Lead — The desk's thesis centers on the implications of recent developments in money markets for the Federal Reserve's quantitative tightening (QT) process, particularly as it relates to the upcoming November refunding. Per the full note from J.P. Morgan, the current dynamics in funding markets suggest that the Fed may need to reassess its QT strategy to maintain liquidity. This is underscored by the recent uptick in short-term rates, which could impact the overall effectiveness of the Fed's tightening measures.
What the desk is arguing
J.P. Morgan's latest insights delve into the current state of money markets, emphasizing that recent trends could significantly impact the Federal Reserve's QT approach. The strategists highlight the critical role that upcoming refundings play in shaping liquidity conditions, which in turn influence the interest rate landscape.
The analysis suggests that if the Fed remains committed to its QT path while navigating these funding nuances, it could lead to tighter financing conditions. This would ultimately challenge the central bank's goal of ensuring smooth market operations, a potential outcome that the desk implies markets may not be fully pricing in yet.
01Money market developments may pressurize the Fed's QT process.
02November refunding could reshape liquidity expectations.
03Strategic insights reflect nuanced survival of smoother financing.
Market implications
The evolving dynamics in money markets, particularly in light of the November refunding, could lead to a recalibration of interest rate expectations. If the Fed's QT process encounters greater friction from these market pressures, we might see a shift in investor sentiment and positioning.
Risks to this view
Potential risks include a more aggressive tightening cycle by the Fed if liquidity issues arise, which could spook markets. Additionally, a failure to manage these refundings effectively might disrupt financial conditions, amplifying volatility in both fixed income and currency markets.
You're listening to At Any Rate, J.P. Morgan's global research podcast, where we take a look at the story behind some of the biggest trends and themes in fixed income, currency and commodity markets today. I'm your host, Jay Barry, head of global rate strategy at J.P.
Morgan. We're now in week four of the U.S. government shutdown and seemingly headed for a new record and thus remain in somewhat of a data vacuum here, and as a result, the delivered volatility in U.S. rates remains pretty low. Nevertheless, we've got a lot to talk about this week, and today we're going to cover kind of a barbell of topics, first on recent developments in the repo markets and how it could influence the Fed's balance sheet normalization, its QT process, and then second, we're going to turn to Treasury supply as the November refunding announcement's now about a week and a half away.
To dig into these topics, I'm joined by our two experts on these subject matters, Teresa Ho, who runs our short duration strategy effort, as well as Phoebe White, who is senior U.S. rates and inflation strategist at J.P. Morgan. Teresa and Phoebe, welcome to the podcast.
If I can, Teresa, first turn to you. There's been a lot that's happened here in the last couple of weeks. Just three weeks ago, September quarter end proved to be a non-event, but since then, the Fed funds rate, after being stuck at IORB less seven for really the better part of the last number of years, has quickly risen to IORB less four, and the moves in the repo market have been even larger.
Just looking at this chart right now, the Treasury general collateral rate, the TGCR, has on average set about a basis point above IORB for the last month, which we really haven't seen on a consistent basis since the aftermath of the supposed ripocalypse back in the fall of 2019. And even a lot has happened since you were last on this podcast channel last week. So what's going on at the front end right now?
Sure. And thanks for having me on this podcast. And so you're right.
Funding pressures have been gradually increasing for some time now, and that was to be expected. We knew that was coming because when we take a step back and think about how we got here, there's been a confluence of factors that have been unfolding and taking place over the past few months. One is that we have a Fed that has restarted a new easing cycle after a pause earlier this year, and this came on the heels of several weeks of weak labor market prints and dubbish commentary from Powell at Jackson Hole.
Two is a rebuilding of TGA, which we knew was going to come in the months of July and August and resulted in a surge in TBO supply by as much as $600 billion or so. So that provided some sort of background in terms of draining reserves from the system as cash moves from reserves to TGA. Three is QT, which was continuing to drain liquidity from the system and still is right now.
And I guess this backdrop, two important things happened. One is that money funds began extending duration, extending their WAMs. So they were deploying new money into T-bills or rotating existing repo balances into T-bills, which then, too, relatedly meant also that balances of the overnight RRP were generally down to zero.
And this was important because then overnight RRP no longer served as a shock absorber. So sure enough, in September, we started seeing evidence of funding pressures, but those pressures were still largely contained and temporary. Then we moved into October, and more specifically over the past two weeks, something changed.
Something felt different. The middle of the month is usually when we have mid-month treasury coupon settlements, and those are fairly routine. We know it adds to funding pressures, but the pressures those settlements ended up adding was much more than anticipated.
