Markets exposed to a potential tax day blind spot7 have overlooked the now-low risk of a squeeze and respected the near-abundance of cash seeping into repo markets, evidenced by Mar/Apr’26 SOFR-FF8 (below) turning positive. As for the rest of the SOFR-FF curve, further compression is possible but will result in slim pickings for traders. For more significant moves, they must venture into stormier regions.

Volatility, thus opportunity, lies outside the comfort of o/n (overnight) markets. As laid out in War & Plumbing, a longer battle is more likely than a quick resolution, with conflict extending beyond late April resulting not just in zero rate cuts but also hikes9, ex-U.S. Despite what POTUS declares via social media, no true TACO remains on the table.

Read more

— created as part of The Warsh Ultimatum (upcoming)

The geopolitical chessboard, however, suggests the latter. Trump had been widely depicted as the first U.S. leader, perhaps in a century, to attempt a transition away from a longstanding unipolar U.S. order. Instead, as shown by capturing Maduro, increasing arms sales, and seeking to annex multiple islands, Trump has turned out to be a unipolar hawk. Instead of ushering in a multipolar world, POTUS has upheld America’s global “security pact”, in which the developed world, ex-Iran and Russia3, “employs” the U.S. empire to enforce clear waters and airways, enabling free trade. Enemies of this security pact face the reality that damaging global trade, financial markets, and the U.S. dollar hurts them more than they hurt America and its allies. Inflicting substantial harm to the U.S. empire must (and has) come from within. Nevertheless, attempts to destabilize the U.S. security pact — such as imposing sky-high tariffs — have illustrated that even the most disruptive U.S. leader ultimately defends the unipolar status quo.

Unable to contest such a power structure, Iran must resort to undermining the U.S. Yet Iran’s recent activities have spurred Trump to embark on another Middle Eastern foray. Most disputes, even during the Trump presidency, have died down quickly. But after eliminating Iran’s Supreme Leader, POTUS can’t simply pack his bags and leave. This looks to be the first “happening” that could take weeks, months, or years to resolve.

Unfortunately for markets, this war will inflict a slow burn on asset prices, with neither the Fed nor the U.S. govt supplying fuel for risk-taking. No sudden market move, not even a spike in Crude, should trigger the U.S. admin’s (low) tolerance level for market instability, one that would trigger the next TACO.

Read more

While repo rates continue to decline, higher interbank volumes have reduced the odds of o/n FF (a.k.a EFFR) printing lower, preserving a narrower1 SOFR-FF basis.

Swap spreads, meanwhile, may have widened (i.e. grown less negative) significantly, signaling easier plumbing conditions. Nevertheless, they could rally further. Fed proposals advocating for increasingly loose bank regulation should emerge by the end of Q1 this year.

As is evident, the Fed’s swift suppression of volatility has transformed money markets for better or worse, depending on the participant. It’s good news for U.S. monetary leaders who’ve compressed the overnight (o/n) rates complex to recent all-time lows, but bad news for some STIR traders starved of both volatility and thus opportunity. After the demise of the Fed’s QT, the Great Compression — i.e. a further narrowing across the SOFR-FF curve, post the Fed concluding its balance sheet reduction and starting reserve injections a.k.a RMOs — has been the only prominent trade. As anticipated, a few relative value plays have also emerged, based on the timing and size of the upcoming reduction in Fed purchases. But none offer, in Conks’ view, sizeable risk-versus-reward. Market participants must look further afield.

Read more

— created as part of The Warsh Ultimatum (upcoming)

With a Great Compression in money market rates underway, it’s the ideal moment for Fed officials to switch to a new target rate. As Part I unveiled in detail, volume in overnight Fed Funds (o/n FF) — the average rate of which the Fed currently targets to influence economic and financial conditions — has collapsed. Remaining o/n FF activities also fail to reflect the actual cost of tapping U.S. dollars while remaining vulnerable to a cluster of emerging rival interbank markets. The mechanism underlying the U.S. central bank’s primary monetary tool is thus approaching its breaking point.

For decades, targeting an unsecured interbank rate, such as o/n FF, made total sense. The Fed Funds market thrived and reflected the genuine price at which banks financed their operations. That ended, however, in 2008, with Lehman Brothers’ sudden demise. Unsecured interbank volumes soon fell by ~80%, while reserves injected by the Fed ballooned by multiple trillions. The latter has served as a stabilizing force for EFFR (i.e. o/n FF), enabling officials to avoid updating the target rate for more than a decade. Regardless, instability has risen sufficiently to warrant the next target-rate evolution.

