Since October 2022 we have seen markets rally hard in the face of what was almost unanimous agreement amongst market participants that a recession was ‘just around the corner’ due to a US dollar that would not stop rallying and out of control energy prices. Many participants were left sidelined these last 9 months asking themselves, “How could this rally possibly be happening?!” The answer (retrospectively) is simply liquidity.
Back in the heady days of 2020 and 2021, you could not help but hear talking heads on financial TV or threadooors on Twitter espousing the age-old adage “Don’t fight the Fed”. The truth is that if you were short in 2020 or 2021 you were not only fighting the Fed, you were fighting the Bank of Japan, The European Central Bank, and The Bank of China too. As such it shouldn’t surprise anyone that in an environment like that the liquidity found its way into EVERYWHERE.
Dog coins, electric vehicles that only work when pushed downhill, bankrupt car rental companies, struggling game retailers, and movie theatres with a questionable path forward in an age of streaming. All of these ‘investments’ found themselves buoyed to unimaginable heights in the mania that stemmed from every central bank on the planet simultaneously throwing caution to the wind and YOLOing balance sheet like there was no tomorrow.
I’m sure you are saying to yourself, “Gee thanks, Captain Hindsight, how does this help me now?” Well, dear reader, let’s put together a framework for thinking about liquidity in the year 2023 and beyond in hopes that we can make better decisions on when and where to deploy our finite resources in the future.
“All models are wrong, but some are useful”
First, we need to clarify what we mean by liquidity. There are many metrics that have been used over the years to define liquidity in the system and as monetary policy and the financial system have changed over the years they are prone to becoming less useful over time. M0, M1, M2, and M3 are all measurements of liquidity that have been used for various purposes to varying degrees of success over the years. At the end of the day, we are looking for a model of liquidity that can be useful in predicting the impact on asset prices, because if your model doesn’t help you make financial decisions then it isn’t very useful, is it?
In 2020 it became quite trendy to measure the movements of assets relative to the size of the Federal Reserve’s balance sheet. From a broad strokes directional perspective, this was (and still is to some degree) somewhat predictive of asset price movements.
So, Liquidity = Fed Balance Sheet
Nice and simple, right? The problem is that once the Treasury General Account (the US Treasury’s bank account at the Fed) ballooned from $430B to $1.8T those watchers of liquidity realized that the amount of liquidity in the TGA was no longer a rounding error against the then ~$7B on the Fed’s balance sheet. So it became necessary to recognize that of the $7B in liquidity, $1.8B of it was sitting in a bank account at the Fed ready to be spent down at any time.
Alright then. Liquidity = Fed Balance Sheet - TGA Balance
Sure, that worked pretty well… for about 6 months. Then, from January 21’ to June 21’ the Reverse Repurchase Agreement Facility (RRP) ballooned from basically nothing to over $1T. The RRP is one of myriad 3 and 4-letter acronym facilities at the Federal Reserve which allow firms to source or sink liquidity.
The RRP specifically can be thought of as a checking account at the Fed that pays an overnight interest rate (currently 5.05%). Or for the more crypto-native types you can think of it as staking rewards for staking your dollars at the Fed. This is the main interest rate that people are talking about when they are talking about the Fed raising rates. If the Fed kept all the other interest rates high and dropped this one there is no way they would be able to keep short-term rates where they want them. But as it stands there are nearly $2T dollars still sitting idly in the RRP. Hardly a sum that can be ignored.
Ok, so where does that leave us?
Liquidity = Fed Balance Sheet - TGA Balance - RRP Balance
Well. Yes, but… there is one major problem that needs to be addressed and it has to do with something that I was tweeting about back in October.
For the first time in 100 years, the Fed began operating at a loss in the Fall of 22’. Now normally, any profits made by the Fed just end up getting returned to the member banks and a portion is sent back to the Treasury. The amounts sent to the Treasury ranged between $55B and $109B in the time period between 2012 and 2021 (see graphic below).
So in the big scheme of things, this is a relatively small sum on the scale of macro monetary flows. However, once the Federal Reserve started operating at a loss. It had to start carrying those losses as a deferred asset on its balance sheet and you can see the cumulative impact of those losses here on the liability side of the balance sheet in the Fed’s H.4.1 release.
