Central bankers often discuss monetary policy in relation to the “neutral rate”—the interest rate that, over time, is consistent with an economy operating near potential while inflation is stable around target. In practice, much of this discussion centers on the federal funds rate. That framework is useful, but incomplete.
The economy does not borrow only at the federal funds rate. A household buying a home typically borrows at a mortgage rate that is linked more closely to longer-term Treasury yields and mortgage-market spreads. A corporation considering a new factory or data center confronts multi-year borrowing costs and its weighted average cost of capital. Equity valuations depend partly on longer-term discount rates. Governments refinance themselves across the Treasury curve.
For this reason, the effective stance of monetary policy should not be judged solely by comparing the federal funds rate with an estimated short-term neutral rate, r*. It should be evaluated through the entire term structure of interest rates and the broader monetary and financial environment.
\[\boxed{\text{Effective Monetary Stance} = f\left( r_{short},r_{long},TP,B_{Fed},Credit,Liquidity \right)}\]
Here, TP is the bond-market term premium: the extra yield investors require to hold a longer-maturity bond rather than repeatedly rolling over short-term instruments. B_Fed denotes the size and composition of the Federal Reserve’s balance sheet. Credit refers to both the price and availability of private borrowing—such as corporate and household credit spreads, lending standards, and bank credit growth. Liquidity refers to the ease with which funding and financial assets can be financed and traded without large price dislocations; relevant indicators include reserve availability, repo and money-market conditions, bid-ask spreads, and market depth. Federal Reserve research has explicitly examined a long-term neutral interest rate, recognizing that long rates can be especially important for spending decisions.
There Is, in Effect, a Neutral Term Structure
The familiar expectations decomposition gives a useful starting point:
\[y_{10,t} \approx \frac{1}{10}\sum_{i = 1}^{10}E_{t}\left( r_{i} \right) + TP_{10,t}\]
The 10-year Treasury yield therefore reflects the expected path of future short-term rates plus a term premium. The latter can move because of duration risk, inflation uncertainty, Treasury supply, investor risk appetite, and other forces.
A 10-year yield can rise even when the Fed does not change the overnight rate. When that happens, financial conditions can tighten without an additional FOMC rate increase. This is precisely why a neutral-rate concept confined to the overnight rate can miss an important part of the monetary transmission mechanism.
A useful conceptual extension is a neutral term structure. At each maturity τ, define the neutral nominal yield as the yield consistent with sustainable real activity, inflation expectations anchored at the Fed’s objective, and a normal or equilibrium term premium:
\[\boxed{y_{\tau,t}^{*} = r_{\tau,t}^{*} + E_{t}\left( \pi_{\tau} \right) + TP_{\tau,t}^{*}}\]
Here y* is the neutral nominal yield at maturity τ; r* is the corresponding neutral real rate over that horizon; E(π) is expected inflation over the horizon; and TP* is an equilibrium term premium. The concept is not directly observable and should not be interpreted as a precise policy target. Rather, it is a diagnostic benchmark: if market yields across important maturities move materially above their neutral configuration, financial restraint can increase even if the policy rate is unchanged.
This does not mean the Fed should target every point on the Treasury curve. It means policymakers should recognize that monetary restraint is transmitted through the curve as a whole.
Three Monetary Levers, Not One
\[\boxed{\text{Policy Rate} + \text{Yield Curve} + \text{Balance Sheet} \rightarrow \text{Financial Conditions} \rightarrow \text{Aggregate Demand}}\]
The federal funds rate directly anchors the very short end and influences expectations of future short rates. Long-term yields influence mortgages, corporate borrowing, capital investment, equity discount rates, commercial real estate, and government financing.
The Fed’s balance sheet operates through a distinct but related channel. Asset purchases can increase reserves and remove duration risk from private portfolios, reducing term premiums and easing financial conditions. Balance-sheet reduction works in the opposite direction by increasing the amount of duration that the private sector must absorb. Its effect on 'liquidity,' however, should be stated carefully: changes in Fed assets, reserves, Treasury cash balances, currency, repo facilities, and other liabilities jointly determine reserve conditions, so a smaller balance sheet does not translate one-for-one into a mechanically equal reduction in broad money or market liquidity.
