TLDR:
Last year, US banks were initially ‘forced’ to lend, then cut lending sharply
They are way below the pre-crisis trend in their levels of lending
Their results showed great profitability from lending
They arguably should be stepping up and there’s some evidence they are
If they don’t step up as govt pulls back that’s a signal for risk off
Most people who follow macroeconomics and finance now agree that the ‘money supply’, if there is such a thing, is determined by the behavior of banks and the government. In the Covid policy response, we saw an unprecedented (in peacetime) expansion of money balances across developed economies driven by government spending. Government spending on its own doesn’t increase money balances necessarily, but it does if accompanied by central bank asset purchases. Net net, this is just the government crediting money to peoples bank accounts. Normally, taxes or bond issuance mean that government spending is a wash with some private money balances reduced and others increased, but QE extinguishes the effect of bond issuance. That has led to an unprecedented increase in monetary aggregates in the US. There is monthly data at the whole economy level, but who cares how much money pepsi and bank of america have on account really. This blog is read by (a few…) human beings not bodies corporate, so here is the slower moving ‘ checkable deposits of households’ series from the US financial accounts:
From a base of 1trio, we now have 3.4trio of balances on US households’ bank accounts. But the other really remarkable thing about this number (apart from sheer size) is that for the past year or so, the effect of banks lending decisions has been going the other way - reducing money balances. Net bank lending increases money balances, and reductions in the stock of lending reduce them. Here’s the stock of US bank lending as measured by the Fed’s H8 release:
Initially, US banks were getting with the program - though not of their own volition. As Covid hit and the US economy locked down, corporates drew down pre-agreed ‘revolver’ credit lines arranged with banks - to the tune of USD 1trio. You can see what the banks thought of being forced to be counter-cyclical providers of credit in what happened next - they cut lending elsewhere to the tune of half a trillion $. I’m not an economist so I don’t have to justify the lines on charts I draw with models, so here is one I drew earlier, on the same series but logged:
A straight line on the log chart means a constant rate of growth. For all the theorising one might do about why banks lend, it appears to be reasonably approximated by 5-6% annual growth all else equal. If you believe that this is a choice variable for banks, and I do, then it’s not unreasonable to think that they didn’t want to be above the line - they were forced to be by the nature of their revolver obligations. These pre-agreed credit lines built a kind of counter cyclicality into the system. The response post june 2020 though shows exactly why the banks can’t be trusted with the overall fate of the economy however. In response to being taken away from their desired rate of credit growth, they not only cut lending back to the trendline but through it - ‘correcting’ their oversupply with a period of relative austerity.
Fortunately, we were not relying on the banks’ public spiritedness. QE + Deficit spending = growth in money balances so whilst the banks were undermining spending power on the one hand, the government was bolstering it faster on the other. Household deposits have grown faster in the H2 2020 - H1 2021 period than in the initial pandemic response as the fiscal response was greater than I, and presumably the banks, thought possible. I don’t say this to excuse my own poor forecasting, it’s both what the banks said at the time in their own forecasts and is reflected in their lending behavior. Credit provision is a strategic game. If you are generous and your competitors are also generous, your loan performance will be good because most of your borrowers incomes will be sufficient to repay their loans. If you are stingy when your competitors are generous, you will lose out on profitable lending opportunities. If you are generous when others are stingy, you will suffer defaults. If everyone is stingy, everyone will suffer. Banks in 2020 were expecting lean times - and said so. They told investors to expect poor loan performance, and borrowed extensively themselves in wholesale markets. Just like me, they didn’t expect the scale of the government intervention - which has meant that so far, their loan performance has been far better than anticipated. A selection of quotes from banks’ latest results:
BofA: ‘Provision for credit losses decreased $6.7 billion to a benefit of $1.6 billion, reflecting a reserve release of $2.2 billion amid an improved macroeconomic outlook’
JPM: ‘Net income was $11.9 billion, up $7.3 billion, driven by credit reserve releases’
Citi: ‘Citigroup’s end-of-period loans were $677 billion as of quarter end, down 1% from the prior-year period on a reported basis and 3% in constant dollars, driven by declines across GCB and ICG, reflecting higher repayment rates.’
My translation? ‘We expected a bunch of losses that never materialised’. Back to our log chart, bank loans have now spent as much time as far below the pre-crisis trend as above it. If I were an investor in these banks, having observed higher loan profitability and fewer write-downs than expected whilst they cut their lending, I would be asking that they please pull their proverbial out and get lending. And indeed, there is some evidence they’ve already begun. My favourite US economic commentator published, as I was just finishing this piece off, his summary of the Senior Loan officer Survey showing increased supply and demand for credit in Q2. As he says, that typicall leads lending by 1-2 quarters and improved materially in Q2. Banks may already be getting going.
The main counterargument I’ve already heard to this view is that credit ‘demand’ is weak. The problem with that as a concept however is that credit demand is functionally unlimited - someone out there will always want to borrow at the going rate - the drummer from my old band reached out recently asking to borrow ten thousand pounds to make a Christian rock album for example. Not only was he offering to pay market rates but rather more! Yet I still felt moved to decline . Banks are offered the choice of borrowers but only from a fixed pool in a given period of time. If macroeconomic conditions or their loan profitability are arguing for more lending, then they will move down the credit spectrum. What they hope to avoid is the next available borrower being my old drummer - better to lend to a slightly better credit now than have to offer him a loan later if the banks CEO demands growth! As oligopoly providers of credit to the economy, banks are in more of a strategic game than a competitive market.
In summary, banks have been demonstrably conservative and pro-cyclical through this crisis as they often are. In 2008, the government response was nowhere near enough to make up for this and we felt the effects for a decade of weak demand - with even the tightening of 2017/18 being enough to spook the Fed into keeping monetary settings loose. 2008 though was a financial crisis - starting on balance sheets. Covid was a real economy crisis, and balance sheets have come out of it looking good thanks to a much stronger fiscal response. Banks will likely take advantage of that strength to grow their loan books, keeping the economic expansion going in the process. If they pull back, that’ll be a great time to short risk.
NB: This post is not investment advice and is not a trade recommendation. The views expressed here are my own and do not reflect those of my employer.