Anat Admati on the US Banking System and the Basel III Endgame

Anat Admati is a professor of finance and economics at Stanford University and is the coauthor of the 2013 book, *The Bankers’ New Clothes: What’s Wrong With Banking and What to Do About It.* Anat is also a returning guest to Macro Musings and she rejoins the podcast to talk about the 2024 expanded edition of the same book, as well as the most recent developments in banking. David and Anat also discuss the effectiveness of post-financial crisis regulations, the design and impact of Basel III Endgame, the fallout from the most recent regional banking crisis, and a lot more.

Read the full episode transcript:

Note: While transcripts are lightly edited, they are not rigorously proofed for accuracy. If you notice an error, please reach out to [email protected].

David Beckworth: Anat, welcome to the program.

Anat Admati: Thank you.

Beckworth: Well, it's great to have you back on the show. Now, I was comparing the new edition of the book to the old one. The old edition had 416 pages. This one has 624. So, you added a good amount of material to it, still an important book, but with more meat and more ideas to chew through. What motivated the additions?

*The Bankers’ New Clothes*: New Additions

Admati: In summer 2022, we were coming upon the decade anniversary of the book, and this was a time at which it was post-COVID, 10 years had elapsed. It was the start of interest rate increases, there was crypto, there were all of these things. And so, we thought, "Okay, well, let's republish it, because the book's still relevant." I basically continued in that fight until about 2015, '16, maybe around the time we first talked, and then I left for a while. There was nothing to do and nobody was talking about it anymore, and now we were concerned that they haven't fixed this system. They didn't listen to us.

Admati: We were going to refresh the content and sound some warnings about this persistently distorted and dangerous system and explain some of what's happening. And we weren't sure we wanted to explain, maybe, more about central banks because, central banks have become ever more important, certainly since the financial crisis, and then COVID, and now they were having to deal with some of the interest rate issues and inflation. We were considering talking about crypto. In the end, we took it on in fall of 2022, and crypto kind of took care of itself during that fall. 

Admati: And then we were still going to explain why the system is fragile and then why too-big-to-fail is still a problem, and then warn specifically about both fragilities that were going to become evident when the interest rates increased, and the too-big-to-fail, specifically Credit Suisse, which was on our radar as a, probably, zombie bank by then, for sure, and maybe crypto. And, of course, we were going to finish it in December, but then we didn't, and we were going to finish it in March, but we know what happened in March. So, in March, the book was just barely still in our hands, and we had to completely go back. And so, that's how it grew so big, because, at that point, we had to include more material on central banks. We had to include the discussion of all of the events that were going on in spring 2023.

Admati: In the end, we didn't do crypto, but we did a lot more central banks and explained, because what happened in 2023 is that the central banks, both in the UK and in the US, sprang once again to save the system, to save the government, to save the banks and others. And we had to explain how they work, so we did that. We explained the balance sheets of central banks, how they work, not monetary policy, but the interaction of central banks with markets and institutions in the private sector.

Beckworth: Well, I guess it was fortuitous that you hadn't completed the book until March. Things unfolded right as you're finishing it up and you're like, "Hey, another chapter to add to the book," so, great timing, if accidental timing. I saw you soon after that, last year at the Hoover Monetary Policy Conference. It was in May, so it was just a few months after the crisis. You gave a talk, I think one of your colleagues gave a talk. Darrell Duffie gave a talk. And that was a really great conference, also well-timed, because we had just gone through the banking turmoil that spring.

Admati: Well, I was moderating that panel, but I did, as a moderator, insert my-- 

Beckworth: You certainly did, yes.

Admati: -And challenged some of the people on the panel, that included my colleague and Quarles on the issues of both the bailouts and the TLACs, the loss-absorbing debt, that they told us we’d do bail-ins instead of bailouts and all of that. 

Beckworth: Alright, well, let's go to your book. Again, you have the new expanded edition, but I want to go back to the original one, which is in the new one as well. This covered the very basic arguments you made the first time, because, I think, you will make the case that they will still apply today. So, I'm going to run down a quick summary, and then you can correct me or tell me if it's right, but here's my summary of your book. And, again, this is not going to do it justice. Go read the book listeners, but here's how I summarize it. First, banks were operating with too much leverage. That is, they were using excessive borrowing or debt to fund themselves. Instead, they should have relied more on equity or capital to fund.

Beckworth: Second, as a result of this banking system having too much leverage, they were more susceptible to crisis, and this, in turn, led to a perverse system where there was growing reliance on government bailouts. Certain banks in particular benefited from these ballots, which helped foster a too-big-to-fail system of big, big banks. And another point you bring out, I'd say fifth, is it created a “heads, I win; tails, you lose” environment for these big banks. So, [there were] privatized gains to private investors while the losses were socialized. And finally, the sixth point is that we need to fix this. So, you wanted to see far more funding through capital. Is that a fair summary of your book?

Summarizing *The Bankers’ New Clothes*

Admati: Yes, we were explaining how this works. We were explaining the incentives of the banks and why regulation, in this issue, is essential. Since writing the book, I gained a lot deeper [of an] understanding of the dynamics of leverage, of corporate leverage more generally, and how addictive leverage can become, which I think we discussed the second time I came on. And so, banking is sort of fundamentally inefficient. It's kind of born inefficient, and the markets are just not going to control that, because the depositors are very fragmented and don't have the right contracts or ability to monitor what is done with their money.

