Todd H Baker is a senior fellow at Richman Center for Business, Law and Public Policy at Columbia Business School and Columbia Law School
A few days back, the news broke that the Federal Reserve Board is ready to weaken its so-called “Basel III Endgame” capital rules, which had projected a huge increase in capital requirements for the largest and most complex banks.
This is no surprise.
Opposition from the banks to capital regulation is no longer limited to rational policy papers, or even the usual antics from K Street lobbyists. It’s gone into the grassroots — or at least into the AstroTurf adhesive that passes for the grassroots in contemporary American politics. Who would have imagined that NFL fans and couch potatoes would be subjected to television ads where a supermarket checkout guy accuses the Federal Reserve of using Basel III to make Mom and Pop pay more for groceries?
And to prove that it was really, really serious about this one, the Bank Policy Institute (the lobby for large banks) hired Eugene Scalia, the Dark Lord of Administrative Law himself, to kill the proposal. Scalia never loses these cases. No wonder the Fed is in full retreat and talking about a tier one capital increase of something like 5 per cent instead of the 16 per cent originally proposed. That guy is scary.
All this drama about bank capital leads to rumination about radical change in democratic societies. In particular, it’s worth considering why one of the most influential books written about banking and systemic risk in the post-2008 era — which urged much higher equity levels at banks — has failed to influence policy. The book is Anat Admati and Martin Hellwig’s The Bankers’ New Clothes: What’s Wrong with Banking and What to Do about It, which was recently revised and updated in a second edition.
The publication of The Bankers’ New Clothes in 2013 was an intellectual thunderbolt, widely lauded and named a book of the year by the FT, the Wall Street Journal, and Bloomberg Businessweek. Martin Wolf called it “the most important book to have come out of the financial crisis”. The second edition drew more accolades from academics, public policy experts and financial commentators.
Its two core messages were fairly straightforward:
-
Less leverage and more bank equity capital — lots more equity capital, 25 per cent to 30 per cent of assets — would radically reduce the failure risk profile of banks without damaging the real economy and eliminate, or at least minimise, the enormous social costs of recurring banking crises.
-
Banks and bankers (and economists, and lawyers) wilfully promote a reality distortion field around capital levels and other matters — dubbed the Bankers’ New Clothes — that is designed to obscure that first message and support centuries of inaccurate conventional wisdom about banking.
Admati and Hellwig emphasise that high leverage in banking is a choice, not a necessity, and that banks are no different from other corporations when it comes to the benefits and risks of leverage for shareholders, management and society.
The authors built their argument on the theoretical insight of Modigliani and Miller that the total cost of issuing debt and equity to finance any given portfolio of investments should not depend on how much is debt and how much is equity.
In practice, there often is an observable difference in the cost of financing investments depending on the mix of debt and equity, usually when there is a third party influencing the difference in costs. In modern banking that third party is the government, which implicitly or explicitly protects a bank’s creditors during crises and uses tax policy to prefer debt financing over equity.
By adding leverage, bank management can increase the value of the state safety net by issuing debt at artificially low cost. Shareholders are better off when things go well, at the expense of the taxpaying public, which is bearing risks that private investors would not accept without being paid a greater interest premium.
Thus, the authors argue, when the cost to the public of providing this insurance is properly taken into account, the total social cost of investment in banks is not increased by requiring more of it to be financed by equity.
In the authors’ view, society would benefit immensely from higher bank equity levels because they would radically reduce the negative impact of systemic financial disruptions on economic activity without affecting banks’ role in supporting economic growth.
Quite a lot has happened in the decade since to book was first published to support the authors’ insights.
The long period of zero interest rates that followed the Great Financial Crisis has finally come to an end, giving money time value and making old risks new again. The Fed has further expanded government support during financial turmoil and solidified its position as “buyer and lender of last resort” for banking and nonbanking finance systems in the US and abroad. And in 2023 we experienced two (in the authors’ view entirely predictable) crisis interventions:
-
The serial failures and associated FDIC resolutions of Silicon Valley Bank, Signature Bank and First Republic Bank, followed by massive, open-ended liquidity support by the Fed for the banking system, and
-
The Swiss government assisted and financed bailout of perpetually troubled Credit Suisse by UBS.
In both cases, higher capital levels called for by Admati and Hellwig would likely have prevented the failures, or at the very least created more time for non-governmental solutions. This suggests that Admati and Hellwig’s core insight about equity levels and bank failures is correct. More capital will lead to fewer failures, and fewer failures will lead to fewer systemic consequences.
So why haven’t the well-reasoned and well-presented arguments in this book about equity capital — arguments that appear to be fundamentally correct as an economic matter — made more headway since 2013? (Let’s exclude the self-interested lobbying of the banks on this question for the moment, even if they end up winning for the wrong reasons.)
The reason that Admati and Hellwig’s proposals appear dead in the water is prosaic but eternal: absent a crisis, transitions which promise vaguely real long-term benefits at the cost of short-term pain are next to impossible in modern, democratic societies.