SOFR ended up printing outside the Fed Funds target range for two days, and TGCR ended printing at the top of the Fed Funds target range also for two days. After that, things normalized a little bit, and we were heading into the GFC period, which we know tends to bring cash into the repo markets between the 18th and 25th of every month, exerting downward pressure on repo rates. But instead of going down, repo rates actually remain quite elevated.
So what does that tell us? Well, I think it tells us three things. One is that funding markets are getting much more sensitive to incremental collateral ads in the system, which is also to say that it's getting harder to find that marginal dollar of cash, that marginal dollar of financing.
And two, repo markets now operate with much more frictions now that balances of the overnight RRP have dwindled to nearly zero. And three, I think the last thing is SRF operations have so far been less effective in redistributing reserves across the system, plugging liquidity holes, and ultimately serving as a ceiling tool to police repo rates. That's great, Teresa.
That's a pretty comprehensive backdrop that you've given us on everything that's happened here. So given what you've told us and this sort of tightening in liquidity conditions that we've seen, what do you think it means for the Fed's QT process? So Lori Logan made a speech in October 2024.
And Lori Logan, who is the Dallas Fed President, she used to be the head of the silver portfolio at the New York Fed. She made a speech back then saying that in the long run, she believes that the Fed should operate with money market rates close to, but perhaps slightly below IRRB. So taking her definition, that definition in terms of how she defines ample reserves, and just looking at where TGCR and SOFR have been trading, we are basically there.
We're basically operating in this ample territory. This thought was further reinforced by Governor Wall's comments that he made last week during an interview at the Council of Foreign Relations, where he said he also believes that reserves are ample now. And then, of course, we had Chair Powell that came out and say that he believes balance runoff should end in the coming months, which in Fed parlance usually means two to three months from now.
But given what we saw in the market over the past couple of days, we do think it's time for the Fed to end QT now. And they can do that next week at the October FOMC meeting. The other consideration to think about, and Jay, you and I have talked about this at length, is that the Fed's tolerance to have another repeat of September 2019 is like zero, particularly considering the criticism that the Fed has veered outside of its core mandate in the last decade.
I mean, you just don't want to give folks any more ammunition to criticize the Fed that it's not able to control short-term rates. So, you know, we do think the Fed is going to be more prudent, more conservative this time around. The risk of letting QT run further is just too high, and it makes sense to end QT next week.
Yeah, I think that makes complete sense, Teresa. Your point about the political economy of this decision is valid, and particularly considering, you know, where we've come on the balance sheet. I mean, the balance sheet peaked just under $9 trillion early in 2022 and is about $6.6 trillion right now.
It's gone from 35% share of the U.S. economy down to 21%, and you're right, it's not as small as it was the last time we saw these funding stresses back in 2019. But if originally you thought it might end later this year, not to be flipping here, but what's another $40 billion of runoff between friends when the risks are bigger to the opposite end of the equation? No doubt.
And I've got one final question for you on this topic. So you've talked about it, that clearly the first step here is to just stop the QT process, and that makes sense. But we know there's also other factors at work, and you talked about TGA.
It's probably going to increase a bit further in the coming weeks on a seasonal basis given positive bill issuance, but also because the shutdown means fewer outlays are going out. So it is somewhat likely that we're going to see further tightening liquidity conditions into November. What other steps do you think the Fed can take here, and in what order would they be deployed?
Sure. So in addition to ending QT, there are a few things the Fed can do, or should I say the Fed needs to do? Because remember, as you said, the seasonality perspective, the TGA is running higher.
But I think over the medium term, I would think funding pressures will only continue to persist as the amount of collateral that requires funding in the system continues to grow as we continue to see more and more Treasury supply in the marketplace. So besides ending QT, the Fed needs to take actions to continue to support the funding markets and continue to provide liquidity into the markets. So what can they do?
They can do a few things. One, they can lower IORB and or the SRF rate to bring repo rates back down into the middle of the Fed Funds Corridor. Two, they can engage in TOMOs, or Temporary Market Open Operations.
Three, they can engage in Reserve Management Operations by buying T-bills in the secondary market. Of the three, I think one and two are most likely at this point in time. Three, you can kind of wait until 2026, because in contrast to 2019, we are in the transition point from abundant to just above ample.
We're not at reserve scarcity yet. So I just don't see an urgent of a need to immediately inject liquidity into the markets and start buying T-bills in the secondary market. With that said, we do have October month-end coming up, which is also Canadian year-end, and there are some large Canadian repo dealers out there.
So to the extent they pull back their balance sheets, and they will, you know, we do think having TOMOs there to help out with these days where we know there are going to be liquidity shortfalls are going to be helpful. And the same can be said with year-end. And I guess you might ask, well, why TOMOs when there is a standing repo facility there already?
