The Bear’s Lair: How to regulate banks

Chorus of Peers from Iolanthe

First Republic board meeting

The collapses of Silicon Valley Bank, Signature Bank and First Republic Bank, all within six weeks, are a drastic failure of both bank regulation and Fed monetary policy. This column has written incessantly about the idiocies of Fed monetary policy and its GOSPLAN approach to interest rate setting; I would now like to focus on bank regulation and make some suggestions about how it might be improved.

The Dodd-Frank system of U.S. bank regulation, instituted in 2010, is manifestly a disaster; it is now responsible for the second, third and fourth largest bank collapses in U.S. history, within little more than a decade of the legislation’s passage. It rests on a fundamentally false premise: that wise regulators can order the economy, in particular the banking system, and forecast reliably what might go wrong in any given period, taking steps to rectify it.

As an example of this, the 2022 Fed “stress tests” imposed by Dodd-Frank, under which banks test their capital in a “severely adverse scenario” had interest rates in the “severely adverse scenario” of zero on 3-month T-bills and 0.75% on 10-year Treasury bonds; it never occurred to the “stress testers” that interest rates might rise to the extent they have. Needless to say, SVB Signature and First Republic all sailed through their “severely adverse scenarios.” If regulators cannot imagine the possibility of substantially higher interest rates, at a time (early 2022) when annual inflation is already around 8% and rising, they serve no useful intellectual function in the U.S. banking system.

We should now differentiate between the three banks concerned. SVB was genuinely abominably run, with no risk management chief for several months of 2022, a Treasury operation that appears to have believed the Fed regulators’ interest rate forecast and bet the store on long-term Treasuries, and a loan book that included all kinds of dodgy loans to Silicon Valley’s feckless entrepreneurs. In the artificial boom of 2009-21, it did very well indeed, melding perfectly with the aspirations and beliefs of its youthful “woke” client base; when reality returned in late 2022, it had already lost all its capital on a “mark-to-market” basis, even before you considered the potential horrors in its loan portfolio.

Signature was less obviously badly run, beyond a fatal decision to emphasize service to the crypto-currency industry, which inevitably left it with large losses after the FTX collapse. Its other emphasis was on the murkier areas of New York City real estate, already in trouble before the bank’s demise.

First Republic was a different case; its business strategy appears to have been based entirely on snobbery. Far from dealing with ordinary plebs, like other deposit-taking banks, it focused on the very rich, of which there were a huge number around San Francisco in the last decade, financing their offices and overpriced dwellings. The bank’s managers might have been House of Lords members in Gilbert and Sullivan’s 1882 opera “Iolanthe”:

Our lordly style
You shall not quench
With base canaille!
(that word is French);

Distinction ebbs
Before a herd
Of vulgar plebs!
(A Latin word);

‘Twould fill with joy
And madness stark
The hoi polloi!
(A Greek remark);

One Latin word, one Greek remark, and one that’s French.

In a period when interest rates are held artificially low, this is not a bad strategy – after all Bernard Arnault of the luxury conglomerate Moet Hennessey Luis Vuitton, is the world’s richest man, and luxury banking plays to the same human frailties as luxury goods markets in general.

The luxury-banking strategy comes seriously unstuck however when interest rates rise and the economy returns to normal. The volatility of high-end house prices and prestige office space is far greater than that of mid-range real estate, so when interest rates rise and prices fall, they fall much further. When that happens, lending against such assets becomes suicidal. Add First Republic’s problem of too many deposits above the FDIC’s $250,000 limit, financing long-term impaired-value real estate loans, and you have a recipe for collapse. At least SVB held warrants on many of the start-ups it financed, whereas the quants have not yet invented an adequate warrant on housing, which would have given First Republic extra profits on any loans that remained good.

To lose one substantial bank in an economic downturn may be an accident; to lose three looks like carelessness. SVB was so badly managed it could have turned turtle in a financial flat calm; the other two banks were victims of the overwhelming error of our time: the attempt by leftist governments to overrule market price signals. Had the Fed not held interest rates far below market levels, First Republic, at least, would have profited from its snobbery, as high-end real estate would not have become drastically over-inflated and high-end consumers would not have become in many cases foolishly over-leveraged.

The same problem appeared last week in a more modest realm of housing finance, loans guaranteed by Fannie Mae and Freddie Mac, available to a maximum of $726,200 in most areas and $1,089,300 in “high-cost areas” (the “high-cost areas” designation is itself an interference with the market, since it encourages further price run-ups in those areas, making the designation self-reinforcing). Rather than have a free-market system of housing finance, in which willing lenders interact with willing borrowers (and, by all means, securitize the results if they wish) the United States has chosen to have this absurd French-style system in which all loans to the middle class must be guaranteed by one of two government-sponsored behemoths, whose incompetence is such that they both went bankrupt in 2008 and may very well do so again.

Now the Biden administration wants to tweak the playing field, by increase the “origination fees” paid by potential borrowers, so that those with good credit scores pay more than the near-deadbeats – the real winners being those with poor credit who borrow more than 95% of the value of their house.

As even SVB’s risk managers could probably tell you, this will have two results. First, there will be a gigantic mass of high-leverage home loan applications by those with poor credit, all of which will default in the next housing downturn, as they did in 2008, thereby causing an even worse banking and economic crisis than the two we have already had.

Second, and much more positively, the increase in premiums for good credit risks should open a market for direct mortgage lending, without intervention by the bureaucracy-proliferating, incompetent Fannie Mae and Freddie Mac. Applying for a direct mortgage loan should be considerably simpler than for a Fannie/Freddie guaranteed loan, because the lending bank will be able to streamline its procedures, thereby attracting more of these good-credit-score customers, who by definition have lower default rates than the average (of course, sudden illness and redundancy can strike anywhere).

I see only two obstacles to this wholly welcome development. First, if the lenders want to securitize the mortgages they create, there will doubtless (because of the securitization procedures) be bureaucracy similar to or even worse than that of Fannie/Freddie. Second, the recent crisis has through removing many of their large (over the $250,000 FDIC limit) deposits, weakened the creditworthiness of the regional banks who provide what competition remains in the lending market, forcing it even more into the clutches of the banking behemoths. If bank lending becomes an oligopoly on a national basis, as seems possible, then there will be no competitive forces leading the banks to direct bank mortgage lending, and the pallid blight of Fannie/Freddie will remain, strangling the market.

A national banking oligopoly will have an even worse effect than bureaucratizing the home mortgage market: it will make it even more difficult for small and new companies to obtain banking services, unless they are connected with the private equity industry, avid creator of dozens of new companies exploiting market opportunities that appeared five years ago. That will move the U.S. economy even further from capitalism, and further towards a corrupt government-direct socialism that will force living standards and productivity into eternal decline.

There are thus two lessons to be drawn. Regulation is useless, and regulators have even less idea of the impending threats to the system than bankers themselves – so abolish it, as far as possible, and make banks hold 20-25% of their assets in capital, as they did in 1928. Second, financial crises are almost all caused by governments attempting to interfere with market forces, often, as with the Biden loan fee tweaks, in pursuit of deluded social goals. Stop doing this, and crises will disappear as if by magic.


(The Bear’s Lair is a weekly column that is intended to appear each Monday, an appropriately gloomy day of the week. Its rationale is that the proportion of “sell” recommendations put out by Wall Street houses remains far below that of “buy” recommendations. Accordingly, investors have an excess of positive information and very little negative information. The column thus takes the ursine view of life and the market, in the hope that it may be usefully different from what investors see elsewhere.)