The Banks Loosen Up
I had given up on writing weekly articles pegged on the Fed balance sheet. I had gotten bored telling you that the banks were hoarding cash, and judging from readership, you were bored too.
But 17 weeks into this, we finally see some signs of a slight thaw. There are three main things happening:
- The Fed throws a lot of digital cash at the banks via QE, repo, and the facilities.
- Reserves have sucked up about half. This is the banks hoarding cash.
- The other half has been sucked up by Treasury through debt sales exceeding spending, which is saying something. This is Treasury hoarding cash.
Every dollar and then some was spoken for.
But now we see a slight thaw, and banks have loaned out $45b more than they have taken in from the Fed, though some of that is likely defaults. The pattern in previous QEs is pretty clear. When the banks hoard cash like in QE 2, it has little effect on the market. When the banks start lending that out in large amounts like QE 1, some of that makes its way into the market, largely through substitution.
So ordinarily, I would be telling you this is a tailwind for stocks if it keeps going in this direction. But this is happening:
So unless that $45 billion becomes $450 billion very quickly, I still think we are nearing the end of this historical disconnect between the economy and the market.
Billions and Trillions
Running down the factors from the Fed and Federal government:
This was a make-up week for QE (blue boxes). MBSs are hard to maintain week-to-week because of early pay-backs. Last week, the Fed’s MBS level was actually down $9 billion, but they made up for it this week by doubling up. Anyway, it’s at $25 billion a week, averaging the last two, the lowest level since last fall.
Repo (green) is at zero. That is not a typo or an approximation, and now down $242 billion since our March 11 start date. Its highest level was $442 billion on the March 18 balance sheet. Almost $2.4 trillion in QE since then has made it unnecessary.
Treasury (yellow) has been grabbing with both hands during all this, with net debt auctions exceeding spending by $1.2 trillion. That is saying something, because they are spending trillions. Every week I’ve been waiting for it to stop going up, and that was this week. They only let go of $32 billion this week and are still holding on to $1.6 trillion, the second highest of all time after last week. But maybe now that money will actually go to stimulating the economy rather than gathering dust at the Fed.
Finally, and most controversially, is the red box, loans and facilities. It seems very tiny for all that fuss. Let’s break that red box down.
The three to the left are the earliest, and they are all coming down week after week. The rest have either stalled or are growing very slowly. The secondary corporate bond facility, the locus of so much popular ire, is at $43 billion, about 17% of where the Fed said it would go.
This is all very small and of little consequence in the big picture.
But the best news for the economy is that the banks have stopped hoarding every dollar that the Fed has given them, and that Treasury hasn’t taken back.
This all peaked on the May 27 balance sheet. Reserves had risen by $1.5 trillion, Treasury had grabbed a trillion, but at that point the Fed had only pumped in $2.3 trillion. The banks were holding on to everything they got from the Fed, plus another $244 billion in deposits.
Now, all the way to the right, you can see the banks have lent out $45 billion more than is coming in since March 11, though some of that may be defaults. That’s $289 billion in net lending since the end of May. This is good! To be clear, the banks are still absurdly liquid right now and still don’t seem particularly interested in changing that.
Digging in on the shapes of those curves in that last chart, it tells the story of how this went over the last four months. Starting with the green column, which is the net of the first chart. It’s mostly QE minus Treasury’s account, but there is also the net change of -$242 billion in repo, and the addition of $211 billion in the loans and facilities.
But the big mover is that the Fed really front-loaded QE, getting up to $1.5 trillion in the first four weeks. At that point, it was outpacing Treasury. But the very rapid pace was done by the end of April, and Treasury kept on going, until this past week.
The blue curve, reserves, peaked at $3.3 trillion on that May 27 balance sheet, but are now down considerably, by $504 billion. But Treasury took $298 billion of that. The rest got loaned out, and maybe there are some defaults in there. These days, I’m all for small victories.
QE 1 Versus QE 2
One thing I like to stress is that all these things are different, and the results have more to do with what the banks do than what the Fed does. People like to ascribe magical powers to the Fed, but the banks and bond markets are much more powerful in aggregate.
The Fed only likes to counterparty banks and other central banks. During these extraordinary times, it has loosened up a bit on that, but still, of the nine facilities, seven are still counterpartied by banks or other financial corporations.
The point is that the big thing that happens in QE is that the banks get a lot of cash. And then they get to decide what to do with it. They can hoard it or they can put it to work.
In QE 1, the banks got a total of $1.2 trillion, mostly for their privately-bundled MBSs, which had caused the whole mess. This one leaked like a sieve.
Federal Reserve; Standard & Poor’s. To be clear, before 2015, Treasury kept the account much smaller, between $25 and $50 billion, so it has very little effect on that green line. I’m keeping it in for consistency’s sake. Chart © 2020 Trading Places Research
What you see here is that Fed piled on the cash, and banks held on to relatively little. By September 2009, the difference was $846 billion. About half of that got sucked into bad loans, mostly mortgages. You see what happened in the market in the same period.
QE 2 was very different. The Treasury component was bigger and faster, but there were no MBS purchases, so the whole affair was smaller by about a third.
This went very differently. The banks decided to hold on to every dime they got from the Fed, plus another $185 billion in deposits. As you can see, the S&P 500 performance in this period was much poorer, below the 16% annual average for the cycle, and far below the QE 1 gains. Moreover, all the gains were baked in by the time the banks decided to hold on to all that cash (dashed lines).
