The debate on bank reform has reached a curious moment. In one half of the conversation, regulators are discussing how to make banks safer for society. In the other half, equity investors are discussing how to make banks safer for their portfolios. If you put the two halves of the debate together, you soon realise that the regulatory conversation is topsy turvy – at least in one crucial respect.
The regulatory discussion generally presumes that "reasonable" reform must leave banks intact. Breaking up too-big-to-fail lenders would do violence to the private sector; by contrast, demanding that banks raise extra capital is a market-friendly way to avoid taxpayer bailouts. But the actual conversation in the markets inverts this presumption. Among equity investors, breaking up banking behemoths is increasingly regarded as desirable; by contrast, boosting banks' capital is anathema. If a "reasonable" reform is one that goes with the grain of preferences in the market, busting up the banks may actually be more reasonable than forcing them to hold capital they absolutely do not want.
It is easy to see why investors are eager to dismember the big banks. The promises of synergies trotted out by empire-building bosses in the 1990s have proved largely empty; clients don't necessarily want to buy underwriting or wealth management services from the same supermarket that provides their ordinary loans. Meanwhile, the risks in empire building are evident. If even the respected JPMorgan Chase can lose billions on a sloppy trade in one wayward outpost, then imperial overstretch is everywhere. "Banks are increasingly regarded as unanalysable and uninvestable," says Mike Mayo, an analyst for CLSA on Wall Street.
Investors' scepticism shows up in share prices. The stock market capitalisations of Citigroup and Bank of America languish at half and three fifths of tangible book value, respectively – liquidating Citi could hand shareholders a gain of 100 per cent. Indeed, because banks' assets include infrastructure that could be sold for much more than book value, the bonanza might be even bigger. JPMorgan's market capitalisation is roughly equal to its book value, but analysts reckon that the bank might be worth about a third more dismembered than intact.
If the attraction of bank break-ups is obvious, so is the hostility to regulatory efforts to require banks to raise more capital. Traditionally, banks have preferred to issue debt rather than equity because the government has perversely subsidised leverage. The double taxation of corporate profits has rendered debt tax-efficient; deposit guarantees have subsidised borrowing from retail customers; central bank liquidity has made short-term wholesale borrowing artificially cheap. However, these long-standing incentives for leverage have now been fortified by two new ones.
The first comes from the realisation of banks' too-big-to-fail status. Since 2008, it has been clear that banks' unsecured bondholders stand a good chance of being bailed out in a crisis. This has created a subsidy for bond issuance, distinct from the old subsidies for collecting deposits and issuing short-term paper. Whereas in the past it was only quasi-government lenders such as Fannie Mae that issued bonds cheaply, thanks to an implicit government backstop, now all the big banks enjoy this privilege.
While bond issuance has acquired a subsidy, equity issuance has grown more expensive. To limit moral hazard, crisis bailouts were crafted so as to make shareholders suffer; thus AIG's shareholders were taken to the cleaners even as the banks that had incautiously bought its derivatives got off scot-free. As a result, equity investors today do not expect the government to rescue them. In contrast to bondholders, shareholders see banking behemoths as scarily complex, not comfortingly protected.
This too-big-to-fail effect shows up clearly in the prices of equity and debt. A non-financial corporation choosing between equity issuance and bond issuance could reasonably go either way. For example, Coca-Cola's shares sell for 17 times next year's expected earnings, meaning that equity investors demand a return of 5.8 per cent in order to buy the shares. Since Coke can issue long-term bonds at just over 4 per cent, equity is only 1.6 percentage points more costly. By contrast, Bank of America, JPMorgan, and Citigroup face a far larger wedge: 7.5 percentage points, 10.4 percentage points and 11.8 percentage points, respectively. Banks frequently moan that equity is expensive. Thanks to regulators' selective concern with moral hazard, they have a point.
Finally, as Stanford's Anat Admati has noted, bank shareholders do not want to cut leverage because of a classic debt overhang effect. Normally, the owners of a healthy company may accept the expense of issuing equity rather than debt. They do so to reduce the risk of bankruptcy, since equity investors, unlike bond investors, agree in advance to forgo payouts in hard times. But when a company is already unhealthily indebted, creditors resemble owners; the odds of bankruptcy are rising, so bondholders also face the danger that their cash receipts may stop. In this circumstance of debt overhang, the effect of deleveraging is to cut the risk borne by bondholders – or by the government that backstops them. Equity holders get stuck with higher financing costs and no compensating fall in risk.
None of this means that bank regulators should stop insisting on more capital. On the contrary, bank investors' losses from deleveraging will be society's gain. But the capital adequacy police should be aware that they are pushing against powerful incentives to evade their edicts. If regulators want a "reasonable" policy that will be accepted by the equity market, they should break up the giant banks.
The writer, an FT contributing editor, is a senior fellow at the Council on Foreign Relations
This article appears in full on CFR.org by permission of its original publisher. It was originally available here (Subscription required).