Admati: The regulators are basically like the debt covenants of the depositors. From that point on, the banks really are unable to commit to being more careful with depositors' money, and they really need the disciplines of a regulator, to correct the distortion, to correct their incentives, their addiction to borrowing. If it was up to them, they would operate with zero equity and just take the money out every time they had it, and then let the system, the Fed, the FDIC cover up the downside. And we explain some of the way that the guarantees and subsidies, the safety net, operate on the bank, and then explain what's wrong with the way the regulations were framed.

Admati: In Basel, before the crisis, where they failed miserably and actually exacerbated the fragilities of the system, the way they do the risk weights, all of this stuff-- You can also talk to Tom Hoenig about the system, the complex regulations, and why they add distortions. In their complexity, which is later weaponized and used to economize on equity ever further by the banks, through the complex modeling and other things in the congregation, the system also encourages banks to lend to governments over businesses, so it's actually an anti-lending system, the way it's framed.

Admati: The risk weights only take care of credit risk, never take care of interest rate risk, so a lot of it is really, really flawed, which is what we saw in the last year. We gave, in the book, specific recommendations, both what to aim for in the regulation and steps [on] how to get there. It's all like a guide taking people on to that, and then also alerting that both of you can have too-many-to-fail, you can have too-big-to-fail, the system is bloated, the system is inefficient, et cetera. And we ended with a chapter called “Other People's Money,” which was a broad governance structure about both people in the banks and people in the government being careless with other people's money, with taxpayer money, with investor money, et cetera.

Admati: In the new part, we update the fact that the system has not changed much, and to some extent got worse. Some of the biggest banks are bigger. The risk weight system was barely a tweak to the ones that brought us 2007 to 2009 and a lot of nonsense. So, *Bankers’s New Clothes,* the title of the book, refers to flawed claims that are made, of which there are dozens by now. In the recent 2024 documents, we now have organized, by type, a list of 44 distinct flawed claims that we debunk.

Admati: And the lobbies have such a big grab bag of flawed claims that they will just take you into the night, and people who want to hear them might be convinced by them. But it starts with the absolute basics about what it is [we’re even] talking about, to the point that you can get senators or journalists explaining that we're talking about cash reserves sitting on the sidelines. And the cash reserves sitting on the sidelines are actually money that they put in the Fed, central bank, earning a high interest, and often borrowing from the Fed itself in lower interest, in a perversion that just happened recently. That's money setting aside, when they take out dividends, that's money not lent.

Admati: That's perfectly good equity money that they take out because of their addiction to borrowing, so, a lot of confusion, a lot of politics, and all of that. So, in the new addition, we expand a lot more on bailouts, both the mechanics of bailouts and the politics of bailouts. So, we have two full chapters on why it is that bailouts persist, and I can explain how they play out, even today, as we speak. And then we expand ever more on the rule of law, where the final chapter is called “Above the Law?” which actually goes through the entire system of ineffective rules, both internal and external rules, and there is weak enforcement internally and externally, and nobody going to jail, and the whole thing about liability regimes, and why they get away with all of it.

Beckworth: Yes, so, listeners, get your new copy of the book and, Anat, I'll share with you that I spent my last year teaching at a university in 2016. And what I did— I believe it was the second to last semester I was there. I was able to teach an upper division undergraduate class, a seminar-type class. We focused on financial crises and, basically, it's a class that I designed for books that I wanted to read, and your book was on that list, so we had a good time, my students and I, reading through that book. And that's where I first got to know you and your co-author, and then it was a real delight to actually get you on the podcast and meet you in person. So, if I did that seminar again, we'd have to take the expanded version of it.

Admati: Yes. I teach undergrads and MBAs from this, so I just finished teaching a short MBA course called “Power in Finance” that was using it. We had Kevin Warsh, and then we then had an event at Stanford launching the book, also with Kevin Walsh--

Beckworth: Very nice.

Admati: -and Amit Seru, my two colleagues and my one colleague, and Amit was moderating a discussion between me and Kevin Walsh about power and politics in banking. And he visited my class. I teach the undergraduates as well in, “Finance, Corporations, and Society”, and they gain some understanding of the issues, because it's not hard. It's just very confusing jargon, very flawed claims, and denial of basic corporate finance, which is where I come from, into this rabbit hole of banking, and it's possible to teach it. It's just that there's so much willful blindness and confusion that it's really hard to get through with the basic truth.

Beckworth: Now, for those listeners who do not know the definition of equity or capital, and I know most of them do, but for those who may be joining for the first time or are just recent additions to Macro Musings, maybe give us an explanation of what equity or capital funding is versus debt funding.

Equity/Capital Funding vs. Debt Funding

Admati: So, all companies have to fund-- A basic balance sheet of a company has, on one side or on top, what they own, assets. On the other side, the liabilities and equity side, they have, essentially, how they fund, what they owe, and who has financial claims against their assets. That's their creditors, the people they owe money to, and their residual being shareholders or equity holders. Every company has owners of its equity. And the way we started in the book, because that's really the entryway for a lay audience, is with a mortgage in a house, and the down payment, which is the initial equity of the homeowner.