While we would all like to imagine that we are Thanos from the Avengers franchise, snapping our fingers to instantly establish a new, stable market equilibrium for the banking industry, a transition in the real world would be brutal in the lived reality of bank shareholders and managers. It took the 2008 global financial crisis for Congress and regulators to ignore howls from the industry and rein in megabank risk-taking, but as soon as the crisis had passed the impetus for further change dissipated rapidly.
We know what the long-term economic benefit of lower systemic risk from higher equity levels at banks proposed by Admati and Hellwig would be. So how bad would the short-term pain inflicted to get to those equity levels really be?
Bank stock prices are determined partly on earnings multiples, partly based on price-to-book value, and partly on growth expectations. If banks were to issue new shares to raise equity to the authors’ preferred levels — 2-3 times the current amount — share count and equity would rise dramatically while growth expectations would presumably be unchanged. EPS would decline by around 75 per cent, albeit with the earnings benefit from repaying some costly debt with the proceeds from the new equity damping the loss to around 50 per cent.
Return on equity would also decline. An average bank’s 10 per cent ROE on 8 per cent equity capital would likely drop to less than 5 per cent at the 25 per cent level, pushing down price/book ratios.
These things together would likely be reflected in a severe and co-ordinated drop in bank share prices — perhaps as much as 50 per cent.
For just the eight largest US banks, a 50 per cent decline in stock value would total around $750bn, somewhere between the GDPs of Sweden and Switzerland. If we were to include all the banks, the number would be well above $1tn.
These projections don’t have to be right for it to matter because, despite decades of economic and markets research and related theoretical modelling, there is still no consensus around how a large increase in the equity component of bank funding affect stock prices. The theoretical finance mathematics and economics formulas at issue implicate contentious and unresolved questions about whether a bank’s lower failure risk might drive its cost of equity down low enough to offset the ROE dilution — an argument made by the authors of The Bankers’ New Clothes — and whether the Modigliani/Miller models the authors rely on have ever been shown to work in practice.
No one should claim to be able to predict the precise impact of doubling or tripling required equity on bank stock prices. The only thing that can be said with certainty is that the outcome is uncertain. And that uncertainty, from a political point of view, is deadly.
This prospect of a near-term $1tn (or $750bn, or $500bn, or $100bn for that matter) crash in bank stock values is certain to prevent any political or regulatory endorsement for the kind of step-function capital change that the authors’ advocate. And, as the authors’ ruefully note, financial stability has no constituency . . . until there is a crisis.
There’s a second reason that this approach is impracticable. Raising banking industry capital levels within any reasonable period through new common stock offerings would effectively be impossible because the project would materially exceed the capacity of our current capital markets infrastructure.
A rough estimate shows that banks would have to raise additional capital equal to two to three times their current $2.2tn in tangible equity (or between $4.4tn and $6.6tn) to meet the authors’ 25 per cent standard.
The low end of that number is almost 50 times the $87bn in total common stock equity issued in all public offerings in the US in 2022, and far beyond what global equity markets could absorb even over an extended period, especially because all banks would be attempting to raise new capital at the same time.
There is another possible implementation path for the authors’ plan, which they note several times during the book. Banks repurchased more than $121bn in their own common shares during 2019, prior to the restrictions imposed by the Fed during the Covid crisis. They paid total dividends of about $45bn during the same period.
This implies that banks could add $166bn annually to industry capital by ceasing buybacks and dividends, all other things being equal. If you add to this the capital effect of increased banking industry earnings from replacement of expensive debt with equity, it should take about a decade to double bank capital levels.
In this scenario, banks’ return on their gradually increasing equity would decline over time while EPS would increase somewhat as outstanding debt was repaid.
But expected earnings growth beyond that would stop in its tracks because balance sheet growth, the main driver of bank earnings growth, would be impossible for nearly a decade.
Shareholders who have historically shown a preference for capital returns rather than forced reinvestment in banks would also have to be convinced that this added equity capital would be used efficiently by banks.
It is reasonable to assume that these things would have a materially depressive effect on stock prices, perhaps the same overall 50 per cent effect as if new shares were issued. Again, we lack the theoretical and empirical experience to predict the outcome with any confidence.
So now we have our answer about the short-term pain that would be experienced to implement radically higher bank capital levels and reduce systemic risk. A lot of pain to bank shareholders. Massive pain. Try explaining the need for all that pain to the voters and fundraisers by referring to Modigliani and Miller and the hidden social costs of systemic financial risk.
The unhappy truth is that the only way that anything close to the authors’ equity proposal could become law and overcome the political influence of the banks — the only time we could raise Thanos’s gloved hand and snap our fingers — would be in the context of a financial crisis so cataclysmic that banking stocks became close to worthless and the banking industry is completely discredited. A crisis worse than 2008 and more akin to the Great Depression, in short. Only then would a fundamental change like the authors’ propose be possible. This is, of course, depressingly ironic, as the whole point of increasing bank equity capital is to avoid just such a crisis in the first place.