You may be asking yourself, ‘Why did I bother explaining this esoteric detail and what does it have to do with liquidity’? In essence, the Fed is currently paying out 5.05% on $2T (roughly $100B per year at that rate) to keep short-term interest rates elevated using the RRP facility we touched on earlier. But the problem is that they are now paying for RRP using ‘future profits’. Another way to look at this is the Fed is printing dollars and filling the Treasury’s coffers with IOUs.
If you want to watch the avalanche of IOUs pile up in real time you can do so using this link or you can use the RESPPLLOPNWW time series on Trading View.
All of this brings us to…
Liquidity = Fed Balance Sheet - TGA Balance - RRP Balance - Operating Losses
Is this a perfect model that will guarantee you riches forevermore? Definitely not. Is this model a useful means of viewing asset prices relative to tides of monetary policy? I would argue yes. With one major caveat, which is also the reason I went to the effort of giving a detailed history of how thinking about liquidity has changed in the last several years.
The financial system is a living breathing organism and the monetary policy levers of choice of the Fed have changed dramatically over the last 2 decades. What works this month as a model of liquidity may not work next month if the major players in the injection of liquidity start finding new places to insert it into the system. The key to keeping this metric up to date will be to monitor for large flows of liquidity, I’d say flows on the order of $50B+ that are also measurable with some regularly posted dataset would be worthy of consideration.
Another important note is that this metric that we have developed here is a measure of DOMESTIC US LIQUIDITY. We haven’t even gotten into measuring global liquidity (and we won’t be today, but perhaps in a future article).
Now that we have laid the groundwork on the key contributors to domestic US liquidity, let’s briefly touch on what the probable glide paths are for each of these contributions.
Fed Balance Sheet
The Fed has signalled that its plan going forward is to reduce its balance sheet by $95 Billion per month. The primary mechanism by that the Fed reduces its balance sheet is that when an asset (such as a US treasury) on the balance sheet reaches maturity the Fed just takes the money it receives and essentially pays off the debt it created when it made the money out of thin air to buy it in the first place. Because there are constantly new assets on the Fed balance sheet maturing every month the Fed has quite a bit of control over the dollar amount of assets they wish to ‘roll off’ the balance sheet vs. how much they will reinvest into new securities. The Fed’s balance sheet roll-off was going off without a hitch until March 23’ when the regional banking crisis forced the Fed to use their balance sheet to stabilize the situation with the creation of the Bank Term Funding Program (BTFP).
By the end of March, the Fed Balance Sheet had resumed its downward trajectory and we are currently sitting around the level at which we were back in March when the crisis happened. We can expect that this trend will continue until something else in the financial system breaks.
Any time that the Fed is attempting to reduce the size of its balance sheet, they claim they are going to monitor the system and stop draining once the system reaches an ‘adequate level of bank reserves’. The problem with this strategy is that they can’t tell you what ‘adequate’ is (many have asked). What they have shown they can do, is drain until they hit ‘inadequate’ and then stop and reverse, or in the case of the recent banking crisis, make a new facility to address what was ‘broken’ and proceed as scheduled. In other words, the strategy went from ‘hike till you break something’ to ‘hike till you break something you can’t fix’.
Treasury General Account (TGA), Reverse Repo Facility (RRP), and Operating Losses
I could probably write an entire article covering only the refilling of the TGA but I will keep this to just the important bits. On June 5th, 2023, Congress passed the ‘Fiscal Responsibility Act’ which officially raised the debt ceiling for the US Government. Prior to the passing of that bill, the US Treasury was in danger of running out of funds to pay for the continued operation of the government.
There were of course many legal avenues that were left unexplored such as ‘minting the coin’ and just ignoring the debt ceiling but we won’t get into those. The important part of this story was that in the months leading up to June 5th, the TGA had dwindled down to less than $40 billion. Then once the bill had been passed, Janet Yellen was once again able to load up the Federal coffers. This means issuing US Treasuries and in return getting those sweet sweet Federal Reserve Dollar Liabilities (read: dollars).