To supplant o/n FF, central bank alchemists will explore multiple configurations: targeting an administered rate (e.g., IORB or o/n RRP), targeting an existing benchmark, assembling a “rates basket” and targeting its average, or constructing and targeting a new benchmark altogether. The ideal successor, though, grows evident quickly.

Read more

— created as part of The Fed’s New Target

The Fed will hope SOFR stays in its sweet spot, as the major $ clearers remain less willing to step in. The “Big Six” banks, led by JPMorgan, have ostensibly front-run cuts by swapping out more of their reserves for USTs (U.S. Treasuries)1, while those bearing unrealized losses in their investment portfolios hoard an extra reserve2 cushion to stem uneasy outflows. Further rate cuts by the Fed will only decrease large banks’ “excess” reserves further but increase their willingness to step in with their remaining balances at lower spreads3, containing o/n rates. Cuts are also an antidote to the weakening macro outlook, a win-win for monetary leaders.

Read more

Given the amount of information required to demonstrate the complete flows of a repo trade, the graphic is tough to navigate without a PDF version. Subsequently, we’re providing one for all readers below.

Conks Infographic The Modern Repo Market

20.9MB ∙ PDF file

Download

Download

Still, even with complete visuals, a written explanation is needed to comprehend such an intricate ecosystem. Welcome to a Repo 101: a brief primer that will complement Part II of our current target rate series, covering repo lingo and mechanics in plain English. This is not the usual Conks style of article. Instead, it’s written to clarify all key repo fundamentals and rates as concisely as possible.

Embrace the Lingo

If you’re an active money market participant or keen observer, you’ll soon discover why jargon in repo markets is unfortunately essential. For instance, “o/n TPR” (pronounced in English as “overnight triparty” — the meaning of which we’ll get to further on) is much easier to say than “the average overnight rate on non-centrally cleared triparty general collateral repos”!

Fortunately, over time, market participants have condensed various elements of the repo ecosystem into a succinct language. Learning this lingo is necessary to avoid going insane, but once acquired, understanding repo markets becomes straightforward. And the flows in the infographic above gain purpose and meaning.

The Repo Machine

For the uninitiated, a repo (short for repurchase agreement) is a cash loan secured against collateral. The cash borrower sells the security and repurchases it at a higher price at a later date. The difference in price, which is converted into a standardized repo rate, is the interest earned and the cost of borrowing cash. Trades are known as a reverse repo (RRP) from the perspective of the cash lender, and a repo from the perspective of the cash borrower, so reverse repos and repos are part of the same transaction, but the terminology is used to define which side of the repo trade someone is on. Even so, everybody calls it the repo market — not the RRP (reverse repo) market. In the opening “leg”, cash and collateral swap hands. In the closing leg, which occurs after the agreed loan period (such as o/n — overnight), cash and collateral are returned to their original owners, along with any repo interest owed by the cash borrower to the cash lender. Voila! A repo transaction.

Two types of repo exist: GC (general collateral) and SC (specific collateral) repos.

General collateral (GC) repos get their name from the fact that the end objective is either to raise cash (from the cash borrower’s perspective) or to earn interest (from the cash lender’s stance). As long as the collateral is sourced from a basket of high-quality securities, such as U.S. Treasuries and Agency MBS, the cash lender doesn’t care what backs the loan. If they can earn yield securely, then “general”, a.k.a. a broad range of securities, suffices. The Bank of New York (BNY) operates the “triparty” platform, the only venue for GC repo trading. “Triparty” and “GC” are therefore used interchangeably or concurrently to describe both the triparty repo market and GC repos. For instance, “triparty”, “triparty GC“, or “triparty general collateral” all refer to the triparty platform where GC repos trade exclusively.

