If ‘QE-Lite’ Reignites Classic QE Trade, Fade It, Goldman Says

Over the weekend, we recapped September’s epic short-term funding squeeze and made a few comments about what’s likely coming next from the Fed in light of this month’s dramatics.

The unwelcome chaos in the repo market gave short-end rates strategists a rare moment in the spotlight, compelled the Fed to respond with emergency liquidity injections and forced the financial punditry to don their money market expert hats and pretend to care about something they know is important but want absolutely nothing to do with (much like UK politics).

The timing left something to be desired. The worst of the crunch came on September 16 and 17, the two days ahead of the September FOMC decision. Given the acute nature of the squeeze, some wanted more from Jerome Powell and the Fed on September 18. Instead, the market got another IOER tweak and a promise to conduct “as need” liquidity ops to smooth things out into month- and quarter-end. Eventually, the Fed released a schedule for the overnight operations and conducted three term repos, some of which were upsized to meet demand. Now, the challenge is to lay the groundwork for balance sheet expansion, starting in November.

Read more: The Funding Storm Has Passed. Now What?

As the conversation continues around the Fed’s next move, Goldman is out with a handy Q&A that addresses recent events and makes some predictions about what’s coming up. The bank doesn’t deliver anything “new”, per se, but it’s a good piece and there are some passages that are well worth highlighting.

After reiterating that the Fed will need to roll last week’s trio of term ops when they expire next month, the bank writes that “we then expect the Fed to announce a resumption in balance sheet growth at the October 31 FOMC meeting to: 1) prevent the passive erosion of reserves due to growth of other non-reserve liabilities on the Fed’s balance sheet (mostly currency in circulation) and 2) rebuild a buffer for reserves, which we estimate to be roughly $200 billion”.

(Goldman)

The specifics of the forthcoming POMOs will depend on how quickly the Fed wants to rebuild the reserve buffer.

A Fed that wants to hurry it along could initially buy $60 billion/month for three months or $35 billion/month over six months, Goldman suggests. After that, we’d see “a more ‘steady state’ ~$10 billion/month of purchases… largely intended to offset the passive decline in reserves”. Obviously, all of that would be on top of the $20 billion/month the Fed buys to replace MBS paydowns.

The problem with the “hurry-up offense” approach (if you will) is that it has the potential to exacerbate the extent to which market participants would already be inclined to conceptualize of these new purchases as “QE”, as opposed to balance sheet expansion to relieve reserve scarcity. You’re reminded that this is one of the major challenges for the Fed going forward. As Nomura’s Charlie McElligott put it on several occasions this month, “the market’s ‘muscle memory’ in the post-GFC period has conditioned many participants into believing that ‘balance sheet expansion = QE'”.

“Front loading the reserve buffer build-up through large purchases may, however, be misinterpreted as QE”, Goldman cautions, on the way to suggesting that one way to mitigate the situation would be for the Fed to purchase $15 billion/month and continue to conduct “as needed” open market operations to ensure there are no further episodes of funding stress, even as that risks a kind of “Whac-a-mole” scenario, where repo chaos flares up and has to be tamped down the next day with O/N repos.

Goldman goes on to explicitly lay out the difference between the kind of balance sheet growth that’s almost guaranteed to be announced in some form at the October meeting and QE proper. To wit, from the bank’s note:

From an operational standpoint, the policies are the same, but the intentions (and probably the outcomes) are different. While the Fed would be purchasing assets in both cases, the primary purpose of POMOs is ensuring adequate liquidity in the system to maintain orderly market functioning, and compression of risk premia is only a side effect. Under more recent incarnations of QE, engineering an easing of financial conditions by compressing term premia is the primary objective, and injection of additional liquidity in the system is a side effect.

Despite the Fed’s best efforts to communicate the nuance, it’s still entirely possible that the market’s “muscle memory” will be too strong to resist. In other words, try as they might to convince everyone that this “QE” isn’t really “QE”, the Pavlovian response may show up regardless.

“The immediate market reaction could be to re-initiate the ‘QE trade’ (long risk assets, long bonds, short the Dollar) irrespective of Fed messaging, particularly if the Fed were to decide to re-up its reserve buffer more aggressively, with say $80 billion of monthly UST purchases over a short period ($60 billion net of MBS run-off)”, Goldman writes. “We would fade such a move”, the bank adds.

As far as the standing repo facility goes, Goldman doesn’t see that as particularly likely given the urgency of the current situation and the time it would take to get things “right”.

“We have long viewed the standing repo facility as an elegant construct that obviates the need for the Fed to gauge the reserve needs for the system [but] there are several operational and design parameters that will take time to calibrate”, the bank says. “The need for reserves to reduce front end rate volatility is more immediate [and] as a result, we view POMOs (perhaps combined with some TOMOs) as more likely”.