Again, the point is that each of these things are different. I don’t even bring up QE 3, because the sequester happened in the middle of it, and that was a big deflationary bomb that threw off the whole thing. In none of the first three was Treasury nearly as consequential as it is in the current round. They are all unique.
Juiced: How The Fed’s Money Gets Into The Market
A few weeks ago, I wrote an article called The Fed Is Not Juicing The Stock Market that many people clicked on just so they could heap some abuse on me in the comments. Fair enough.
But back a few weeks ago, it was clear where the Fed’s money was. It was in two places: in bank reserves and with Treasury. That is pretty much still true, except for that $45 billion, but what if that gets much larger like in QE 1?
About half of the banks’ bounty went to bad loan write-downs in Q1, but the rest made its way out into the economy as new loans, but mostly new debt securities – Treasuries and real AAA MBSs. So how does that get into the market?
The first way is directly through margin loans. This is pretty straightforward.
But the more important way is indirectly through substitution. For example, a family wants to put an addition on their house. They have plenty of money in the stock market, and could pull some out, but instead take out a HELOC at a low interest rate, and pay with that. The loan did not go directly into the market, but it substituted for large equity sales.
Now think of a small corporation with a mixed shelf. They need more funding and they have a choice of debt or equity. If they take the loan, that substitutes for diluting the equity of shareholders.
But we are still talking about marginal amounts relative to the grand scheme of things. We are probably talking about something in the range of $50-$100 billion in equities, out of a pool of about $10 trillion in non-financial corporate market cap back then, with the market at its low.
So the more important effect is psychological signaling:
- QE 1 set the expectation for the “Fed Put.” The Fed put a bottom on equities and drove a very fast initial rebound in March 2009, or so the thought goes. Keep doing the thing that worked until it stops working.
- Banks putting a lot of money to work in the economy and setting good terms for margin signals risk onto a variety of participants.
The Fed has leaned into that first bullet with its policy I like to call “Very Large Numbers In Press Releases.” It began with repo in early March. By the end of the month, it could have had $4 trillion in contracts out there, but never got close to that. The facilities were all announced with giant limits, but have come nowhere close. I think this is all by design.
The Fed has been remarkably consistent on two points since we began this journey:
- There is no bottom to this well.
- We can only stave off the worst – a huge wave of bankruptcies, foreclosures, bankruptcies and evictions. This is going to be terrible no matter what we do.
People have been focusing on that first bullet, but not the second. The Fed wanted to establish confidence in debt markets generally, because the alternative was a depression. The knock-on effect is people forgetting about the second bullet.
We just saw a terrible Q1 for corporate America. BEA’s broad corporate income and profits measures just came out for Q1. This covers only about three weeks of the pandemic, though there were supply chain issues all quarter coming from China.
First income, roughly equivalent to revenue, was off only 1% QoQ, but that was masking some deep drops in the splits
But the supply chain issues and then three weeks of pandemic destroyed profits, roughly equivalent to net income.
S&P 500 profits were down 67% QoQ in Q1, cratering from $35.53 in Q4 to $11.88 (S&P’s still preliminary estimate). Companies are about to report the June quarter, and it will be worse, containing as it does the three worst months. So far. At some point that has to start mattering.
Also starting to matter is this…
The Country Is On Fire
Most of the focus regarding the current surge in COVID cases has focused on the Sun Belt. Southern California, Arizona, Texas and Florida are leading the charge here, but every single state in the Sun Belt is seeing their case rates rise, mostly pretty quickly.
But the blue line, the cold-weather states, is now turning up. This is centered on the middle of the country:
Even the northeast and mid-Atlantic may be turning back up in the last couple of weeks.
The whole country is on fire. Everyone is focused on four states, because they are really terrible right now, but there is the potential for the whole country to become like that.
Israel provides a cautionary tale on how quickly the incredible progress of places like the northeast can turn around:
I highlighted May 13, the day Israeli schools reopened. That day the moving average of new cases was only 34. The case rate began rising about two weeks later. Yesterday the case rate was 1,111. It is re-closing much of the economy now. This is a cautionary tale.
The public health crisis causes the recession, not the shutdowns.
The Fed Is Not God, It Is A Bank
Under normal circumstances, my ears would be up right now. The Fed is handing out cash to the banks, and it is starting to open up the purse strings. Full steam ahead.
But this is not normal circumstances. Even without the bungled response, the recovery from this recession was going to be very slow and painful, and we have now made it worse. All the high frequency data is showing a double-dip beginning in mid-June. Just the latest data point I’ve seen is Chase’s (JPM) credit card spending tracker:
The Real-Time Population Survey, restaurant and retail data, and mobility data all point to the same thing. As cases rose in mid-June, people stayed in again, and slowed spending. And Chase noted that it was not just in the states where case rates were rising fast, it was all over the country.
Be prepared for a double-dip this summer. I had been predicting mid-August, but it may come sooner.
The Fed can only do what a bank can do which is provide liquidity. It cannot provide solvency. According to Bloomberg, we have already seen 112 COVID-related bankruptcies.
Retail, restaurants, energy and travel/leisure/entertainment are the big losers here. There is more coming in July and August.
I continue to believe that the closest historical analogy is 2000-2001. The big similarities:
- Valuations have exploded, including a large group of money-losing tech companies.
- There is a large corps of investors who sincerely believe the market will never go down again. This time, it’s different.
Spoiler: It’s never different.
For what it’s worth, here’s the comparison, starting on the day of the all-time high of the S&P 500.
See you at the bottom.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.