regulation is the scientist, supervision is the monster
SVB, considered as a specific case of a general feature of systems
(I have been consciously patting myself on the back all week for not doing a substack on the Silicon Valley Bank failure, being true to the “slow news” manifesto and all that, then today I realised I was just being annoying. There is a lot to say within the context of “back of mind” and its subject matter, and the high level principles here are if anything more important than the tick-tock of the specific failure).
If you are a fellow veteran of / survivor of / person out on parole from the world of central banking, you know that although “bank regulation” and “bank supervision” are used interchangeably or as “second uses” for each other, they technically mean different things. Regulation is the rulebook, and supervision is the activity of enforcing it, checking compliance and giving instructions when rules are breached.
(That means that there are distinct policy divisions in places like the Bank of England. “Regulatory policy”, where I worked, is the having of deep thoughts about the principles and economics of the regulatory system. “Supervisory policy”, on the floor above me, was the thinking of thoughts about general approaches to how the rulebook was going to be prioritised, which things were going to be pushed hard and which let slide, what exceptions could be made and how people could be pushed into better compliance. As a young person, I thought I was in the interesting bit, but maybe I wasn’t. There’s a professor at Harvard called Malcolm Sparrow who makes a case that supervisory policy – the structured use of discretion in the enforcement of rules – is the key to a lot of governance. More, much more to come on that).
A consensus is developing among people who know what they’re talking about that SVB was more of a failure of supervision than of regulation. I am not going to gainsay that consensus, even though my first reaction was to write about regulations that hadn’t been implemented properly. But the more I look at what happened, I end up thinking… not that it’s wrong, but that the distinction itself is less clear than I’d realised.
At a high level and in very brief summary: there have been regulations since the 2010s which are meant to set limits on the ability of banks to “borrow short / lend long”. They’re not meant to prevent it – how could you prevent a bank from doing liquidity transformation, that’s what it’s for! But they put limits on it. In an important sense, most bank regulations are really just codifications of best practice (in this case, sensible management of a bank’s treasury). Their purpose is really to draw a line under best practice, in an industry where people who don’t follow good practices can gain a lot of market share really quickly.
The USA didn’t do a good job of implementing these regulations – it only applied them properly to a category of banks which didn’t include SVB. You don’t really need to know much more than that about the global regulatory politics.
But I just said that these rules only codified good practice! That means that the responsibility for the failure lies on the management who ran a core banking function in a really stupid way. (I agree and I think I know why). And it lies with the supervisors at the San Francisco Fed, who ought to have stepped in as soon as they saw what was going on. That’s the gist of the argument.
And I … agree? There certainly needs to be a number of interviews without biscuits – this thing had unrealised losses bigger than its shareholders’ funds, on a portfolio of 30 year bonds financed with overnight money. That’s many levels of “come on, you can’t be doing that”. But …
My half formed thought here is that the regulations aren’t just regulations; each regulation contains a bit of supervisory policy. That’s simply because each regulatory ratio is enacted with a bit of legislation which also specifies things like “exactly how the ratio is calculated” and “how often do you report the ratio”. This last one is really, really important.
The act of calculating the ratio is itself a tool of supervision. When you don’t implement the ratio, you also fail to implement a whole schedule of reporting.
That means the supervisors of SVB didn’t have access to a set of regular reports which would have summarised the state of liquidity management practices at the bank. To get that information, they would have needed to step in and find it out themselves.
Which would be difficult. The US supervisory system, outside the very big banks where they have to be a bit internationally consistent, is set up for small community banks. Small community banks have relatively simple books, and relatively unassuming executives who will usually treat bank supervisors respectfully and can be intimidated if they don’t.
A big bank (SVB was 16th largest in the USA) is a more difficult proposition. If someone wants to sit down and argue that their treasury practices are just fine, that you’re a dumb bureaucrat who doesn’t understand (and they will), it is bloody difficult to gainsay them. Institutionally – their boss probably has a better relationship with the top management of the Fed than you do – but also informationally.
“No, your treasury practices are insane and dangerous” is a difficult technical argument which might have to be extracted from megabytes of inconsistent reports. “If it’s so great, why can it not produce this simple ratio, compliance with which is required by law” is a much easier conversation.
I said in a previous mail that every accounting system is a prison, but an accounting system can also be a battleship. Supervision and regulation are like (because they’re specific cases of the general concepts of) action and information. They’re mainly separate things, but the ways in which the distinction breaks down might be more interesting than the distinction itself.