In a GC repo, the cash borrower doesn’t want to trade with the cash it receives (as accounts on the triparty repo platform are for raising cash balances, not for active trading of securities; dealers have other accounts for that outside triparty), while the cash lender only wants to earn interest and doesn’t want to trade the security pledged to them in secondary markets. Thus, cash and collateral are locked in “boxes” on the triparty system (a.k.a “on the books” of a triparty custodian) to provide extra security for both parties. What’s more, since the lender doesn’t care (within reason) what security backs the loan, the BNY’s triparty system allocates a random security (hence general collateral) from the cash borrower’s account to back the loan. Ultimately, GC, a.k.a triparty, repos enable seamless, secure cash lending and borrowing. Dealers keep their overall cash levels in the green, and money market funds earn interest.

In contrast, specific collateral (SC) repos get their name from the fact that the main goal for the cash lender (also a securities borrower) is to obtain specific assets in their trading accounts, which they will use to enter trades outside repo markets, such as in the UST cash market (a.k.a. the secondary Treasury market). Unlike GC repos, cash and securities can be accessed and moved between participants during the opening and closing legs. The vast majority of these repos finance relative-value and basis trades in U.S. Treasury markets that require specific securities (hence, specific collateral) to execute, e.g. delivering a bond into a futures contract.

To facilitate these trades, Fedwire — the Fed’s settlement system — is (almost) always the venue for SC repos. DVP (delivery vs. payment) is the settlement mechanism used by Fedwire to securely transfer cash and securities between accounts. Accordingly, like with “triparty” and “GC”, “SC repos” are also called “DVP repos”. Another term for SC repos is “bilateral” repos, as there’s no third-party custody of cash and securities during trades. To reiterate, “DVP”, “bilateral”, and SC repos are the same thing. Participants, however, overwhelmingly refer to SC repo markets as “DVP repo”.

One type of repo (GC) tends to finance the other type (SC), as GC is a pure funding vehicle and SC is a trading vehicle. The major hurdle is gaining access to the markets required to execute elaborate fixed-income trades, which now hold the fixed-income ecosystem together.

A Great Divide

Repo markets are fragmented. How trades are cleared and settled determines where certain types of participants trade. Each type (GC and SC) has its own market for “centrally cleared” and “non-centrally cleared” dealings — which participants shorten to “cleared” and “uncleared” markets for simplicity. Central clearing providers (a.k.a. central counterparties or CCPs) provide dealers, brokers, and banks with risk management, such as absorbing losses from a counterparty default, in exchange for fees and the posting of margin. Membership is required, costly, and granted only after meeting strict criteria, usually met by large banks, brokers, or dealers. Consequently, only major banks, dealers, and other large players are likely to become members of a clearinghouse, and cleared markets (to reiterate, this means centrally cleared; all trades are cleared in some capacity) are classed as interdealer markets — and even interbank markets, as all large dealers tend to be subsidiaries of banks (known as dealer banks for short). All trades that go through central clearing — hence interdealer markets — will pass through the accounts of the FICC (Fixed Income Clearing Corporation), as it’s the sole central clearing provider in U.S. repo markets.

To trade GC (general collateral) repos, participants must have access to BNY’s triparty repo platform and be established firms. Money market funds are the major cash lenders, and cash borrowers are the largest dealers that money market funds are willing to lend to. The uncleared triparty segment, simply “triparty”, is “open to all”, as trades aren’t centrally cleared by the FICC. Alongside the main triparty market, though, is GCF (General Collateral Financing) Repo, the interdealer segment of triparty, in which trades are centrally cleared via FICC’s GCF Repo service. Dealers consequently trade GC repos with money funds in triparty and with fellow dealers and banks in GCF Repo. A great divide.

As for trading SC (specific collateral) repos, participants must have access to either the FICC’s DVP Repo service — the interdealer market for SC repos, or “NCCBR” (non-centrally cleared bilateral repo), a name that encompasses all the aforementioned jargon. This remains the uncleared market for SC repos. Similar to triparty and GCF Repo, dealers and banks trade in DVP repo, and dealers and hedge funds trade in NCCBR. Money funds are barely active, since they prefer tri-party repos, specifically designed for “basic” cash lending.

A repo market liberation, however, has gotten underway. In both GC and SC segments, members of the FICC (dealers and banks) can more easily sponsor or act as agents for their clients, who do not hold an FICC membership, to trade in centrally cleared markets. These services, all created by the FICC, are called Sponsored Repo (Sponsored GC and Sponsored DVP) and the Agent Clearing Service (DVP ACS and Triparty ACS). A growing number of repo segments and services now exist to support the evolution toward most trades being centrally cleared, thereby reducing systemic risk. That leaves NCCBR to eventually collapse in volume, as traders depart for cleared DVP markets, NCCBR’s centrally cleared rivals.