Admati: So, basically, shareholder equity for a corporation is the equivalent of the equity somebody has in their house, which is what they would have if they sold and paid off the mortgage, the leftover. Now there are accounting issues about who puts numbers on balance sheets. That's a whole other rabbit hole about who puts the numbers and whether we're talking about assets minus liabilities, which is an accounting definition, or they were talking about market value of assets, fair value of assets versus face value of liabilities, which is really your distance to default.

Admati: There are all kinds of issues related to how you measure leverage, but let's just say the following: An underwater— this is an important concept— an underwater homeowner, which a lot of homeowners found themselves being, meaning somebody who owes more on their mortgage than the house is worth, meaning that they could not pay the mortgage if they sold the house, that is the equivalent of an insolvent corporation; a corporation that cannot use its assets to pay its debt. Now, insolvency is a tricky concept, because it has to do with the valuation of the assets today or when the debt is due. Can you pay your debt? It has to do with your ability to pay.

Admati: Now, what's happening in banking is that for most corporations, they cannot operate with very little equity. Investors just won't go for it. In banking, what happens is, you can be insolvent for a long time and hide that insolvency. Banks are zombies. They're walking dead, almost like it's a Monday morning, like it's a state of affair in which they have an enormous debt overhang to the point that they are showing all of the symptoms of insolvency all of the time. An insolvent borrower, like a corporation on the verge of bankruptcy, would have very flawed [or] distorted incentives, because they're conflicted.

Admati: If they're still in control of the assets, they're conflicted with the creditors who are actually the effective owners, or would be the effective owners, once the corporation is in bankruptcy. In banking, the bankruptcy doesn't happen, because they don't default and because they have a lot of supports from the outside, from third parties. The depositors are not running unless they're uninsured and realize that the bank is insolvent, as happened in SVB Bank, Silicon Valley Bank, in March. In that case, the bank exposed its own insolvency, because even as it was allowed to maintain the valuation of Treasury, government securities at full value that they paid for them, the value of those securities went down when interest rates increased.

Admati: When they were forced to sell, they were forced to actually recognize a $1.8 billion loss. That exposed their weakness. Then, in order to survive, they had to raise, or they tried to raise, equity and couldn't. Now, a corporation that cannot raise equity is definitely insolvent. A corporation that can raise equity is not necessarily solvent, because the equity can always have some upside chance, but if they cannot raise equity, if investors will not give them a penny for their equity claim, that would be committed to paying the debt and take the rest, is definitely underwater and insolvent.

Admati: When everybody realized that it was obvious at that point, that's when the run started. It wasn't a pure “panic and run” like some people wanted to present it, a social media version of a bank run from the movies, or whatever. That's not what happened in SVB, that's not what happened in Credit Suisse or Signature or First Republic. These banks became insolvent because of bad decisions, because of a lot of risk, and because of too little equity to absorb the losses. What happens with equity when you have a house and you bought it, say, with a lot of equity, [is that] if the house goes down in value, you won't be underwater. The equity will absorb it.

Admati: Now, if you become insolvent, you can inform a bank, depending on your contract, [and] either there'll be a no recourse loan, and you can just walk away from it and leave it to the creditor in a foreclosure, or, in some countries, they have recourse to your assets, so it's not limited liability from the house, but corporations are limited liability. So, basically, the debt can only be paid from the assets of the corporation, and if they don't have enough assets, then something else will have to happen. Either the creditors would lose or there will be a bailout. A third party would come in and prevent a default. That's all there is to it. In other words, if the corporation loses money, either the shareholders lose or somebody else loses, and that's all there's to it.

Beckworth: So, we want to have more equity, more of a cushion to absorb shocks. So, I can speak from firsthand experience. My first home was bought in 2007, and I didn't have a lot of equity, so my cushion wasn't very big. When I had to sell and move, I was hit hard. I had to actually pay money to sell my house, which was not a fun ordeal to go through.

Admati: Yes, that's what happened. On the upside, per dollar, you'll do great, because the leverage magnifies the risk on the upside, but it also magnifies the risk on the downside.

Beckworth: Right, so, I learned that lesson the hard way, but the key insight, I guess, is that we want to have banks to have enough of a cushion, an equity cushion so that they can absorb losses or whatever, shocks--

Admati: -And keep lending in a downturn and all of that.

Beckworth: Yes, so whether it's something either idiosyncratic or something macro, you need to be able to absorb those shocks. Here's my question to you, then, just kind of stepping back from this kind of detailed discussion. What is a number that you would like to see equity be at? And maybe it's complicated, because [there are] different banks, different ratios, but in general, where would you like to see bank capital levels be, and where are they now?

Building Higher Levels of Bank Capital

Admati: Okay, so right now-- We in the book and in letters signed by top academics in corporate finance and banking, the number of digits that is in the total equity relative to total assets, we aim at maintaining it between 20% and 30%. These numbers sound crazy, because the current numbers are in single digits. They quote you double-digit numbers, but those are not relative to total assets. Those are relative to something called risk-weighted assets, and a lot of assets get ignored. In particular, government bonds are ignored as if they don't have any risk even though they carry the risk of interest rate changes, even if they don't have default risk.

Admati: When First Republic lent mortgages to rich people, there wasn't a lot of credit risk, but they lent at 2%, and that's what they were going to get, but they then had funding costs that were increasing into the 5% or more, and that's a formula similar to what happened in the savings and loan [crisis] when interest rates were high for funding and low for the 30-year mortgages. So, the point is that right now, the capital ratios are all very complicated. The leverage ratios, which is what I am talking about, are in the 5% [range]. Academics were saying 15 at least. You won't even begin to see the cost. You only see benefits.

