Effective Fed Funds vs Target Rate Corridor: The IORB / ON RRP Band Mechanics and FOMC

When the FOMC announces a target range, it does not directly steer the effective Federal Funds Rate: it sets two administered rates, IORB and ON RRP, that bound a corridor within which the market settles.

This post-2008 architecture replaced the open-market operations regime that prevailed before the financial crisis. Two stress episodes (September 2019, March 2020) empirically tested its limits.

The Post-2008 Corridor Architecture: From Symmetric Regime to Floor System

Before October 2008, the Federal Reserve’s steering of the effective Federal Funds Rate rested on a fundamentally different mechanism from the current system. The Fed maintained a relative scarcity of bank reserves as a functional prerequisite, and the open market operations desk at the Federal Reserve Bank of New York intervened daily through Treasury purchases and sales to adjust available reserve quantities, and thereby the equilibrium price of the interbank market. If the effective rate threatened to deviate from the FOMC target, the desk injected or drained reserves until convergence. The corridor was symmetric: the discount window (the rate at which banks could borrow directly from the Fed as last resort) served as the logical ceiling, and the deposit facility rate, then unremunerated, served as an implicit floor at zero.

This regime assumed that banks actively sought to minimize their excess reserves, since those reserves earned no return. This constraint made the Fed Funds market highly active and sensitive to desk operations. In October 2008, the situation changed radically. The Federal Reserve began paying interest on bank reserves (Interest on Reserve Balances, IORB) at an administered rate. Simultaneously, the Quantitative Easing program flooded the system with reserves: the total stock exploded from less than $50 billion in 2007 to $848 billion by end-2008, then $2,948 billion on 30 September 2026 per the Federal Reserve H.4.1.

With this new structural overliquidity context, steering through open market operations became inoperable: relative scarcity cannot be enforced when all banks collectively hold massive excess reserves. The Fed therefore shifted to a floor system, where monetary policy transmission flows through the administered rates bounding the corridor, and no longer through interbank arbitrage mechanically reacting to desk operations.

The current architecture, dubbed the “ample reserves regime” in Fed communications (Federal Reserve, Policy Implementation Note), thus rests on two administered rates that the FOMC sets in parallel to the target range announcement: IORB and the ON RRP rate. The open market operations desk retains a residual function for fine liquidity management (notably via the post-2021 standing facilities), but no longer steers the effective rate in the way it did before 2008.

This article describes the mechanics of this floor system, its empirical limits exposed by two stress episodes (September 2019 and March 2020), and the institutional evolution that followed. For the effective rate definition itself and its NY Fed-published computation methodology, see the cluster’s foundational article, which logically precedes this reading.

IORB and ON RRP: The Two Administered Rates Bounding the Corridor

The post-2008 corridor rests on two administered rates that the FOMC sets simultaneously with the target range at each monetary policy meeting. how central banks steer rate cycles retraces the successive episodes.

Interest on Reserve Balances (IORB) is the rate the Federal Reserve pays to banks on their excess reserves. No eligible bank has reason to lend in the Fed Funds market below that rate, which the Fed pays without counterparty risk or horizon constraint. The market therefore consists mainly of banks borrowing from Federal Home Loan Banks, which have no access to IORB (Federal Reserve, FEDS Note of 27 February 2020): borrowing below IORB to deposit at IORB earns the spread. That arbitrage holds the effective rate below IORB, which acts as the corridor’s logical ceiling without being watertight: the effective rate exceeded it by 1 to 10 bp almost every day from March to October 2019.

At the other end, the Overnight Reverse Repo (ON RRP) rate is accessible to a set of eligible non-bank counterparties: primarily money-market funds, but also certain GSEs (government-sponsored enterprises) and foreign institutions. These counterparties place their liquidity with the Fed against Treasury collateral, at an administered rate the Fed sets. No money-market fund has interest in lending to a commercial bank below the rate at which the Fed itself borrows. The ON RRP rate therefore constitutes the corridor’s logical floor: the effective rate should not durably fall below the ON RRP rate.

Between these two bounds, the effective rate settles per the arbitrage conditions of the residual segment. Where it settles depends on how ample reserves are. From 2022 to 2024 the effective rate averaged 7 bp below IORB, about 4.5 bp below the middle of the target range; the gap narrowed to 6 bp in 2025 and to 2 bp in 2026. In May 2026 the target range was 3.50-3.75%, IORB 3.65%, ON RRP 3.50% and the effective rate 3.62 to 3.64%. Since the hike decided on 16 September 2026, the range is 3.75-4.00%, IORB 3.90%, ON RRP 3.75% and the effective rate 3.88% (5 October 2026, New York Fed).

This architecture presents three important properties. First, it is asymmetric in practice: stress episodes observed since 2008 have shown up as ceiling breaches (September 2019, then March 2020), while the effective rate has not fallen below ON RRP since 2016. Second, it is technically robust but empirically conditional: its functioning relies on the effective presence of arbitrageurs with sufficient balance sheets to exploit the gaps between the two administered rates and the Fed Funds market. When that condition fails (reserves too low, tight prudential constraints), the corridor briefly derails. Third, the IORB to ON RRP gap is not fixed: 25 bp from December 2015 to June 2018, then 20, 15 and 10 bp through the technical adjustments of 2018-2019, and 15 bp since 19 December 2024.

To understand how this mechanics inscribes into the broader Fed reaction function, see calibration through the Taylor Rule, treated in a dedicated cluster article.

The September 2019 Repo Spike: When the Corridor Breaks Through the Ceiling

On September 16 and 17, 2019, the U.S. financial system experienced the first major stress of the post-2008 corridor. The effective Federal Funds Rate, which had been running 2 to 4 bp above the 2.10% IOER since August, reached 2.30% on September 17, 20 bp above IOER and above the top of the target range (2.25%). On the adjacent collateralized repo market, SOFR reached 5.25% on September 17 and its 99th percentile 9%, the series’ high since its April 2018 launch (New York Fed).

