on its performance in the prior year's stress test. 

Warren urges FDIC to reject plan to pay assessment fee with Treasuries

But there are problems with the leverage ratio as well — problems that the stress capital buffer was designed to solve. Some banks, specifically custody banks, have large volumes of liabilities and invest in the safest possible assets, and so the leverage ratio would unfairly hurt their bottom line. Former Federal Reserve Gov. Daniel Tarullo found that argument sufficiently persuasive to present in his farewell address in 2017 that "a leverage ratio serving as the sole or dominant form of prudential regulation would probably have to be set considerably higher, at a level where the impact on financial intermediation could be quite substantial." 

Risk-weighted and leverage-based capital requirements are designed to work hand-in-hand, and in the best of times they do. But perhaps one innovation that can come out of the current crisis is doing away with the concept of "riskless" assets entirely — certainly some assets are riskier than others, but nothing — and I mean nothing — is entirely risk-free. 

Part of the reason for raising the floor for risk-weighted capital is that banks tend to load up on assets that tie up their capital the least — MBS and sovereign debt in years gone by, long-dated Treasury securities today. That concentrates bank investments in low-yielding but "safe" assets rather than spreading those investments around — and pretty soon everyone is doing the same thing. In other words, if you designate almost anything as "riskless," before too long market concentration will create risk where there was no risk before. It's a kind of Murphy's Law for risk weighting — when you depend on a single thing to bear no risk, eventually it will. And it doesn't take a nuclear physicist to tell you that.