Zoltan Pozsar On Money Market ‘Singularity’ And War Finance

If you’re concerned the market will choke on the coming deluge of Treasury bills thereby putting upward pressure on global dollar funding rates, you should probably note that Zoltan Pozsar does not share your concerns.

And, as many market participants are aware, if what’s under discussion is money markets, and your view doesn’t conform to Pozsar’s, that divergence of opinion is tantamount to you being wrong.

In his latest note, out Thursday, the man who’s been variously described as the “oracle” of funding markets (or the “spider in the middle of the web”), builds on a foundation laid in his last two pieces, in the course of explaining why the market will digest $1.25 trillion in T-bill supply over the next two months with relative alacrity.

Zoltan begins with a flourish, as he’s wont to do. “In astrophysics, singularity refers to infinite density at the center of a black hole, and the earliest state of the universe — before the Big Bang”, he writes. “In money markets, we define singularity as an ‘infinite’ tightness of spreads at the zero bound”.

The money market “singularity” is “nigh”, Pozsar declares, and “its gravitational pull can drive the US dollar Libor-OIS spread tighter from its current level, to the 10-20 bps range by the end of June”.

One can, of course, delve as deeply into Pozsar’s work as one desires. It all depends on how many hours you’ve blocked off on your daily calendar for consideration of market plumbing dynamics and the issues currently being debated by those who study and critique the performance of the plumbers.

The thrust of Pozsar’s latest is actually pretty straightforward as far as Zoltan goes.

“While $1.25 trillion of Treasury bills over the next two months sounds like a lot, it’s actually not that much relative to the scale of the Fed’s liquidity injections, and the availability of a standing repo facility and dollar swap lines at low rates”, he says, on the first page of what, if you include Zoltan’s trademark “primary-colors-on-blinding-white” charts, is a 14-page assessment.

He starts with a Libor-OIS “morbidity review”, which revolves around a discussion of the three private funding channels foreign banks turned to when prime money market funds began hemorrhaging during the worst days of the crisis. Prime funds, you’re reminded, bled nearly $150 billion during the final weeks of March.

If you followed the evolution of the crisis and the Fed’s response, it’s easy enough to follow Poszar’s latest.

You’ll need to recall the chronology of what I’ve characterized as the “Whac-a-Mole” phase, during which Jerome Powell moved swiftly to unfreeze the commercial paper market, roll out a primary dealer credit facility, unveil a backstop for money market funds (aimed at addressing criticism that the PD facility wasn’t sufficient to prevent an exodus from prime funds) and enhance/ expand swap lines.

Here’s Poszar on how foreign banks “toggled between three private funding channels” over that tumultuous period:

First, during the last two weeks of March, they relied on U.S. primary dealers to issue short-term U.S. dollar unsecured debt (CD/CP), while prime funds struggled with outflows. Second, during the first half of April, they raised local currency and swapped it for dollars, enabled by the rapid normalization of the FX swap market thanks to the dollar swap lines. Third, from mid-April, they started to issue CD/CP in volume to prime money market funds once again, using rapidly falling FX swap implied funding rates to lock in low CD/CP rates.

As Zoltan notes, “these three funding phases correspond to the rapid widening, the inflection point and the equally rapid tightening of the U.S. dollar Libor-OIS spread during the past two months”.

After an in depth (and I do mean in depth) treatment of each phase, Poszar reiterates that “the main theme of phase three is the return of liquidity to the U.S. CD/CP market and foreign banks using the ongoing fall in FX swap implied costs of U.S. dollar funding to extract low rates on CD/CP issued to prime money market funds”.

The two figures below illustrate the state of dollar funding markets during the worst of the crisis versus today.

(Credit Suisse)

Unsecured spreads are now low, and if one assumes inflows into prime funds persist, they’ll go lower, due to a combination of the altered “laws of physics” given the proximity of the ZLB and, less esoteric, due simply to the supply/demand dynamics created when prime funds are seeing inflows and foreign banks are raising via swap lines. To wit, from Zoltan:

Low yields on U.S. Treasury bills will continue to drag CD/CP rates lower, as at the zero bound the “laws of physics” change… …anything with a substantially non-zero yield has strong demand for it. Strong demand from prime funds for CD/CP also coincides with a decline in the structural supply of CD/CP — think of the fact that foreign banks have raised $450 billion through the swap lines, and some of that funding naturally reduces the amount of CD/CP foreign banks need to issue. Strong inflows into prime funds driving demand for U.S. dollar CD/CP, combined with the decline in the structural supply of U.S. dollar CD/CP as foreign banks use the swap lines points to a perfect storm where U.S. dollar Libor-OIS spreads can compress further still.

He then moves to consider how much scope there is for further spread compression.

The actual numbers are likely to be less interesting for the casual reader than the overarching message, and with that in mind, I’m going to skip over what I’m sure Poszar would say are a lot of crucial details. The thrust of the message conveyed in the remainder of the note is captured in this line from the beginning of the second section:

Money markets are like a cake… …most of the time, but they can be like pancakes too when we’re at the zero bound and we are using the tools of war finance.