And the reason for TOMOs is that it's a more efficient tool to add liquidity into the markets on a temporary basis, because there's really no stigma involved. There's no, you know, procedure involved where someone needs to get approval. On the other hand, SRF is a little bit more complicated.
You know, don't get me wrong. It's been very encouraging to see that we're using – that we're seeing SRF usage there, but there are still some frictions involved, which limits its use. So we do think having TOMOs around kind of on an ad hoc basis is needed to kind of fit those liquidity shortfalls is needed.
And so to that end, we also think lowering the SRF rate may be useful in seeing more usage at the facility and bringing repo rates back down to the middle of the corridor. I think that certainly will incentivize dealers to get a little bit more, you know, involved in terms of intermediating in the markets. I also think if you lower the SRF rate to something inside, just slightly below the Fed Funds corridor, I think that provides a little bit of a better bargaining chip when the dealers are negotiating with money funds in terms of where they get the tri-party funding.
So again, as I said, I think of all the tools that the Fed can do, I think, you know, adjusting the SRF rate makes sense, and then offering TOMOs as a complementary tool to SRF also makes sense as well. Yeah, that seems like a pretty reasonable way to approach things, especially because if, you know, the Fed has spent the post-2019 era trying to get the SRF up and running and then increasing its efficacy, and it's done a reasonable amount of work on that over the past year or so, you know, it would make sense to rely on those two before you actually turn to bill purchases, particularly because we're not at a point of scarcity. So thank you, Teresa.
That was great. Now, Phoebe, let's turn to you, and I mean, it's the other end of the curve, but it's the whole curve as well, and the upcoming quarterly refunding announcement is coming on November the 5th. We just published our refunding preview last night.
Today is Friday, October 24th. So can you lay the groundwork and really start with what's happened since the last quarterly refunding announcement in late July, and what do you think this means for Treasury's financing needs and for the supply outlook? Sure.
So I think there are two main factors to consider in terms of changes since our last refunding. The first is just we see a slightly better budget outlook. We have revised our forecast for the deficit for fiscal year 26 to $2.035 trillion, so $90 billion smaller than where we were three months ago, primarily reflecting higher tariff revenue than we previously thought.
The second thing is we now expect QT to end five months earlier than we previously forecasted, as Teresa just outlined. So that should reduce net privately held borrowing needs in fiscal year 26 by $25 billion, right? All maturing Treasuries will now be passively reinvested as add-ons, and then on top of that, we still think that once runoff ends, all MBS paydowns will now be reinvested into T-bills in the secondary market.
So that adds $15 billion or so of demand per month beginning in November. We also expect the Fed to buy an additional $8 billion or so of T-bills per month beginning in January. So altogether, we now see the Fed buying about $280 billion of T-bills in the secondary market next year.
In addition to those two changes, so on the QT side and our deficit forecast, I think we should also recognize there is some uncertainty around the Supreme Court's decision over IEPA tariffs. If we do see the Supreme Court strike down those tariffs before the end of the year, there could be more than $100 billion of tariff revenue subject to reimbursement. Whether or not those tariffs are reimbursed will depend on kind of the process that's set up.
If companies need to file lawsuits in order to get those refunds, it's possible that payments end up being considerably lower than the total amount eligible for reimbursement. So acknowledging some uncertainty there, on net, we project Treasury will announce $564 billion of privately held net marketable borrowing in the current quarter, $639 billion next quarter, assuming an end-of-quarter cash balance of $850 billion for both. Thanks, Phoebe.
That's at the stage pretty well. So given everything you've talked about, the smaller budget deficit coming from larger tariff revenue, the shift in the QT sort of expectations, but also the Treasury's ongoing comfort with the guidance that it's given us, do you think that that guidance that it anticipates maintaining coupon and FRN auction sizes for at least the next several quarters, does that still hold? And after that, kind of the follow-on question is, when do you think Treasury next needs to change its coupon auction sizes and why?
Right. So we think that Treasury will maintain that guidance at the upcoming refunding, and we have adjusted our forecast and now see the first round of coupon auction size increases coming in November of next year. So six months later than we previously thought.
So part of that is the updated sort of outlook both on the deficit side and also how we're thinking about QT, but I think we're also considering the political perspective here, as you mentioned, and guidance from Treasury officials continues to send a message of no urgency here. I think part of that is also the administration remains very optimistic about the medium-term growth outlook, but we're really not sensing urgency from the department at this stage. That said, you know, eventually increases will need to be made.
We think they will come before the end of next year because we still see a fairly large funding gap over the coming years. So what do we mean by that? The main point here is that the amount of funds that Treasury can raise under the current coupon auction schedule is set to decline pretty materially in the next few years as maturities are set to pick up.