Admati: If all we're doing is removing some subsidies, that could be a good thing, because subsidies are paid by somebody, and the industry could be very bloated with the subsidies. So, if the industry shrinks, it's possible that the industry is too big and too bloated and too comfortable in its own size, and that we're not getting what we need at the appropriate price from this sector altogether. So, something is very wrong with the ability to live so dangerously and benefit from outside subsidies, and then threatening that you won't make some loan if we take any of these subsidies away, which is what's happening now in the policy debate, still, today, and was always the problem.

Beckworth: So, right now, we're around 5% in terms of the leverage ratio, a very basic measure of equity to debt, and you would like to see it somewhere between 20 and 30. Is that fair?

Admati: They would take you gradually to that. 

Beckworth: You wouldn't go overnight and shock the system.

Admati: [inaudible] earnings and issuance.

Beckworth: Fair enough. 

Admati: In other words, I'm not giving you a ratio, because adjustments to the ratio can be inefficient themselves.

Beckworth: Sure.

Admati: I know exactly what to do, but they're not doing it.

Beckworth: So, let me ask this question. I believe that one of your colleagues, John Cochrane, has called for even more radical proposals. I believe that he would be happy to go— and correct me if I'm wrong— to 100% equity-funded banks. What's wrong with doing that? I guess that's my question. If the argument is that you have a cushion, why not go to 100%?

Admati: First of all, part of the business of banking is taking deposits, so immediately they have debt. They're born with debt already if you're going to have them have deposits. There are narrow banking proposals-- We have a document now called, *The Parade of Bankers' New Clothes Continues,* and we do have a section there about more extreme solutions. When John Cochrane reviewed the book in the Wall Street Journal, he ended by saying, “How much equity until it doesn't matter?” And this is sort of what I think. I think the reason they scream so much is because they have so little equity, so now it [really] hurts.

Admati: If you're addicted to something, it hurts when I start taking it away, but once they had 20% to 30%, then the risk will primarily be theirs to bear, and then things will begin to calm down on that. So, 100% reserves or 100% equity is more extreme than we want. We think it's enough to maintain it at a level if you are watching it, if you are effectively supervising it way away from insolvency. So, you can intervene in time, everything is slower, [and it] isn't like every little bit wipes them out, the equity. So, you have time to adjust, you have time to build back up, you have time to sell assets that might not be essential. You just have time. The way they live on the edge is just incredibly unhealthy for any corporation.

Beckworth: Well, let me throw out the reasons why I wouldn't want to go to 100% equity-funded banking. I've got three at top of my head. One is, I don't want to have a bank account where each hour I check-in, it swings in value. 

Admati: No, I agree with that. So, if you're talking about a mutual fund, no, I think deposits are useful. I want to maintain the ability to make deposits, I just want to reduce the footprint--

Beckworth: -Or you want the cushion, right. I agree with you on that. That'd be my first concern. The other concern is, if we went so high-- again, I'm going to the extreme here just to illustrate a point. What would happen, I think, would be that you would see more deposits, money creation kind of shipped out into the shadows. It would be that you would push the risk at that high of a level. My third concern would be, what would happen to the money supply, again, at this extreme 100% equity-funded bank level?

Admati: So, that shadow banking thing is this bugbear that they bring up. A lot of activity happens in shadow banking. I'm sure that the colleague that was in that conference you mentioned, and in our event as well, shows that shadow banks that make loans and are not funded with deposits have a lot more equity. It's really just that the banks have so much access to subsidies and guarantees that they just want to have a little equity and get away with it. We're talking about corporations everywhere, even in the financial system, that are living more in markets than with this coddling.

Beckworth: Okay, so you're saying that shadow banking— shadow money creation, actually— they fund with more equity, typically?

Admati: They have more equity. They have twice more equity than the banks, and that's despite the tax subsidy that we give [to the] debt of all corporations, but we don't regulate the leverage of corporations more generally, and yet we never see corporations as indebted as banks. So, there's something wrong with it. Investors push back against it. It is just that the banks get away with it, because their creditors are abnormal, [and] it falls on the regulators who are also lame. So, the regulators are not doing their jobs. The supervisors [and] regulators [are] continuously failing. Auditors are not doing their jobs. So, they get away with this and, all of a sudden, it gets exposed every so often.

Beckworth: Okay, so, one of the objections you have received many times is that funding with equity is more expensive. And you really go at length to explain why that's wrong. So, what is the argument for why it's not true?

Is it More Expensive to Fund with Equity?

Admati: When you say expensive, expensive to whom is the question. The thing is, it's very expensive for the bankers, but it's really cheap for the rest of us. There's a fundamental conflict of interest between people who find it expensive. For example, let me just say it in this way. It's expensive for me if I can't reach into your pocket and steal your wallet. We still have laws against my getting into your pocket and stealing your money. Basically, what's happening is, it's expensive for them to lose subsidies and guarantees. What's being done is to reduce their ability to ship the cost and risk to others. That's it, plain and simple.

Admati: There are laws of conservation of taxes and risks. If they don't bear it, somebody else bears it, and that's all there's to it. That's it. So, when they scream, they're only saying, “give me subsidies and trust me to use them for you.” Well, they'll do what they'll do once you give them the subsidies. They'll, first of all, take off the top for their own bonuses, but there's no social cost at all between 5% equity and 25% equity. None of the costs are social, all are private costs.

