The Second Great Bailout

This post covers a new article, “Ending Bailouts, at Last,” with Amit Seru in the Journal of Law Economics and Policy, and the associated WSJ oped “Preventing Bailouts is Simple, but it Isn’t Easy.” As usual, I’ll post the full oped in 30 days.

After 2008, politicians and regulators promised the Dodd-Frank Act would stop bailouts. They failed. We document the massive bailouts of 2020-2023. But this time nobody is even promising to do anything about it. Too big to fail has spread everywhere. The basic architecture of allowing highly leveraged finance but promise that regulators will stop risks has failed. We have now tried everything else, it’s time for equity-financed banking and narrow deposit taking.

Important points:

“Financial stability” has come to mean the possibility that someone, somewhere, might lose money, even just on a mark to market basis, that an interest rate might rise, a price might fall. It has come to mean some business somewhere might undergo bankruptcy reorganization, or an individual bank might experience a run. No. A financial crisis is a systemic run,

In the same issue of the Wall Street Journal, the editorial board points to a new Treasury initiative to extend “systemically important” designation to non-bank mortgage servicers, providing them too big to fail status and the tender mercies of “consumer protection” regulation. This is a classic example of the ever-expanding definition of “financial stability.”

Another point to stress, from the WSJ oped:

How much of this bailout was necessary isn’t our point. If it was necessary, as some surely was, the question is why the financial system was still so fragile that it needed another bailout. And in its wake moral hazard has unapologetically expanded. If you saved and bought a house with cash, if you saved and went to a cheaper college rather than taking out a big student loan, if you repaid loans promptly, if you kept some money around to buy on the dip, you lost. The lesson: Borrow. Buy risky debt.

Again, the point here is not to criticize the decision to bail out. The point is that Dodd-Frank, and the army of regulators, failed at their mission to make the economy robust, so that leaders would not feel that bail outs would be needed.

A final teaser:

We have seen the benefits of an abundant-reserves regime, in which banks no longer scramble to just meet reserve requirements, and in which reserve requirements no longer constrain bank lending and deposit creation. We need a parallel abundant-equity regime…

*****

The introduction and conclusion of “Ending Bailouts, at Last,” edited a bit, and without footnotes:

In 2008, we had a financial crisis. Our government responded once again with bailouts. Bailouts keep existing business going, and most of all protect creditors from losses. The instruments vary, including direct creditor guarantees like deposit insurance, mergers of failing companies with sound ones sweetened with government money or government purchases of bad assets, or government purchases, guarantees, and other efforts to prop up security prices and thereby cover up losses. Since actual or promised (contingent) resources flow from taxpayers to financial market participants, we include all of these interventions as “bailouts.”

Ex-post protection breeds ex-ante risk taking or moral hazard, however. If deposits are guaranteed, depositors have little incentive to seek out safe banks. If banks, financial institutions, and other companies will receive bailouts and are therefore unlikely to default on loans, creditors have little incentive to seek out safe companies, and companies have less incentive to make safe investments.

Recognizing this danger, and responding to public outrage over bailouts, our government promised during the 2008 financial crisis to address moral hazard once the storm had passed. It made good on that promise with a vast expansion of financial regulation under the Dodd-Frank act. Similar approaches were followed internationally, under the Basel international regulatory umbrella. Whether or not one approves of the outcome—we are mostly skeptics—at least one must grant the effort.

2008 was not the first time…

It just happened again. Fearing another crisis due to the natural and policy-induced economic dislocations of the COVID-19 pandemic in 2020, our government bailed out, breaking many of the Dodd-Frank promises. And again. Silicon Valley Bank and First Republic suffered runs in 2023, triggered by old-fashioned interest rate risk that somehow the army of regulators had completely missed. Credit Suisse failed, and its regulators threw out the resolution plans. These events laid bare that the basic architecture of current financial regulation—allow fragile financing, but count on regulators to contain risk—has failed.

