Editor’s note: This column is a lightly condensed and edited version of remarks Greg Baer made at a Brookings Institution roundtable on requiring banks to hold more capital in early December.
The basic notion has been for some time that we want to have risk-based capital standards, and we want to risk-weight the assets. Instead of treating them all the same, we will treat them differently based on their probability to default. That’s what the newest Basel proposal is all about. It’s trying to make the system a little more risk-sensitive. The thousand pages of that proposal are really about trying to make a series of choices about different assets and how much risk they hold.
I don’t think there’s any debate — certainly among anybody in the markets, any regulators — that as you increase the cost of capital for a given asset, you provide the bank an incentive to move out of that asset or reprice it up to this point.
You can talk about it in the macro sense of more capital is good. And I think you can debate that, but that’s actually not what this proposal is about. What this proposal is about is choosing at a micro level which types of assets or exposures to advantage over others. And you won’t see banks raising capital or doing an equity offering. They’re going to optimize the balance sheet. That means they’re going to get out of things that have high capital charges relative to the economic value they put on that asset.
That’s why banks are a small minority of the mortgage market now. It’s being pushed into a sector that charges more and is less likely to continue providing that credit under stress. Because banks have and pay for the great advantage of deposits. That makes them durable through the cycle lenders, and that’s why you want them to keep making loans. That’s what’s at stake in this proposal. It’s not some esoteric debate about the optimal capital level.
When they announced the Basel accord, they said they weren’t intending to raise capital requirements. Since 2010, you’ve had any number of economic cycles where the banks have shown no sign that they are significantly undercapitalized, as this proposal presumes. If you live in the real world, it is hard to see anything since 2010 to indicate that they are currently significantly undercapitalized.
The other really odd thing about this proposal is that the most granular and accurate way to assess risk is to use the bank’s internal models that actually look at loan-level detail since 2014. The very largest banks have been doing that. It has been, by all accounts, a success. There’s never been any accusation that they’re misstating the risk. There’s a whole system of compliance and audit and examiner oversight and back-testing to supervise that process.
Equally important, the Basel accord of 2017 continued to allow internal models because they’re more granular and more accurate. But just to make sure that there’s not unwarranted volume variability, it set a parameter around that.
What the U.S. proposal has done without explanation is to say, no, we’re not going do that at all. We’re just going to put everything in 100 percent risk weight. We’re not going to allow use of internal models at all. If this proposal is adopted, we’ll be the only country in the world where your capital requirements for credit are set solely by government models.
What’s even odder is that the regulators actually allow banks to use their internal models for market risk, which is significantly more complicated than credit risk, and for the Fed stress test for credit risk. So it’s hard to escape the notion that perhaps they’re not using internal models simply because they’re producing a lower capital charge than they might prefer for political or other reasons.
If you look at what’s happened over the last 10-15 years, is the problem that there’s not enough capital in capital markets for the dealers? Or is the problem that market depth is way down and in fact, we have gross concerns about liquidity in those markets and we’ve had the Fed having to bail out the Treasury market both in 2019 and in 2020? On the credit side, again, are we seeing banks with massive credit losses that are bringing them down? Are we seeing credit migrate out of banks to other institutions? It seems like, almost without exception across asset classes, banks are amply well-capitalized.
The Fed puts a lot of stock in its Fed stress test. They’ve been running it every year since the crisis. Every year it says the banks have ample capital. Were they wrong all along?
Whether it’s that experience or whether it’s what’s actually going on in the market, I think if you do a reality-based assessment of bank capital, we’ve got plenty.
Greg Baer is CEO of the Bank Policy Institute, a nonpartisan public policy, research and advocacy group, representing the nation’s largest banks.