Beckworth: Are you invoking kind of a Modigliani-Miller theorem idea here?

Admati: No, so I'll explain that. Modigliani-Miller is only a result that says— So, right now, as we speak, there is an explainer of why equity is expensive on the Bank Policy Institute's website. Every student taking every corporate finance course in the land is going to fail, saying what they're saying. This is a basic policy. The policy is the following, and we explain it in the book right away from the beginning. The more equity you have, the lower is your return on equity.

Admati: On the upside, the higher is your return on equity. On the downside, the risk is less per dollar; less risk, less required return. So, again, there is a conservation. Now, if we give a tax subsidy to more debt, that we’re giving it to all corporations, and that's a distortion by itself. I’m against that policy. Now, we can settle the amount of taxes, but in any case, if they pay more taxes, somebody else pays less, and vice versa. If we want to subsidize first-time homeowners or some businesses, we can do that.

Admati: Passing subsidies to the economy through the banks is extraordinarily inefficient. We used the banks to pass on PPP loans, and there was a huge windfall for them. We used the banks for all kinds of supports, because they clip their coupon on the way to anything. So, the fact that they stand in the middle and take off the top of every subsidy that we might want to pass is already bad, and we should find other ways to subsidize the economy if we need to subsidize anything at all if the markets don't work, not always through the banks or through corporations. And that's also true for the airlines [during] COVID and whatever.

Admati: In other words, we had a petition of 300, I don't know, 250 academics during COVID [that were] saying, "Find ways to give support [during] COVID to people and not to corporations." In Europe, they took over the payroll. Here, they didn't take over the payroll. They told the corporations not to fire people, but they did fire people. We couldn't get our luggage from the airlines when airlines came back because they did fire people, but the executives were fine.

Beckworth: Okay, well, let's move on to the present. So, those are some of the issues covered in the book, the old version and the new version, so we'll, again, encourage listeners to check it out. And I want to start with this. You've touched on it already, but just more explicitly, how well did the post-financial crisis regulations work? 

Evaluating the Post-Financial Crisis Regulations

Admati: Poorly, very poorly. What they claim are tough reforms are not at all tough reforms. I sometimes quote Martin Wolf saying, "You triple nothing, you get nothing." They always talk about percentages relative to before. Even now, they say, "Oh, 20% more." Well, 20% of nothing is not a lot. So, instead of saying, "Oh, the speed limit was 98, and now it's 96,” that doesn't sound like a big difference. If you say, "Oh, my equity was 4% to risk-weighted assets, or 2% to risk-weighted assets, and now it's 7% to risk-weighted [assets],” they say, “Oh my God.” And that's still nothing, because that's even to risk-weighted assets. Anyway, tripling nothing is nothing, so they didn't change the approach.

Admati: They tweaked the levels, but the measurements are bad. This is the biggest problem. The biggest problem is that the measures are very dated, and according to these measures, all of these well-capitalized banks failed the next day. The banks that failed in the spring, like SVB and First Republic, they got glowing regulatory ratings right before they failed. They got auditors signing off. They have these ratings called CAMELS. At least, I know from First Republic that they got hugely spectacular CAMELS just before they failed. How can you have a well-capitalized bank fail? What does your measure of capitalization mean if you rank a bank well-capitalized and then it failed?

Admati: In the financial crisis, the banks that needed the most bailouts were well-capitalized through the crisis, so the measures mean nothing. If that's the rate you are measuring, you are missing the crisis. In the market, you're going to see the stock price go down, but on these measures, they're useless. So, the point being, we have, in the law, prompt corrective action statutes and authorities to regulators from the beginning of time. Before Dodd-Frank, before any of this, they were always there, and they just keep failing to use them. They just don't dare the banks. It's all basically a spectacular regulatory failure, and capture, and intimidation of the supervisors by the banks.

Admati: The Fed failed. The FDIC failed. These people just failed to do their supervision job, and then it's the big “oops,” and we have got to give them liquidity support and all of that. They're basically bailing out insolvent institutions. When they announced that they were going to recognize the collateral of the banks at par, with Treasury putting $25 billion behind it on March 12th, that's when we knew that we had to explain how central banks work, because central banks are meant to help financial stability. Instead, they became enablers of financial fragility.

Beckworth: Let me go back to SVB, Silvergate, and that banking turmoil since you touched on it. As you noted earlier, you could point a finger at the fact that they were just poorly run institutions. They were very dependent upon specific sectors, Silicon Valley or crypto. I think everyone agrees with that. Now, I guess, going back to the role of the supervisors, though, the bank regulators, do you fault them, or do you fault the fact that they were forced to follow these poor metrics, as you outline?

Admati: All of the above.

Beckworth: All of the above, okay.

Admati: They had the authority to do more, and they failed to do it, very plain and simple. Now, they're not the first or the last to fail in the supervision of banks. It's, really, pretty outrageous how they failed. What are they good for if they're going to let the bank with obvious flags continue to pay dividends and then continue to persist until, all of a sudden, it becomes obvious that it's insolvent? It's been insolvent for months. It had been insolvent for months when it failed, SVB.

Beckworth: So, you think that the supervisors should have seen it?