Federal Reserve staff estimated (Crawley et al., 2022) that a one-time permanent reduction in Fed holdings of 10-year-equivalent Treasury securities equal to 1% of nominal GDP raises the 10-year term premium by about 10 basis points in their model. The same tightening could be generated, in that model, by raising the average expected federal-funds-rate path over the coming decade by about 10 basis points. The authors stress that this is a model-dependent rule of thumb, not a fixed conversion.
In a broader scenario, the same 2022 FEDS Note estimated (Crawley et al., 2022) that reducing the balance sheet by roughly $2.5 trillion relative to a no-runoff counterfactual was approximately equivalent to a little more than 50 basis points of sustained policy-rate tightening. This provides useful quantitative evidence for treating the balance sheet and long-term rates as part of the overall monetary stance.
\[\boxed{\Delta B_{Fed} < 0 \Rightarrow TP \uparrow \Rightarrow y_{long} \uparrow \Rightarrow \text{Financial Conditions Tighten}}\]
From Rates and Liquidity to Growth, Employment, and Inflation
The transmission mechanism can be represented more completely as follows:
\[r_{short} \uparrow ,\quad r_{long} \uparrow ,\quad B_{Fed} \downarrow\]
\[\Downarrow \quad\text{Financial Conditions Tighten}\quad \Downarrow\]
\[\text{Credit Growth and Interest-Sensitive Spending} \downarrow \Rightarrow C + I \downarrow \Rightarrow AD \downarrow\]
\[\Rightarrow \text{GDP Growth} \downarrow \Rightarrow \text{Labor Demand} \downarrow \Rightarrow \text{Wage/Price Pressure} \downarrow \Rightarrow \pi \downarrow\]
In this chain, C denotes household consumption, I denotes business and residential investment, and AD denotes aggregate demand. Higher short rates affect floating-rate borrowing, bank funding, money-market returns, and near-term spending incentives. Higher long rates raise mortgage and corporate financing costs and increase discount rates applied to long-lived assets and investment projects. Balance-sheet policy can affect term premiums, reserve conditions, and the quantity of duration held by the public.
There are important lags and nonlinearities in every arrow. The Fed also does not mechanically control broad money supply through the size of its balance sheet. Commercial-bank lending creates deposits; households and firms determine desired money balances; Treasury operations shift reserves and deposits; and velocity changes the amount of nominal spending supported by a given stock of money.
A reduced-form representation is:
\[\Delta Y = f\left( \Delta r_{s},\Delta r_{l},\Delta B_{Fed},\Delta Credit,\Delta M,\Delta V,\Delta Z \right)\]
Here ΔY is the change in real output or growth; Δr_s and Δr_l are changes in short- and long-term real or nominal rates as appropriate; ΔB_Fed is the change in the Fed balance sheet or relevant securities holdings; ΔCredit captures changes in the price and availability of credit; ΔM is money growth; ΔV is the change in money velocity; and ΔZ represents other shocks, including fiscal policy, energy prices, productivity, exchange rates, and geopolitical events. The equation is deliberately reduced-form: the coefficients are state-dependent and cannot be treated as stable mechanical multipliers.
The monetary identity MV = PY provides another useful organizing framework for understanding why monetary conditions cannot be inferred from the quantity of money alone.
\[\boxed{MV = PY}\]
In this identity, M is broad money (for example, M2); V is money velocity, defined as nominal spending per unit of money; P is the general price level; and Y is real output. The product PY is therefore nominal GDP. In growth-rate form, the identity can be written approximately as:
\[\Delta\ln M + \Delta\ln V \approx \Delta\ln P + \Delta\ln Y\]
or, more intuitively, money growth plus velocity growth is approximately equal to inflation plus real-output growth. This is an accounting identity, not by itself a behavioral theory of inflation. It becomes economically useful when combined with an explanation of how interest rates, credit conditions, portfolio choices, and confidence affect M, V, P, and Y.