Except, scandalously, this time neither government, nor Fed, nor other regulators have even acknowledged that anything was wrong with these bailouts. There are no “What went wrong?” inquests, no acknowledgement that bailouts induce moral hazard, there are not even promises to mop up moral hazard someday in the vague future. As unproductive as it would be, there is no concerted effort to reform the rule book once again to contain moral hazard or to pre-commit against ever larger bailouts. (The massive “Basel III endgame” rule expansion is not motivated by the failures of 2020-2023.) The main reaction is a self-congratulatory pat on the back for saving the world by spreading out immense amounts of bailout money. Bailouts are the new regime, the new norm, and expected by financial market participants in the next crisis. Perhaps the lack of another popular revolt at “bailing out the banks” led to the unusual quiet, but a technocracy which only reforms when the peasants are outside with pitchforks is not healthy.

Too big to fail is now enshrined. But small companies get bailed out too, and their creditors. Industrial companies, not just financial companies, are protected. Too leveraged to fail might be the summary of our new regime. But our authorities subsidize leverage, with tax deduction and regulatory preferences for debt. As a result, there is every incentive to take risk, to borrow and to lend, with confidence that the government will backstop debt, prop up prices, and keep companies afloat should any serious crisis develop. There is little incentive to issue equity rather than borrow, to keep cash around to provide liquidity or hunt for bargains and thereby prop up prices with private money in the next moment of stress. Why be ready to bargain-hunt when you know the government will front-run you and keep prices from falling?

…

The bailout-and-regulate spiral must end. The promise of Dodd Frank to finally regulate away risk and bailouts has failed. Inflation shows us that the government is near its limit to borrow and print money to fund bailouts. We have one last chance to construct a bailout-free financial system. Fortunately, plans for such a system are sitting on the shelf. They need only will to overcome the large private interests that benefit from the current system.

…

Conclusion:

Fortunately, there is a straightforward way out. We can construct a financial system that is immune from private sector financial crises and hence the need for bailouts. It can be as or more innovative and functional as the current one, giving savers ample returns and borrowers ample access to credit and investment capital. The blueprint has been around since the 1930s. Arguably, modern information, communication, and financial technology makes it even more easily achieved than when first conceived.

First, we must restore clarity on just what “financial stability” means, and what events are, genuinely, in need of a regulatory response. “Financial stability” has come to mean the possibility that someone, somewhere, might lose money, even just on a mark to market basis, that an interest rate might rise, a price might fall. It has come to mean some business somewhere might undergo bankruptcy reorganization, or an individual bank might experience a run.

No. A financial crisis is a systemic run, when people run to get cash out of short-term promises all over the financial system, including healthy institutions, and the capacity of the financial system to function is imperiled. This is what happened in 2008. Other events are not crises.

Likewise, “contagion” has become overused, a dark yet vague fear that somehow any ripple anywhere might bring down the financial system. To laypeople it sounds like a technical term, but it has evolved to no meaning beyond this vaguely stated fear. Contagion requires a mechanism. If there is a run at one bank due to losses in one particular kind of security, other banks with similar exposure might suffer runs. That’s a sensible “contagion,” though propping up the first bank might do little to stop such a run at the second. But if other banks do not have similar exposures, and that is well known, such “contagion” will not happen. Central bankers spoke of “contagion” from Greece to Italy. Why? Italy did not own any Greek debt. At best we learn from a Greek default whether or not the rest of the EU will bail out Italy, but that is not the usual meaning of the word. We should only use the word “contagion” along with an explicit mechanism.

With this understanding, it is possible to pre-commit against many bailouts.

But other bailouts loom for good reason. Once a run is underway, a creditor bailout is really the only way to stop it, and governments will (and must) stop it. …

But protection leads to too much risk taking by investors and by bankers. So in our sequence of financial crises, over and over again, authorities bailed out creditors to stop a run and then passed regulations to try to constrain risk taking so another larger crisis would not break out. The Dodd-Frank and Basel approaches were not anything new, they were just the latest in a centuries-long cycle.

Events since 2020 do not break this history by the bailout. They break this history by the unusual lack of any interest in containing moral hazard so the next one is not larger.

How can we escape the treadmill? The ingredients are simple, First, risky financial investing, like risky corporate investing, must be financed by equity and long-term debt, which are securities that cannot run. When a stock-financed company loses money, you can’t run to get your money out and bankrupt the company when it can’t pay you. The price goes down instead. Second, any run-prone securities such as deposits must be fully backed by interest-paying reserves or short-term treasury debt.