Admati: And not only the Fed, by the way, the safety nets go to the federal home loan banks and others. Everybody's keeping [them] alive, these zombies.

Beckworth: So, let me go, then, to the bank term funding program that you touched on where, at the time, it seemed very shocking that the Fed would accept the Treasuries at par value. I've heard this pushback against that concern, that that concern may not be as worrisome as some had said, because when you take your Treasury to this facility and you use it there, and you borrow against it, you're still borrowing at high market rates. If you were to, for example, in theory, roll it over, you're losing money, because you're getting that market discount bearing that drove the low price in the first place, which caused the problems. No?

Admati: No way. The losses are meaningful. You may claim that you are going to hold it to maturity, but you may not be able to hold it to maturity, which is just what happened to SVB, because you are basically bleeding. You are paying the high interest, and your assets are not worth as much. Eventually, it catches up with you, which is what happened in the savings and loan. You had interest rates very high that you had to pay in markets. You had mortgages paying lower interest. It's a formula for failure. You maintain these institutions-- you know, in the savings and loan, they loot. The fact that they pay dividends is pure looting.

Admati: In a bankruptcy court, it would be called fraudulent conveyance. You cannot take money out of an insolvent institution and pay dividends and bonuses to yourself, but that's what they're doing. This makes no sense. There’s no way that the Fed or any central bank is supposed to lend to an insolvent institution. This is part of the lender-of-last-resort mandate. The fact that they're getting backing from the Treasury suggests that it's a bailout.

Beckworth: That's fair.

Admati: It's a clear bailout, and with the bailing out of the uninsured deposits, that was also a bailout. It doesn't matter if the industry pays, it's still a bailout. If you and I make an agreement, and somebody, a third party, comes to make sure you don't default, you are being bailed out, and I'm being bailed out by a third party.

Beckworth: Okay, I see where you stand on that. Well, let's move on to some of the fallout from it, I guess, or the suggested changes in dealing with banks. Let's first touch on those mid-range banks, which these were, in the US at least, SVB, Silvergate, First Republic. They're kind of between small and big banks, and so they had a carve-out. Was that consequential? Will that change much?

Fallout From the Regional Banking Crisis

Admati: Not a lot. I think that the stress test didn't even envision the increase in the interest rate that was already happening. The change was bad, and we wrote letters against it, that's true, but I don't think that was meaningful. I think that they had authorities anyway, they failed to use the authority. It's true that if it was included in something, maybe somebody would have noticed more, but I'm not sure of that at all. I don't think it was that meaningful. I think it was a bad move, but I think that the failure goes deeper than that.

Beckworth: So, it wouldn't have made much difference if they had a little bit more of an equity cushion, because it would have been--

Admati: No, I think it wasn't about equity at all. It was just about supervision and stress tests. The equity levels were not that much higher above 100 versus 250. I think it was more about the stress tests and other things that they would have to have, and now they're proposing long-term debt, which is absolutely crazy given Credit Suisse. It didn't work at all [with] Credit Suisse. None of these debt-absorbed losses, they were all stuffed into UBS except for these 81 that had the ability to wipe them out ahead of equity. It was crazy, absolutely crazy.

Beckworth: So, another development which I find intriguing— and I like this, and I'll explain why in a minute— is this idea of encouraging banks to more vigorously use the discount window, so park collateral there so that you are, one, more familiar, there's no stigma, you can draw upon it. Then, secondly, you can use that drawing capacity from the discount window to count towards your liquidity requirements. You're shaking your head, so I'm taking that as a hard pass from you.

Admati: I'm shaking my head because I'm not a fan of liquidity requirements, number one, at all. We created central banks to solve pure liquidity problems. If a bank is insolvent, it doesn't have a run, because it has a central bank. I view liquidity requirements, the liquidity coverage ratio, and all of that, as costly on good days and useless in the run. The cost-benefit is just not there. Capital requirements have every cost-benefit working in the right direction. In other words, lots of benefits, no costs, [or] virtually no costs, because you're not touching their investment, no stigma.

Admati: I think that a bank that needs support should have stigma, they should manage their liquidity better and their capital better, and the fact that I want to throw money at you, and “please, please, come to the window--” I mean, they were accepting collateral at face value, even for the discount window. I checked into that, and that was shocking, so, too much coddling, too much [of a] safety net, too much lending.

Beckworth: So, I love what you said in terms of describing these liquidity requirements, that they're costly in the best of times, and they're useless in runs. That's a pretty strong view, that they don't seem to matter as you see them.

Admati: I'll tell you why. They're saying, "Have enough liquid assets to satisfy a 30-day demand." Now, first of all, you have to define what's liquid. Is Treasury liquid? Cash is liquid, that we agree [on]. Cash is the ultimate liquidity. So, you can have cash reserves, but now you're parking money. Now, you're parking money. That money is on the sidelines. Now, how much is enough for 30 days? Well, in a run, that's out the window, and what's liquid one day, if you expand from cash, is frozen the next day. So, in other words, in a bad situation, it won't help you, and on a good day, it's very costly to comply and enforce the rule. So, I don't see why we need liquidity requirements when the cost-benefit is not clear. But I do see why we need the equity requirements, big time.

