I used to expend quite a bit of energy (spilling copious amounts of digital ink in the process) describing the relationship between dollar liquidity and global trade.
Then it became readily apparent that nobody cared, despite the concept being both critically important and eminently easy to grasp. Better to stick with comparative arcana. The audience loves to be mesmerized with indecipherable desk chatter, but cares little for straightforward macro topics.
I doubt seriously whether the preferences of readers have changed since mid-2018, which was around the time I penned dozens of dollar liquidity articles, with global trade and developments at Steve Mnuchin’s Treasury as the backdrop.
Nevertheless, I want to touch on this again, because that interplay — between dollar liquidity, trade and Treasury — is front and center.
I’ve taken to quoting Nordea’s Andreas Steno Larsen recently. He consistently puts out work that’s compelling and easily digestible, no small feat when it comes to the subject matter at hand.
Steno Larsen says “it has been a hot topic on FinTwit… whether the Fed has printed enough Dollars to send the USD into a bear-market spiral”. I’ll take his word for it on that. I do, of course, have a Twitter presence, but truth be told, I can’t stand the place. “FinTwit” is a nightmarish echo chamber, where the population seems to skew heavily towards would-be Austrian “economists” who have a disconcerting propensity to traffic in academically dubious “theories”, and that’s when they’re not retweeting propaganda and bad memes. The rest of FinTwit appears to me to be a mix of real and imagined buysiders and Bloomberg reporters.
Anyway, let’s just assume Steno Larsen is correct and that lots of folks are talking about digitally-conjured dollars and the prospect for currency debasement. Whether or not Twitter is talking about it, I can say, definitively, that traders and analysts are debating whether the twin deficit “problem” and the fact that the Fed’s aggressive actions have at least facially narrowed the policy divergence with NIRP economies, together have the potential to undercut the greenback. Layered atop this discussion (or maybe “undergirding it” is more apt) are questions about Treasury’s massive cash balance.
Steno Larsen manages to capture a number of critical points in very few words. That’s admirable. If there’s one writing skill I do not posses, it’s brevity.
“The increase of digital USDs in circulation in the commercial system was nothing but bizarre in March and April but the USD creation from Fed has slowed markedly already”, he writes, in a note out Sunday evening. “USD liquidity has tightened as USD swaps rolling off foreign CBs balance sheets paired with a build-up of the cash-buffer has removed USDs from the system”, he goes on to say, noting that Mnuchin “holds a potential USD liquidity bazooka if he and the US Treasury decide to bring down the TGA”.
The TGA discussion is important.
Treasury’s cash balance is now some $1.6 trillion, and the timing on when it will be brought down has been the subject of a vociferous debate among those who spend their days steeped in the funding markets.
“Treasury has increased spending this year, but not near the amount most expected”, Morgan Stanley wrote last week, in a long piece that looked at when this money will be disbursed and where it will go.
For Jerome Powell, this is somewhat problematic (or, if that’s too strong, call it “vexing”). In terms of liquidity, it effectively works at cross purposes with the Fed’s efforts, and could lead to a funding crunch. Treasury had said it planned to run down its cash balance to $800 billion in relatively short order.
The discrepancy is related to amounts earmarked for first-loss buffers in the Fed’s various emergency facilities and write-offs of what amount to grants for the Paycheck Protection Program (PPP). As a reminder, Treasury budgeted for $680 billion of SBA loans to be forgiven.
“We identified two sources of unspent, but budgeted for, expenses: capital pledged to the Fed facilities and SBA loans”, Morgan Stanley says. “In particular, $111 billion and $587 billion could be spent for each of these programs, respectively, totaling $698 billion”.
The credit protection for the Fed facilities is self-explanatory (the pledged capital reduces the TGA). “The half-trillion dollar question” (as Morgan calls it) is “When will SBA loans get forgiven?” Here’s a quick explainer from the bank:
The SBA has already allocated $513 billion of the $680 billion initially budgeted for the SBA PPP program as of June 16,2020, so how can the Treasury still outlay an additional $587 billion in association with the program? Furthermore, how have TGA outlays to the SBA only amounted to $93 billion when $513 billion has been allocated? As shown in Exhibit 9, Treasury does not disburse money until after the loan recipient has met the criteria for loan forgiveness and the loan essentially becomes a grant. At that point, the bank can receive payment from Treasury for the full amount of the loan. Banks can also receive payment from the Treasury for any loans that are defaulted on, as all PPP loans are 100% guaranteed by the SBA. This payment from Treasury (as denoted by the teal color) is the only time the TGA is impacted.
The timing of these payments (or, equivalently, “SBA disbursements”/ “TGA withdrawals”) is uncertain, as is the total amount, for a laundry list of reasons associated with the vagaries of the PPP.
In addition, any further fiscal stimulus out of Washington will obviously have implications for the TGA, which muddies the waters even more.