War finance is a reference to… well, to war finance in the context of the virus, but it’s simultaneously an allusion to Poszar’s April 14 note, documented at some length in this pages.

Financing the “war” against the “invisible enemy” entails issuance and as noted, Treasury will issue some $1.25 trillion in bills by the end of next month. The question, Zoltan writes, is whether that flood of supply will “pressure bill yields and hence Libor-OIS wider, and ruin the pancake party”. (And, yes, he says “pancake party” in the original.)

The answer is “maybe, maybe not”, with a lean towards the latter, for a variety of reasons, not the least of which is that the Fed won’t allow it, because it would amount to countenancing an outcome that is starkly at odds with what policymakers have worked so hard to achieve lately. To wit, from Poszar:

In the extreme, the Fed has the option of buying bills or to cap bill yields at OIS or thereabouts. At a “macro” funding level, it would not make any sense for the Fed to launch all these new liquidity facilities to bring rates down to zero, and then watch passively as bill supply pushes rates away from zero. The Fed shouldn’t, and we think won’t let that happen. It will buy bills if it has to.

But it won’t have to – or at least probably not. There are four sources of liquidity that should absorb the supply. They are (truncated and edited for brevity):

1. The Fed is still buying $35 billion of U.S. Treasuries a week, which will add around $300 billion of reserves by the end of June. Most of these funds will end up with government money market funds, which will recycle them into bills or repos [so] either government funds that will buy $400 billion in bills directly or dealers and RV hedge funds that fund via repos.
2. Foreign central banks still park $300 billion at the Fed’s foreign repo pool, earning… zero rate of interest. While this pool of money has been remarkably sticky, some of it could move into the bill market to earn some positive interest -let’s assume $100 billion.
3. Primary dealers were just given more balance sheet through the exemption of reserves and U.S. Treasuries from the calculation of the SLR. This rule change can easily provide $200 billion of balance sheet for running bigger U.S. Treasury inventories.
4. Asset managers that lend in the FX swap market could start buying more U.S. bills if their yield exceeds the FX swap implied yield of holding German, French and U.K bills. If that will stress the FX swap market, the swap lines will soothe it at OIS+25 bps. Let’s assume $100 billion of bill demand from this source.

So that’s $800 billion. Throw in the ~$400 billion difference between Treasury’s target cash balance at the Fed for end-June and the level right now (figure below), and you end up with $1.2 trillion or, as Poszar puts it, “voilà, the financial system absorbed $1.25 trillion of bills without a penny purchased by the Fed”.

Prior to running through those four liquidity pools, Poszar makes a crucial point about the Fed and its tolerance for bill yields. This is well worth a lengthy block quote (and the three minutes extra it will take you to read it). To wit, from Zoltan:

As far as core funding markets like U.S. Treasury repos and FX swaps are concerned, the Fed is demonstrably intent on providing backstops at extremely generous rates. The price of the U.S. dollar swap lines at OIS+25 bps is perhaps the best example of that, as it suggests that the Fed doesn’t want the cost offshore dollar funding above OIS+25 bps. Given that the offshore dollar funding market — the FX swap market — is the “outer rim” of funding markets, if the outer rim is backstopped at OIS+25 bps, all other funding markets should trade below that rate too: repos for sure and if prime money fund inflows continue, unsecured funding markets too. Furthermore, the Fed’s willingness to backstop banks’ and dealers’ term funding rates at 25 bps flat through the discount window and the PDCF suggests that the Fed doesn’t want to see the cost of onshore funding north of 25 bps. But then if onshore and offshore U.S. dollar funding markets, and U.S. and foreign banks and dealers are backstopped at 25 bps flat and OIS+25 bps, respectively, why exactly does the market worry about the funding needs of the U.S. Department of the Treasury? Given the above construct, we’d be surprised to see U.S. Treasury bill yields above banks’ backstop funding rate, which means that bill yields can go a maximum of 10 bps higher from here before the Fed decides to cap them at OIS or OIS + some small spread.

Part and parcel of war finance, Poszar says, is making absolutely sure the supply of collateral doesn’t overwhelm the supply of reserves – being keen to create and facilitate an environment where there’s sufficient balance sheet to accommodate the ballooning credit needs of the US government.

If, for whatever reason, bill supply needs another sponge, the Fed can go the “nuclear” route, he writes, where that simply means “outright purchases of bills and capping bill yields”.

And really, who cares? Or, as Zoltan puts it, “what’s the big deal about bill purchases?”

Consider one final passage in that regard:

U.S. Treasury bills are the easiest asset for the Fed to buy — it’s the Fed’s natural habitat (and for money-fundamentalists, it should be the Fed’s only habitat). If bills are in excess, and this excess were to push global dollar funding rates up and away from the zero bound, the Fed will simply buy the excess. If the Fed is backstopping the credit market, why wouldn’t it backstop its natural habitat?

Good question.

---