The Repo Gods

Out of the chaos, four key rates have arisen: o/n TPR, o/n GC(F), o/n DVP, and o/n NCCBR. Repos trade across multiple timeframes and are collateralized by various types of securities, yet the common practice is to publish and track rates based on overnight (o/n) trades backed by U.S. Treasuries — as the Fed does with its repo benchmarks (TGCR, BGCR, and SOFR).

o/n TPR (overnight triparty) refers to o/n rates on triparty repos a.k.a uncleared GC repos. In English, o/n TPR measures the average overnight rate at which dealers finance their operations from money funds. There are many other lenders and borrowers in triparty, but these two parties are the major rate-setters. o/n TPR is observable via the Fed’s TGCR (triparty general collateral rate) benchmark. Meanwhile, o/n GC refers to the average rate in the interdealer triparty market (the FICC’s GCF Repo Service), where dealers and banks lend and borrow cash among themselves. o/n GC is observable via the DTCC’s GCF Repo Index benchmark.

Meanwhile, in SC markets, o/n DVP refers to the average rate in DVP Repo, the interdealer DVP market (the FICC’s DVP Repo Service), where dealers and banks source securities among themselves. It also includes rates in Sponsored DVP and DVP ACS. o/n DVP is observable via the OFR’s Cleared Repo benchmark.

Finally, o/n NCCBR refers to the average overnight rate on non-centrally cleared bilateral repos. In English, o/n NCCBR measures rates at which dealers lend to hedge funds without going through central clearing. Unfortunately, the obscure nature of this segment means no published rate is available to track, but it likely trades somewhere between o/n DVP and o/n GC.

Adding context to these rates are the flows among participants on the “bid side” (i.e. those raising cash by pledging securities) and “offer side” (i.e. those lending cash secured against collateral). In one chain, dealers borrow cash from money funds at o/n TPR and lend it to other dealers in o/n GC or o/n DVP, or to hedge funds in o/n NCCBR. In a more specific example, a dealer who has access to cleared markets sponsors a hedge fund to trade in Sponsored DVP, allowing the hedge fund to go long and short two similar Treasuries on leverage in a relative-value trade.

As is evident, the repo market’s intricacy has created an abundance of jargon, but it’s justified to stop participants — and Conks — from going insane. Hopefully, learning the lingo above will help you, too, to navigate money markets with less friction. Armed with this primer and our mega-infographic, you should now be able to analyze repo markets with confidence, knowing what’s occurring beneath an obscure exterior.

Subscribe now

Give a gift subscription

Chartbook (24th Jan 2026)

• Changelog: SOFR-FF basis monitor

• WIP: STIR monitor, money market volumes monitor

If you act on anything provided in this newsletter, you agree to the terms in this disclaimer. Everything in this newsletter is for educational and entertainment purposes only and NOT investment advice. Nothing in this newsletter is an offer to sell or to buy any security. The author is not responsible for any financial loss you may incur by acting on any information provided in this newsletter. Before making any investment decisions, talk to a financial advisor.

EFFR, OBFR, SOFR, TGCR, and BGCR are subject to the Terms of Use posted at newyorkfed.org. The New York Fed is not responsible for publication of tri-party data from the Bank of New York Mellon (BNYM) or GCF Repo/Delivery-versus-Payment (DVP) repo data via DTCC Solutions LLC (“Solutions”), an affiliate of The Depository Trust & Clearing Corporation, & OFR, does not sanction or endorse any particular republication, and has no liability for your use.

— created as part of The Fed’s New Target

A decent zoom feature will be required to navigate this extensive infographic, but the image viewer on Substack does not seem to be sufficient. Consequently, it’s recommended to download and view the graphic outside of the Substack website on a retina desktop or a high-quality tablet — such as in Apple Preview/Microsoft Photos. Flows start in the top-right (with the green “start here” box) and move from right to left. Plus, it’s best to read the chart legend to learn all the necessary lingo. The Fed’s New Target: Part II will provide more details soon. Enjoy!

