Silicon Valley Bank is a very American mess

Daniel Davies is Managing Director at Frontline Analysts, the author of lie for moneyand co-author of The Brompton.

If you think back to the Great Financial Crisis, one of the big things you will remember is that for the first time in living memory, the big banks were subject to liquidity rushes.

Banks like Northern Rock, HBOS and Dexia had gone too far, funding long-term assets from short-term liabilities, and either demanding massive central bank bailouts or inflaming themselves. You may also remember that in the aftermath of this crisis, regulators assured us all that the rules had been changed to prevent this sort of thing.

So what happened with Silicon Valley Bank (and Silvergate)? You will laugh.

To some extent, of course, borrowing short to lend long is what banks do. For. Thus, the rules that were adopted under the Third Basel Accords (“Basel III”) were never intended to completely ban the practice, but simply to impose reasonable limits on it and to say “come on, be reasonable “. There were two dimensions of being sensitive that had to be applied, with two corresponding ratios. (Regulators think in ratios).

The first is the “Liquidity Coverage Ratio”, which aims to measure emergency financing capacity. Simply put, it’s the ratio of the amount of “high quality liquid assets” you have, compared to a rough estimate of cash outflows you could experience over 30 days if your wholesale funding dried up and some (but not all) of your retail and business deposits have been executed. The ratio is assumed to be greater than 100%; it roughly corresponds to a “survival horizon” of thirty days.

The second is the “net stable funding ratio” – also supposed to be above 100% in a good bank – which measures of construction funding liquidity. It’s basically a ratio of the two sides of your balance sheet, after each side is adjusted for liquidity risk.

Assets are weighted according to their ease of turning into cash. This means that short-term Treasuries get 100% weightings, while longer-term corporate bonds get 50% weightings and loans get zero. Liabilities are weighted by the likelihood of someone wanting cash back immediately. Thus, overnight repo is weighted at 100% and retail deposits at 90%. Hot money wealth management and corporate deposits only get a 50% weighting, and long-term bonds/deposits get zero. If the ratio of the first number to the second is greater than 100%, then, generally speaking, your long-term, illiquid assets equal an equivalent amount of long-term, stable liabilities.

This all sounds pretty sensible. So why didn’t this settlement prevent SVB from . . .


This is from the latest 10-K from SVB. When the Fed implemented Basel III in October 2020, it took advantage of the fact that strictly speaking, the Basel accords are only internationally agreed to apply to “large internationally active banks.” While most jurisdictions apply the Basel rules to their entire banking system anyway, the United States has a strong and powerful community banking lobby, and US community banks are generally quite aggressive in their use. of the borrow short/loan long business model.

Thus, the Fed adopted a rule that only very large international banks were subject to the full Basel NSFR requirements (several of these large banks are in fact holding companies for foreign institutions). It adopted a second level, under which the ratio was only to be 85%, and a third level where it was calibrated at 70%. And even then, the majority of US banks are not required to follow NSFR or LCR standards at all.

Despite being the 16th largest bank in the United States by balance sheet size, SVB was apparently not subject to “no more Dexias, no more HBOS” regulations. The reason, as implied by the 10-K disclosure above, appears to be that a bank is only required to follow the NSFR and LCR rules if it has a certain amount of “short-term wholesale funding” and the passive side of SVB was dominated by deposits from corporate clients.

Of course, as we can now see, the fact that a risk is not covered by a regulatory ratio does not mean that it does not exist.

Despite grumbling and moaning in the 2010s (NSFR compliance in particular has been a big drag on European bank profitability), European and UK banks have essentially managed to comply with funding rules, which have consideration merely codifies reasonable cash management practices.

The fact that the major domestic banks in the United States are apparently allowed to manage such large funding mismatches (and in effect, to trail with massive unrealized losses on portfolios of held-to-maturity securities, which is perhaps a regulatory story for another day) are likely to be a source of embarrassment for the American authorities during their next visit to the great Swiss tower.

Leave a Comment