Shadow banks are not outside the banking system

Unlock the Editor’s Digest for free

Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.

Good morning. Yesterday, ADP’s private payroll report landed with a splat, showing a loss of 32,000 jobs. There are always complaints about the reliability of the ADP numbers, but tough luck: until the end of the government shutdown, there will be no payroll reports from the Bureau of Labor Statistics. Readers with advice about writing a finance column in the absence of official economic statistics should get in touch: unhedged@ft.com.

Shadow banks II

Earlier this week Unhedged pointed out that, in the US, bank lending to non-bank financial firms — “shadow banks” — accounts for all of the growth in bank lending this year. At $1.7tn, shadow bank lending now accounts for about 13 per cent of all US bank lending, and a significantly higher proportion than that at the largest banks.

How much should we worry about the wild proliferation of lending from banks into shadow banks (known, in the jargon, as NDFIs: “non-depository financial institutions”)?

When a bank provides credit to an NDFI rather than a business or consumer, it is moving from direct to indirect lending. A bank might lend to a private capital fund, rather than to businesses; it might provide bridge financing to a credit manager to securitise credit card receivables, rather than lending directly to consumers; or it might provide prime broker financing to a hedge fund, rather than engaging in proprietary trading itself. And so on.

With indirect lending, three things might leave the bank and the banking system: risk, return and capital. Ideally, the three leave in proportion. When the private credit fund intermediates the lending, it takes on some of the risk — the fund’s investors are in line for the first losses on the loans. The fund earns some profit in return for taking that risk, over and above what it pays the bank for the financing. And the bank, because it is taking less risk, can hold less capital than it would have had it made the loan directly. 

The problems arise when the transfer is not in proportion — when, for example, risk stays with the bank, but profit and (especially) capital leak out anyway.

In the NDFI lending boom, is risk staying in the banking system, while returns and capital leave? From the outside, it is extremely hard to tell.

NDFI loan disclosures in the weekly “call reports” required by regulators don’t tell you very much about the banks’ economic exposures. The banks have to report their NDFI lending in five subcategories. Here’s how NDFI loans broke out by category as of the end of the second quarter, for banks with more than $10bn in assets:

That’s fine as far as it goes. But that’s not very far: other than mortgage credit, how much do these classifications tell you about what sort of underlying assets a bank is exposed to when it lends to an NDFI? Is it exposed to commercial real estate? Subprime auto? Supply chain finance?

One large regional bank, for example, carries $62bn in NDFI loans — about a fifth of its loan book. Ninety-five per cent of that is to “business”, “private equity” and “other”. Knowing that, are you any wiser about what the bank’s NDFI exposures are, ultimately? Not much, and don’t look to the bank’s quarterly disclosures for help. The 10-Q reports don’t mention NDFI lending. The bank’s supervisors can see the individual loans, of course. Investors just have to hope the regulators are on top of things.

Next, banks’ exposure to NDFIs will vary depending on the condition of the economy, individual industries and particular creditors. This is in part because of the way banks and NDFIs work symbiotically, as Viral Acharya, Nicola Cetorelli, and Bruce Tuckman argued in an excellent paper last year. They make the point that while NDFIs can do lots of the things banks do, they cannot compare to banks as repositories of liquidity, because they can’t take deposits and they don’t have access to official liquidity backstops. So the NDFIs acquire “liquidity insurance”, usually in the form of credit lines with banks. These lines are the “umbilical cords” connecting the NDFIs to the banks, as Acharya put it to Unhedged.

At moments of stress, these credit lines get drawn, and banks’ exposure to the NDFIs and whichever they are lending to rises. Indeed, you can see a small example of this in the first chart above, in the little bump up in NDFI loans in 2020. In the early days of Covid-19, NDFIs drew on the credit lines with banks, to have cash on hand just in case. The particular worry in cases like this is what bankers call “wrong way risk”. A subprime car lender, say, is most likely to draw on its bank credit line at a moment when the value of subprime credit loans is under acute pressure. So the bank’s exposure to the subprime lender increases just as the collateral backing its loans gets less valuable.

How much variable exposure to shadow banking does the banking system have now, given the fast growth of NDFI lending? And how much wrong-way risk is out there? I don’t know, and I’m not sure anyone else does, either.

One good read

Art is not a store of value.

FT Unhedged podcast

Can’t get enough of Unhedged? Listen to our new podcast, for a 15-minute dive into the latest markets news and financial headlines, twice a week. Catch up on past editions of the newsletter here.