Money markets have experienced yet another quiet year-end. Just a week after ending its balance-sheet taper — where officials shrink the central bank balance sheet via QT1 — the U.S. central bank has commenced an onslaught of reserve injections, marking the start of a Great Compression in money-market rates. Year-end will soon turn into a distant memory, with repo rates (SOFR) set to print well below the Fed’s new fixed SRP (Standing Repo) rate while emergency Fed volumes top out. Large U.S. Treasury (UST) issuance combined with tighter liquidity was destined to make rate spikes a regular feature of the o/n (overnight) rates complex. Yet, the Fed’s RMOs (reserve management operations) have thwarted any chance of a plumbing “hard landing”2. The subsequent Great Compression should not just contain rates within the U.S. central bank’s limits but enable officials to enforce a fully secured standard, no longer targeting overnight Fed Funds (o/n FF) but one of its numerous repo benchmarks. A target rate evolution is looming.

Just as the Fed’s new permanent open market operations (POMOs) have commenced3, a lengthy balance-sheet winter has concluded. Banks' balance sheets, upon the year-end “air pocket” in money markets4, are now destined to expand rapidly, with expected plumbing relief priced in. The lengthy “tug of war” in (swap) spreads, Conks’ real-time gauge for measuring UST market frictions, has ended with an easing of constraints as the victor. This outcome was decided5 in part via one of the most noteworthy6 central bank leaders proposing a shift from targeting an unsecured rate (o/n FF) to targeting a “new-age” secured benchmark (repo) — a move intended to subdue volatility while granting the Fed even greater dominion over money markets7.

Upon delivering a speech at the end of September last year, Dallas Fed President Logan8 has initiated a shift toward the U.S. central bank replacing EFFR — the benchmark for unsecured9 interbank (i.e. o/n FF) rates10 — with a repo benchmark as its “policy rate”, the Fed’s primary mechanism it uses to influence U.S. (hence global) financial and economic conditions. The last prominent evolution, the switch from LIBOR11 to SOFR12, was completed just a few years ago, a decade after private and public-sector behemoths13 sought to oust an unsecured standard, led by LIBOR, from the mainstream. Back then, a potential successor was whittled down to OBFR14, a semi-global unsecured dollar rate, and SOFR, a secured overnight (repo) rate, with the latter crowned as the leading global benchmark for pricing trillions of dollars in financial assets. Yet, despite SOFR becoming LIBOR’s successor, monetary leaders still retained part of the old standard, preserving o/n FF (EFFR) as their desired target rate.

To this day, it has endured. But not for much longer. The near-death of the unsecured interbank market15 has set the stage for monetary leaders to complete the shift away from targeting rates based on unsecured dealings. Central bank officials have grown wary of the frail linkages, primarily opportunistic arbitrage trades, binding the Fed Funds market together.

Even so, rising uncertainty among major interbank players will be just one motivation for Fed alchemists to change targets. A fully secured standard awaits.

Read more

— created as part of The Fed’s New Target

A swift return to near-abundant reserves is looming. Yet, the Fed’s post-QT asset purchases will be devoid of any meaningful easing outside money markets. Equities will need to rely on other forces to further climb the wall of worry. Rather than absorbing duration (i.e. interest rate) risk by buying longer-term USTs and slowly unwinding those purchases, the Fed is set to purchase short-term USTs3 to preserve waning interbank liquidity. As revealed by Chair Powell in the latest FOMC presser, officials will imminently deploy reserve injections to build up a cushion against several incoming “blindspots” arising from a volatile TGA4, the big risk centered around April’s infamous tax day5. Following this year’s interbank flows threatening to induce numerous blindspots, the Fed — now much closer to the system’s lowest comfortable level of reserves (LCLoR) — won’t take any chances.

Read more

Market participants believe in the Fed & Treasury’s efforts to calm the short end, much like interventions across the rest of the curve. Treasury buybacks, deferred increases in duration (i.e. note and bond) issuance, and the U.S. Empire continuing to avoid fiscal dominance have placed a cap on yields. Yet, their latest incoming alchemy, the rework of a key regulatory ratio that banks can adopt as early as Q1 2026, may not deliver the “juice” leaders envision. The Fed’s Relief Valve, a reduction of its tough SLR (supplementary leverage ratio) — a ratio that requires banks to hold extra capital whenever they expand their balance sheets — will only be partially opened.

As expected, despite calls for greater balance sheet relief for the largest market makers in U.S. sovereign debt, the U.S. regulatory complex has opted to implement the least influential change to the SLR, choosing to deliver only tiny incentives for banks to acquire and absorb more USTs.