Alright now pay attention! This is the part that investors need to be aware of going forward as we proceed into the second half of 2023. Over the course of 3Q2023, the Treasury is scheduled to issue another $671 Billion in US Treasuries. When this happens, the money to buy those Treasuries has to come from somewhere. But where it comes from means everything from a liquidity perspective. Let’s take a quick gander back at the equation from before.
Liquidity = Fed Balance Sheet - TGA Balance - RRP Balance - Operating Losses
Now if for example the TGA sucks in $671 Billion and all of that money actually comes from the RRP, what does that do to liquidity?
Example 1: TGA drains RRP
Liquidity = Fed Balance Sheet - (TGA Balance+ $671B) - (RRP Balance - $671B )- Operating Losses
As you can see the $671B would just net out entirely leaving liquidity right where it was before. In fact, likely better, because the Fed’s operating losses are largely being driven by paying out 5% on RRP balances, the liquidity situation would actually IMPROVE if the UST sales were entirely funded by way of the reverse repo facility.
Ok so let’s look at the other extreme, let’s imagine that the RRP stays right at the $1.9T level that it is sitting right now. In this scenario, all of the money to purchase the Treasury’s wares would have to be pulled from available liquidity. Let’s go back to our equation.
Example 2: RRP remains unchanged
Liquidity (-$671B) = Fed Balance Sheet - (TGA Balance+ $671B) - RRP Balance - Operating Losses
In order to net out the $671B being used to fill up the TGA, that money would necessarily have to come from our liquidity metric since everything else is remaining unchanged in this example.
So which of these two will it be? The truth is no one knows for sure how this plays out, myself included, so we have to just keep an eye on it as things unfold.
The likely scenario is that you see some combination of the two. As you can see in the chart below, the TGA has already been refilled to the tune of ~$350B while we have seen a corresponding drop-off in the RRP that accounts for most of that drop-off. If we continue to see RRP drop off as more USTs are issued then liquidity should not be too greatly impacted.
It is also incredibly important to recognize that at the end of the day, we are measuring only a small sample of the total number of dollar liabilities in the world economy. So while this sample may currently be representative of the movement of dollar liquidity tides, there are certainly issues that could put this tool’s predictive power in jeopardy. With any macro trading strategy, it is important to ask yourself what important signposts you should look for along the way to either corroborate or invalidate your thesis.
A good mental model for understanding liquidity and how it impacts prices is a bucket with holes in it.
Pouring in from the top you have the Fed and the Treasury making new money (you also have private credit and other forms of private money which are not captured by the liquidity model). The problem is that the bucket has many holes. These are the deflationary pressures. The greater the deflationary pressure the bigger the hole and the faster the bucket will be drained when the liquidity pouring in slows to a trickle.
Holes could be things like:
- A Commercial Real Estate deal that went bad and now the lender is stuck with a mountain of bad debt and an illiquid piece of property.
- Auto lenders that underwrote loans on new and used cars at exorbitant prices caused by the supply chain issues in 2021 and 2022 and now the debtors are underwater and can’t or won’t pay.
- Venture Capital firms marking to market illiquid assets they regretfully bought at the peak of the 2021 FOMO.
- Workers paying down student loan debt after a 3-year hiatus.
There are quite a few deflationary pressures that are growing as we get deeper into 2023 and deeper into this rate hike cycle. As companies that financed debt at near-zero interest rates are forced to refinance at higher rates, many will succumb to the destructive part of ‘creative destruction’. I expect zombie companies to be slayed. That is IF the Fed is able to stick to their goals of ‘higher for longer’.
What could end up preventing the Fed from meeting that goal is a swift and powerful deleveraging and deflation event. These events have happened before and when given the choice between letting asset prices find price discovery or support them (and the institutions under their regulatory purview that would be inevitably rekt in the resulting deleveraging), somehow the Fed always chooses to swoop in and save the day by being a buyer of last resort.
As we move forward into the second half of 2023, we need to keep a vigilant eye on the liquidity in the system. But to get the true picture of whether the tides are going in or out, and thus asset prices will move with us or against us, we need to make sure that there is at least as much pouring in the bucket as is leaking out the holes.