So between fiscal years 2026 to 2030, we estimate a cumulative funding gap of $5.5 trillion and if Treasury made no changes to coupon issuance, leaning solely on T-bills in those years to address this funding gap, the T-bill share would rise very quickly. We project it would rise to about 27% by the end of 2028. You know, is that problematic?
Well, TBAC dove into that question in depth just last year, highlighting that running a higher T-bill share involves higher volatility of deficit financing, increased rollover risk, and it means that Treasury would need to run a higher TGA to protect against the loss of market access. So last year at least, TBAC concluded that Treasury should target a T-bill share averaging 20% over time. The current T-bill share, of course, is already sitting a bit above that, right around 21.5%.
You know, the fact that the Fed is expected to buy roughly $280 billion of T-bills next year probably gives Treasury a bit of flexibility around that long-term target, but even in our baseline, we still see the T-bill share rising to just over 23% by the end of 2028. That's great, Phoebe, and I think just kind of square around things out on that side and you know, Teresa, you were part of this in the webinar and podcast that was recorded by the macro research team earlier this week. The wild card is clearly that I think there is some belief that the growth of the stable coin market will contribute to more demand for T-bills over time, and while that's likely to be a positive, you know, here you've talked about it.
We still have the T-bill share drifting higher from its current level as well. So final question now for you, Phoebe, that when Treasury does start increasing that coupon issuance next November as you've talked about, do you think it's going to be across the curve? And as a corollary, because you've talked about the T-bill share continuing to rise over time, how do you see the wham of Treasury's debt evolving in the next few years?
Right, so we still think that Treasury will focus increases in the front end and belly. We think they'll leave 20 and 30-year auction sizes unchanged, and there's a couple of reasons for that. One is I think Treasury will kind of acknowledge the structural demand trends in the market.
We have been seeing slowing demand for long-duration Treasuries, and actually that's not just a story in the U.S., but globally as well we've seen, you know, kind of worsening demand for long-end government bonds. On top of that, you know, TBEX debt optimization framework has also shown that rising term premium and structurally higher deficits both justify leaning more heavily on the front end and belly of the curve. So I think, you know, both of those factors justify not increasing long-end sizes.
Of course, we could learn more about how Treasury is thinking about both of these things, both, you know, structural demand trends and the debt optimization framework at the upcoming refunding potentially, you know, through T-back charge questions. But I think, you know, it's worth highlighting, as you mentioned, under this forecast, even with the later start, no long-end increases and that T-bill share drifting a bit higher, we still project the wham of Treasury's debt to fall only gradually from here. So we see it falling to 68 months by the end of 2028 from 71 months currently, so a pretty small change.
Yeah. Thanks for that. And I was actually just joking.
I've got a bonus question for you because I figured there's one more thing we should talk about with respect to the refunding. So last quarter Treasury made some adjustments to its buyback program, doubling the size of the buybacks in the 10 to 20-year and the 20 to 30-year sector. What do you think should happen with respect to the buyback program at this next refunding announcement on November 5th, or should it make any changes at all?
In short, no, we don't think they will make changes. So last quarter they introduced a framework to assess kind of which sectors could benefit from increased buybacks. That framework looked at a few different metrics.
It looked at offer-to-max ratios, yield dispersion of off-the-runs and then also an on-the-run off-the-run spread as a measure of liquidity preference. And just based on that framework, we have seen the buyback score come down at the long end of the curve. The 7 to 10-year bucket is really the sector that has the highest buyback score suggesting potentially it could benefit from increased buybacks, but if you look at the details there, the offer-to-max ratios have actually been pretty low and on top of that at the last refunding, the 7 to 10-year bucket was actually highlighted as a place where maybe decreased purchases could be warranted.
So we don't think it's likely they would turn around and increase buyback sizes immediately. So just kind of considering that framework and everything we learned in the last quarter, we don't expect any changes to buyback sizes in a week and a half at the November refunding. Yeah, that's sensible as well.
So we've covered a lot here starting with the funding markets in QT and we think it's likely that the Fed will curtail QT on Wednesday next week and then a week after that with the refunding announcement, we're expecting very little, the guidance to remain unchanged and the Treasury to focus on the medium-term funding picture. So I think this is a perfect place to end the podcast and thanks Phoebe and Teresa to both of you for appearing today. Thanks to our listeners for tuning in as well and stay tuned for more episodes of At Any Rate, J.P.
Morgan's global research podcast series. This communication is provided for information purposes only. Please read J.P.
Morgan Research Reports related to its contents for more information including important disclosures. Copyright 2025, J.P. Morgan Chase & Co., all rights reserved.