Read more

Reserves may arrive sooner than expected. The market has priced out delayed liquidity injections from the Fed, with an earlier balance-sheet expansion baked into markets. Trading desks now expect the Fed’s RMOs (reserve management operations) — where it buys bills with newly minted reserves to maintain ample liquidity1 — to begin closer to January.

This is observable via relative SOFR-FF (basis) spreads, which provide a contrast of dollar funding conditions at various points in time. Since SOFR-FF spreads rise (or grow less negative) when the market believes the SOFR-FF basis will narrow in one month compared to another, the latest market moves show traders are betting on earlier RMOs. They could be pricing in a multitude of catalysts. Nevertheless, SOFR-FF in Jan’26 tightening compared to March ‘262 likely suggests participants desire a swifter end to tighter money markets.

Read more

During mid-November, New York Fed President Williams convened with the heads of primary dealers2 at the annual U.S. Treasury Market Conference. The meeting took place not to resolve — presently nonexistent3 — liquidity issues but to remind key dealer banks they could tap cash via the SRF freely without stigma: the fear that other banks may conclude you’re in trouble upon borrowing Fed liquidity. SRF take-up (i.e. usage) appears to be worse than it was half a decade ago, even before the facility became official4. Much like the Fed’s leaky lower boundary (the bottom of its target range), the SRF has failed to contain money market rates, which officials now consider to be routine on month-ends. Rather than forming an immaculate ceiling, the Fed has come to admit that its repo counterparties won’t tap the facility en masse, despite vastly cheaper rates. Lackluster usage, even with persistent nudges from officials, reveals stigma has not only endured but proliferated.

Officials are already losing the battle to convince prominent market makers to use the SRF in size. But officials must also repair other defects with the current repo backstop. Stigma is only one of numerous obstacles preventing truly seamless access to central bank cash.

Read more

As for the November cash flood, admitting defeat, Conks’ 8bps target for the Nov SOFR-FF basis, i.e., the anticipated gap between SOFR and FF by November month-end, is now out of contention. Among the usual drivers, a reduced trading week paired with expected upward rate pressure from continued UST settlements means SOFR is scheduled to print just within the target range and the Fed’s SRF minimum bid, the cheapest rate to borrow via o/n (overnight) Fed repos, until month-end.

Read more

Money Market Commentary: The Basis Trade Scare

• Regulators and commentators have, once again, started to talk about the basis trade — a shorthand for the cash-futures basis trade (illustrated by Conks above), the primary glue holding UST (U.S. Treasury) cash, futures, and repo markets together, fueling the most efficient sovereign bond market on the planet. Data on basis trade activity remains scarce, but it is likely to exceed $1 trillion in exposure. Subsequently, as hedge fund leverage reaches new heights, monetary leaders have grown nervous. So anxious, in fact, that many media and financial commentators have started to join in. The outcome has been a series of posts and articles that have overlooked the recent evolution in the most systemically important trade and instead focus on potential market doom. The cash-futures basis trade “scare”, however, remains overblown.

Read more

• The Fed’s plumbing put awaits, with the U.S. central bank set to expand its balance sheet once again. Even so, monetary leaders initiating RMOs (reserve management operations) will likely unleash more flawed takes on “liquidity” than stimulate financial markets, let alone the real economy. What the Fed is about to unleash isn’t QE (quantitative easing), stealth QE, or even stealth easing. These involve the Fed swapping reserves for safe assets (USTs), absorbing duration risk in the hope that the private sector liquidates its most risky assets and spends into the real economy. This time around, monetary leaders will merely boost reserves to offset forces outside balance sheet runoff (i.e. QT) that will eat away at banks’ reserve balances and potentially (but unlikely) induce “repocalypse” 2.0.

• “RMOs” (reserve management operations) better describe what the Fed is about to undertake, namely boosting banks’ reserve levels, which will continue to wane, even with QT no longer in effect from December 1st. Despite the Fed’s balance sheet runoff ceasing, interbank liquidity will slowly wither away by either temporary or “permanent” means. Reserves can be temporarily removed (i.e. neutralized) from the banking system via an increase in overnight (o/n) Fed RRPs, or from foreign central banks investing their overnight dollar liquidity in the Fed’s FRP (foreign repo pool).

Read more