Beckworth: Well, let me give a little push for this then. I probably won't win you over, given what you've said, but something that I think a lot about, and that I wish we would see more of, is a shrinking of the Fed's balance sheet, the size of it, and the Fed's footprint getting smaller over time. And in some ideal world, we return to a small corridor system as opposed to the floor system, which I know is not going to happen any time soon, but this would be a step in that direction. If banks knew that they could draw from the discount window, they would have less of a structural demand for reserves on their balance sheet.

Admati: That's fine with me. That's totally fine with me, that they would go to the window to solve pure liquidity problems. So, it's under budget, they are solvent, they are paying a penalty rate with good collateral, no problem with that. No stigma with that. It's fine. And I agree with you. The balance sheet of the central bank grew unbelievably, and we expose the distortion that that creates. So, I completely agree with you that the central bank should go back to a more manageable size and stop buying up the whole place.

Beckworth: Okay, well, let's move to Basel III Endgame. We touched on Basel earlier, and I want to read a quote from my colleague, Tom Hoenig, that we talked about. I interviewed him a while back about Basel III Endgame, and I asked him what he thought, and he goes-- this is an excerpt from a longer conversation, but he says, "I have absolutely no use for the risk-weighted capital program at all. It is misleading. Let me give you an example.”

Beckworth: “You quoted that the G-SIBs, the largest banks, have 12% capital now. They may have to go to 14%. Of what? 14% of what? Risk-weighted assets, not total assets. It's much smaller than the 12 to 14 would have you believe. That's number one. On average what is their basic equity, tangible equity to total assets? Not supplemental leveraging, basic equity to total assets. It's 7%. It's not 14, it's 7%. They manage their balance sheet. I said yesterday, [at a different place], that the Basel capital rules are not rules. They are a game to be played, and the largest banks are master chess players at [it].” Thoughts?

Evaluating Basel III Endgame and its Impact

Admati: It's what I said before. I told you that it's about the measurement. I told you that it is about the risk weights. We are thinking the same. My levels that I told you were to total assets, leverage ratios, but not at the levels that we have them right now, [but] at levels multiple times that. There's nothing crazy about this. This is considered crazy. There's nothing crazy about it. Nobody in 15 years gave me an answer of, why not? Not one answer that was satisfactory.

Beckworth: So, what exactly is Basel III intending to do?

Admati: Here's a few of the things. All they do is they don't touch the top-level numbers, the capital ratios. They just tweak the risk weight. That's all they do. And so, by tweaking the risk weights a bit up, they are increasing. Tom Hoenig did not submit any comment letter on Basel III Endgame. I submitted two. One was supportive, with 29 other academics, but I had to submit my own, because I'm critical of the whole thing, and I was saying that I hope it's not the endgame, because, otherwise, we're going to have a big financial crisis, and only then will we talk about it, because I objected to the risk weights and I objected to the levels.

Admati: What they're doing is-- there's an issue about whether the banks can use their own models, and they have manipulated the models. That's the chess game that Tom Hoenig was talking about. They are using the models to economize on equity, and they are playing that game. They increased some requirements against operational risk. What operational risk? The banks are paying-- the last chapter of my book is called “Above the Law,” and it's about all of these billions and billions and billions of fines that they are paying, hundreds of billions of dollars of fines, and they still are alive, the cost of doing business.

Admati: Now, that's operational risk. You're money laundering, you helped Epstein and Madoff, and you name it. The rap sheets are enormous. All of the fines that they pay all of the time and the damages… the list is endless. It's like a daily thing. And so, that has to come out of somewhere. You had the whole fiasco where Eric Holder, back in 2013, gave his too-big-to-jail speech, in which he was concerned with giving them too many fines, because it'll be collateral harm, and then they're proud of all of the fines.

Admati: But the banks can always fund more and more and can extract more and more, so they swallow the fines. Nobody's going to give them fines that would kill them, because God forbid. So, when I was asked about some claim that they make about too much equity for operational risk, I said that the overall levels are so ridiculous anyway, so any increase is good for me. I was supporting it as a step in the right direction, only because it increases and it removes some gaming possibilities, a little bit, and that's all. It's very small. They make a huge deal out of it. I don't know why. Their explanations are full of these flawed claims that we take one by one, and it's a depressing scene of nonsense in politics. That's what it is.

Beckworth: Is it likely to come to fruition, in its current form, in the US?

Admati: The Fed Board is usually unanimous on interest rate decisions, but it can be divided on these kinds of things. So, it depends if vice chair for supervision Barr keeps or modifies the proposal, and whether he can get four people to vote for it. I don't quite know where it stands. There were people who were on the Board who were skeptical but supportive. It wasn't clear where it stood, and then it opened for comments, and that was a whole scene. That's why I submitted my two comments. At the same time, they also made a proposal for long-term debt, which is really, really, really ridiculous.

Beckworth: Explain that proposal.

Admati: So, that proposal, basically… you can fund with equity or you can fund with long-term debt. Now, the long-term debt, that's not even deposits. That's just a way to fund. There is absolutely zero and no justification for long-term debt relative to equity, absolutely nothing. So, if you want absorption of losses, then you have equity. If you have long-term debt, the only way it will absorb losses is in a resolution process, or in some kind of bankruptcy-like process after failure.

Admati: Well, they're not proposing to do this for the $100 billion or more banks, they're already proposed to do this, but those proposals for 15 years, many of them said, are entirely dominated by proposals for more equity. All of this is on the funding side. There's no reason that Basel Endgame is even separated from long-term debt, because long-term debt is considered tier-two capital. It's a lower level capital, whereas the top level is supposed to be equity and equity-like instruments. So, it's entirely flawed.