Equity-financed banking and narrow deposit-taking (we avoid the word “narrow banking” on purpose) is well described elsewhere, as are clear responses to all the standard objections. No, borrowers will not be starved for credit. They can get as much as now, and at good rates. The converse is one of the most persistent fallacies surrounding equity. Jay Powell himself, said of a two-percentage point capital increase “raising capital requirements also increases the cost of, and reduces access to, credit.”

This is simply not true, as Admati and Hellwig and many others have proved time and again. Additional equity has no social cost, and indeed has a social benefit. It carries a big private cost to banks and their current shareholders, which lose too-big-to-fail bailouts and guarantees courtesy of taxpayers. Banks predictably decry any attempt to raise capital, and are persuasive to regulators as well.

…

We have seen the benefits of an abundant-reserves regime, in which banks no longer scramble to just meet reserve requirements, and in which reserve requirements no longer constrain bank lending and deposit creation. We need a parallel abundant-equity regime. For example, the turbulence in Treasury markets in 2020, in which dealer banks refused arbitrage opportunities, has been chalked up to the fact that they were up against capital budgets, and, crucially, they were not willing to get more capital even to finance arbitrage opportunities. The debt overhang keeping banks right at capital constraints disappears when capital is abundant.

The Federal Reserve and international banking regulators are now finalizing a “Basel III endgame” proposal to strengthen big-bank regulation. It’s a large and complex proposal. Most of it is a long addition to the hundreds of thousands of rules we have now, adding to risk assessment rules that just failed so miserably at SVB and Credit Suisse. David Wessel writes perceptively, “The proposal fills 316 pages of small type in the Federal Register . . . Few people besides regulators, executives of banks that would be affected, and their lawyers understand the details.”

The headline 16-percent increase in capital sounds like a lot, but 16 percent is only 2 percentage points, since capital is so low already. Abundant capital requires 20 or even 50 percentage points more capital. How much, exactly? So much that the precise number doesn’t matter, because banks will never fail. How do we get there? We need not reform the current giants, or rewrite the current rule book, taking another years (from Dodd-Frank to Basel III). It would suffice to simply get out of the way, to allow equity-financed banking and narrow deposit taking to emerge, with the light regulatory touch such run-free institutions require, and let the flowers bloom. If the market value of equity and long-term debt is more than, say, 80 percent of the value of liabilities, the financial institution needs no asset regulation and can do what it wants, regulated no more than any other company. If a bank instead wishes today’s capital structure, it faces today’s regulations. A simple regulatory tax on short-term debt financing can also gently provide a nudge.

The Federal Reserve, which has been on a legal warpath against narrow deposit takers, could simply follow its legal mandate and allow them. A gold star for “can’t possibly cause a run” would be nice too, instead of the current silly claim that allowing this enhanced form of money market fund would spark runs elsewhere.65

Simply allowing equity and long-term debt financed investment companies and narrow deposit takers and transactions service providers to operate would allow them to expand.

The absence of government guarantees would also have a salutary effect on financial stability. Your fire sale is my buying opportunity. There is little incentive now to hold some cash aside, as the Fed will jump in during any bad time and outbid you. When prices can fall without fear of a systemic run, then there will be lots more private capital available to jump in and make sure prices don’t fall.

Naturally, it would also be a financial system in which new innovative entrants can come, and old dysfunctional businesses can go.

Standing in the way, of course, is a vast armada of financial institutions that profit from the current game, that have invested hundreds of millions in regulatory compliance/barriers to entry, and that profit from risk taking in good times knowing they will be protected in bad times, along with a lot of obfuscation from financial market analysts.

Also the regulators, whose livelihood depends on deep human capital of the current system, their relationship to a financial industry, and their presumption of technocratic competence to manage even tiny details of the financial system will surely not be pleased at such a fundamental reform. Capture goes both ways. Their ability to tell financial firms where to invest will collapse as well.

But this is politics, not finance. If we could just get to the point of agreeing that there is a problem, that the current system will collapse, that there is a clear solution, and all that stands in the way are vested interests, then we would have made a lot of progress.