So, with that (necessary) tangent, allow me to steer this back on the dollar liquidity track. Here is Morgan explaining why this matters:
In general, though the impact on the real economy should be limited, when the TGA declines, it should be beneficial for funding markets with one caveat. When the TGA declines due to increased capital pledges to the Fed, this will not impact the economy or funding markets, because that money merely becomes “equity” for the Fed to later be remunerated to Treasury. In this case, a decline in TGA is met with an increase in Fed capital. However, when the Treasury repays the principal amount on forgiven SBA loans to the loan originators, this will directly increase reserves, as the decline in TGA (one liability on the Fed’s balance sheet) is offset by an equivalent increase to reserves (another liability on the Fed’s balance sheet). Though reserves have already swelled to above $3.1 trillion, they have begun falling slightly from their peak of $3.3 trillion the week of May 20,2020. This decline is likely due to the fact that TGA continues to grow while Fed purchases have slowed – as a result, reserves have fallen on the margin. Thus, a future decline in TGA would cause an increase in reserves. This, in turn, would be supportive of funding markets.
Again, this situation is the very definition of “fluid”, and it injects quite a bit of uncertainty into the front end. BMO’s Jon Hill recently called it the “gorilla in the room”.
For his part, BofA’s Mark Cabana says that with bill rates low, “Treasury may see its TGA balance as ‘free money’”. He sees the TGA staying elevated through mid-July, peaking as high as $1.9 trillion, before it starts to decline towards Treasury’s $800 billion target in September.
Looking out to the second half, Nordea writes that “USD liquidity will likely remain tight from a momentum perspective unless Mnuchin unleashes his liquidity-zooka during H2”.
“With $600 billion of planned issuance in Q3, the Fed will continue to fall short of ‘buying it all’, which could lead to less benign USD funding conditions compared to recent weeks”, Steno Larsen goes on to write, adding that “usually the USD only weakens if the Fed comes close to ‘out-printing’ the US Treasury”.
My buddy Kevin Muir (formerly head of equity derivatives at RBC Dominion, and better known for his exploits as “The Macro Tourist”) framed the discussion in somewhat Machiavellian terms in a recent note. To wit, from Kevin:
Maybe by increasing the TGA balance during the economic downturn, this has forced Powell to offset the liquidity withdrawal with more QE. It’s not a large leap to deduce that if TGA hadn’t been increased by $1.5 trillion, then Powell might not have needed to engage in as much QE.
Coming full circle to the discussion on USD liquidity and trade, Nordea reminds you that “while the Fed is the direct source of USD liquidity, world trade is the main source of USD liquidity velocity”.
That is crucial. The dollar generally weakens when world trade is robust, and vice versa.
“The running trade deficit of the US is a material source of USD liquidity for foreigners, [which is] why USD liquidity becomes scarce abroad when trade flows are subdued”, Steno Larsen went on to say Sunday. “This is the exact reason why the USD always gains during a US recession as the lack of imports lead USD liquidity to dry out outside of the US”.
So, it’s possible that the dollar continues to get pulled in both directions. Previously, depreciation pressure from the Fed’s printing spree was offset by subdued trade flows due to the pandemic (and in March by the acute panic to raise USD cash). Going forward, improving trade flows (as the global economy recovers) could exert downward pressure on the greenback, even as the Fed eases off the gas pedal. Here’s Nordea on more time:
If i) the Fed keeps the printing press running, ii) Mnuchin reduces the issuance pace as too much debt has been frontloaded (the US Treasury could just decide to bring down the TGA) and iii) world trade rebounds markedly, then we have the perfect USD weakening cocktail in place. For now, only assumption i) holds fully true.
And then, one more time from Morgan:
We think the modest, but not complete, drawdown of the TGA cash balance will weigh on USD, amplifying the already-present USD-negative global trends. The buildup in TGA in anticipation of future spending has been a drain on narrow liquidity, partially counteracting the Fed’s increase in excess reserves as it has purchased assets through its permanent LSAPs and provided liquidity through its temporary programs. The expansion in excess reserves has an important (albeit indirect) transmission mechanism to USD weakness by supporting liquidity conditions and risk sentiment. Thus, if excess reserves are likely to rise over the coming months, we think that is likely to aid USD weakness. Excess reserve growth is likely to come both from the increase in the Fed’s balance sheet (through its LSAPs) but also through the gradual reduction in TGA. The increase in TGA has almost completely counteracted the increase in the Fed’s balance sheet in the past couple of months – such that total reserves in the system are almost unchanged since April.
To close on a humorous (or not, depending on how you want to look at things) note, Muir wrote last week that “if one were truly prone to conspiracies, you might even argue that the TGA has been increased so that it can be strategically released to coincide with Trump’s re-election”.
Yes, one might indeed argue that.