Admati: Tom Hoenig submitted comments on a previous proposal on that, and I cited, in my comment on long-term debt, all of his writing about this. He was sufficiently frustrated that they're never listening to us, and this goes back to the time that he was at the FDIC and I was on the FDIC Systemic Resolution Advisory Committee, where I had more reasons to develop skepticism over the ability to resolve a systemic institution, especially cross-border. In the book, we explain very clearly why that resolution is a pipe dream and is not going to happen, and we might as well recognize that they will not do it, and they didn't do it for Credit Suisse, which proved our points once again.

Beckworth: Along those lines, there were proposals in the past, and maybe they've been tried, but proposals for CoCos, Contingent capital--

Admati: That's the same thing.

Beckworth: Oh, that's the same thing? Okay.

Admati: Some of them have high triggers and they are supposed to convert even before default.

Beckworth: So, debt that converts into equity when there's a crisis?

Admati: But the triggers are all problematic. In the US, if they trigger outside default, they are not considered debt, they are more equity-like. And so, in Europe, they're more popular, but they still didn't work, and we were promised that they will work. That's a clever proposal by academics and others, and I had many interactions with many people on it, and there is not one iota of justification for these proposals, not one.

Admati: We have a section in the original book called “Anything But Equity.” We go much more into detail about why we're skeptical, and not ever more skeptical than we were back 10 years ago, that these will work and they will not work, to the point that we are saying that if banks are too-big-to-fail— which they are, the cross-border banks— then we should really reconsider global banking in the sense of one corporation controlling [and] being systemic in multiple jurisdictions. Because there's no way, cross- border, that countries are going to agree to let each other control a single point of entry and all of the stuff that they tell you. [There’s] just no way. And I was immersed in this discussion, and I can assert to you that this is not going to work despite what they tell you.

Beckworth: Alright, well, in the few minutes we have left, I want to circle back to your chapter in your new version of the book on the rule of law and banking. I know it's broader than banking, but let's stick to banking. And you mentioned how banks will just pay fines, and they never really get into trouble. They just find ways to keep financing themselves out of any particular harm that they've caused. And a previous guest [of the podcast] years back was Jesse Eisinger who had a book on this very thing, [about] how the Department of Justice and the US justice system has gone from a world where you actually go after top executives [and] criminal prosecutions to focusing more on fines for the organizations. Do you think we need to go back to that approach, where top executives do face criminal prosecutions to incentivize better behavior of these corporations?

Reforming the Criminal Justice System's Treatment of Financial Entities

Admati: Definitely. There are multiple books written even after Jesse Eisinger's book on the subject, including by Judge Jed Rakoff, who is my key main mentor on this topic. Judge Rakoff is a senior judge here in the Southern District of New York and he wrote a book called, *Why the Innocent Plead Guilty and the Guilty Go Free: And Other Paradoxes of our Broken Legal System.* This is somebody who saw the system as a prosecutor, as a defense lawyer, and now as a judge for over 25 years. What he will tell you is that you could not actually motivate high level people to even comply with the law if they are completely immune and have shields from ever being criminally prosecuted, because even civil litigation against individuals is all paid by insurance and they all have it. The only thing that would get attention out of executives is a threat of criminal prosecution.

Admati: They obviously will defend themselves vigorously, and they deserve to defend themselves in court, but the fact that indictments didn't come after the financial crisis, including in some cases [where] they were almost wrapped in a bow by the SEC— for example about Mozilo and other people— is a real mystery. Jesse takes that on. It is quite clear that there are issues that are either in the way the law is written, or in the way it's enforced, with the ability to reach top level executives for corporate wrongdoing, including crimes.

Admati: I have a view on this-- And this is an area in which I've focused my intellectual interest in recent years, since I got a bit tired out of this debate, only to be a little bit back in it, because I wonder about the massive rap sheet of many corporations, just the harm that they cause, and why it is that the people who control them don't get into trouble for that. And that would include Boeing and Purdue Pharma and PG&E and others that sometimes kill people or endanger people, and then either use shareholder money to pay fines or to pay damages and all of that. But, somehow, some of this harm seems preventable and could have been prevented.

Admati: And so, in some cases, it's a regulatory failure. Look at FAA and Boeing, et cetera. But, anyway, I think that there are ways. I have my own ideas about it, but I'm not a lawyer. I still want to keep working on it to enable potentially lowering the criminal bar. It would not be strict liability, where an executive would go to jail for anything that happened under their watch, but it would be for, at some point, when they could have and should have prevented harm or crimes or fraud, that it would fall on them too. One day, when it's repeated and it's beyond a reasonable doubt that they could have and should have prevented-- I call it a responsible officer. There are some willful blindness statutes that might be used. So, I've immersed myself in this puzzle of compliance by corporations, and how do you compel a corporate person to obey the law.

Beckworth: Okay, well, with that, our time is up today. Our guest has been Anat Admati. Her book is titled, *The Bankers' New Clothes: What's Wrong with Banking and What to Do about It.* If you have the old edition, buy the new one. Read the new chapters.

Admati: And it’s cheap, mention how cheap it is.

Beckworth: It is much cheaper, this is true. Anat, thank you for coming on the program again.

Admati: Thank you very much for having me.