The immediate cause was a temporary scarcity of available bank reserves. Several factors converged: simultaneous settlement of a large Treasury auction (which drained bank cash to the Treasury), quarterly corporate tax payments (again draining bank accounts to the Treasury General Account), and a total reserves level fallen to roughly $1.4 trillion, historically low for the post-QE regime. Below this threshold, certain commercial banks decided arbitraging the Fed Funds / IORB spread was no longer profitable, because their own balance sheet constraints (notably the Supplementary Leverage Ratio, SLR) made the operation costly in regulatory capital.

The Fed reacted in two phases. Immediately, the NY Fed desk launched massive overnight repo operations, several hundred billion dollars over the week, to directly inject missing reserves into the system. In the short term, the Federal Reserve restarted T-bill purchases at an initial pace of about $60 billion per month, announced on October 11, 2019 to run at least into the second quarter of 2020, explicitly presented as a technical balance-sheet management operation and not as a new QE round (Federal Reserve and New York Fed, October 11, 2019). The effective rate was back inside the target range from September 18, but stayed above IOER until December 2019.

The episode revealed a structural vulnerability of the floor system: it depends on the effective presence of arbitrageurs with non-saturated regulatory balance sheets. When post-Basel III prudential constraints (SLR, LCR) make arbitrage costly for near-saturated banks, the corridor ceases to be mechanically binding: it becomes conditional on bank capital availability and on the absolute level of available reserves. This institutional lesson directly informed the creation of the Standing Repo Facility (SRF) in July 2021, addressed below (Federal Reserve, FEDS Note “What Happened in Money Markets in September 2019?”, 27 February 2020).

March 2020 and the COVID Episode: The Ceiling Gives Again, the Floor Holds

The second major corridor stress, during the dash for cash of March 2020, pushed the effective rate in the same direction as in 2019. On March 12 and 13, 2020, the effective Federal Funds Rate printed 1.10%, level with IOER and 10 bp above ON RRP (1.00%). After the emergency cut of March 15 it printed 0.25% from March 16 to 18, at the top of the new target range and 15 bp above an IOER lowered to 0.10%, before easing back to 0.10% on March 25 (New York Fed).

The floor was never tested. The ON RRP facility, capped at $30 billion per day per counterparty since 2014, took in only $21 billion in total on March 12 and $285 billion on March 31 (New York Fed): nothing points to saturation. Across the whole daily series, the effective rate has printed below the ON RRP rate once, on 31 December 2015 (0.20% against 0.25%).

The Fed response combined several levers. Immediately, the FOMC convened an extraordinary session on March 15, 2020, and lowered the target range to 0.00-0.25%, effectively eliminating the need to defend a floor above zero. Swap lines with major foreign central banks were activated at full capacity, and massive QE began the following week: the Fed’s Treasury holdings rose by nearly $1.45 trillion between March 11 and April 29, 2020 (H.4.1). The ON RRP per-counterparty limit was only raised a year later, from $30 billion to $80 billion on 18 March 2021 (New York Fed), then to $160 billion, the limit in force in 2026. Sixteen unscheduled decisions, that extraordinary session among them, appear in the register of every FOMC decision since 1990, unscheduled ones included.

The lesson is therefore not symmetric: in 2019 as in 2020, the corridor gave way at the top, under a demand for liquidity. The lower bound was approached only in a regime of excess liquidity, after the 2021 limit increases: from 2022 to 2024 the effective rate stayed 3 bp above ON RRP, and ON RRP balances peaked at $2,554 billion on 30 December 2022, the only day above $2.5 trillion (New York Fed, ON RRP operations).

The Standing Repo Facility and the Institutional Evolution Post-2021

The institutional teaching of both episodes led the Fed to formalize, in July 2021, the creation of the Standing Repo Facility (SRF): a permanent instrument designed to place a hard ceiling on repo stress episodes and, indirectly, on the Fed Funds corridor.

The SRF operates on a simple principle: primary dealers and certain eligible banks can borrow overnight from the Fed against Treasury collateral, at a rate set by the Fed itself (the top of the target range, i.e. IORB + 10 bp: 0.25% in July 2021, 4.00% since 17 September 2026). The mechanism is analogous to a collateralized discount window, accessible daily and standardized. The SRF therefore creates a theoretical ceiling on repo operations: no primary dealer has interest in paying more on the private repo market than the SRF costs, which bounds SOFR spike episodes.

Indirectly, this ceiling also protects the Fed Funds corridor. Transmission is mechanical: when SOFR spikes, SOFR-Fed Funds arbitrage becomes unbalanced and the effective rate can follow upward, as in September 2019. By limiting the amplitude of SOFR spikes via the SRF, the Fed indirectly limits the probability of an IORB ceiling breach by the effective Fed Funds rate.

The post-2021 corridor architecture therefore rests on three coordinated instruments rather than the two initial administered rates. The addition of SRF as a third pillar, combined with the 2021 increases in the ON RRP per-counterparty limit, substantially strengthened the system’s robustness against future stress episodes. The SRF went almost unused until 2025, then drew demand at period ends: $29.4 billion on 31 October 2025, $31.5 billion on 31 December 2025, $18.5 billion on 17 February 2026 (New York Fed, repo operations). The amounts remain modest, but they come with an effective rate back to 2 bp below IORB in 2026, a sign of less ample reserves. For the in-depth analysis of this integrated architecture and its role in the integrated reading of FEDFUNDS, see the cluster’s hub article, and for the architecture of Fed tools as a whole, the parent sub-pillar.

Last updated — 6 October 2026