First, the Fed's balance sheet is not the same thing as broad money. When the Federal Reserve purchases a Treasury or agency security, it normally creates central-bank reserves. Those reserves are settlement balances of depository institutions; they do not mechanically become an equal amount of household or corporate spending, nor do they generate a fixed multiple of bank lending or deposits.
\[B_{Fed} \uparrow \Rightarrow \text{Bank Reserves} \uparrow \mspace{6mu} \Rightarrow \not{}\mspace{6mu}\text{Proportional }M \uparrow \text{ or }PY \uparrow\]
Broad money depends importantly on commercial-bank lending and deposit creation, but also on borrowers' willingness to borrow, banks' willingness and capacity to lend, Treasury operations, and household and corporate portfolio decisions. Thus an increase in central-bank liquidity can ease financial conditions without producing a proportional increase in M2, nominal GDP, or inflation.
Second, velocity is an essential part of the transmission mechanism. Even if broad money is unchanged, nominal spending can weaken if households and firms choose to hold larger money balances relative to expenditures, postpone purchases, or reduce investment. In the identity, this appears as lower measured velocity.
\[M\text{ unchanged},\quad V \downarrow \Rightarrow PY \downarrow\]
For example, if broad money grows by 5%, velocity falls by 3%, and real output grows by 2%, the arithmetic of the identity is approximately consistent with zero inflation:
\[5\% - 3\% \approx \pi + 2\% \Rightarrow \pi \approx 0\%\]
The example is intentionally simplified; it illustrates why money growth alone does not determine inflation. A decline in velocity can offset part or all of the spending impulse associated with a larger money stock.
Third, higher long-term rates can tighten monetary conditions even before measured M contracts. Suppose the 10-year Treasury yield rises materially while M2 remains roughly unchanged. Mortgage rates and corporate bond yields can rise, the discount rate applied to investment projects can increase, and equity and real-estate valuations can fall. Households may borrow less and consume less; firms may postpone capital expenditure; lenders may become more selective; and asset-related wealth effects may weaken.
\[r_{long} \uparrow \Rightarrow \text{Credit Demand} \downarrow ,\ I \downarrow ,\ C \downarrow ,\ \text{Asset Values} \downarrow \Rightarrow PY \downarrow\]
In the MV = PY framework, weaker nominal spending with an initially unchanged money stock is reflected in a decline in V. Over time, weaker credit creation may also slow M itself. This is precisely why the stance of monetary policy cannot be read from M2 or the Fed balance sheet in isolation.
A necessary caution is that velocity is not an independent policy lever. Because V is defined as PY/M, observed velocity partly summarizes the outcome of spending behavior rather than independently causing it. Interest rates, credit availability, confidence, wealth, payment technology, portfolio preferences, and economic activity jointly influence spending relative to money balances. The causal story therefore runs through those behavioral and financial channels, with V serving as a compact summary of their effect on the use of money.
The broader implication for monetary neutrality is important: the Fed does not have to shrink broad money directly for monetary conditions to become more restrictive. A rise in long-term rates can reduce the willingness of households and firms to put money and credit to work. Conversely, a larger central-bank balance sheet does not guarantee strong nominal demand if credit demand is weak or money velocity falls.
September 2026: The Bond Market Has Already Tightened Policy
This framework is particularly relevant ahead of the September 15–16, 2026 FOMC meeting. Since July, the long end of the Treasury curve has repriced sharply. The July average 10-year Treasury yield was approximately 4.60%, while the yield was around 4.95% on September 10; the 30-year yield was roughly 5.36%. Depending on the specific July observation used, the increase from levels around 4.45% is even larger.
The movement is economically significant. A 35–50 basis-point increase in long-term rates is not equivalent in every respect to a 35–50 basis-point increase in the federal funds rate. The duration, persistence, source of the move, credit-spread response, and effect on real rates all matter. But it should not be treated as irrelevant simply because the FOMC did not initiate it.
Mortgage rates, corporate discount rates, and the cost of long-duration financing respond to the market curve, not merely to the overnight policy rate. The bond market has therefore delivered a meaningful degree of de facto monetary tightening ahead of the September meeting.
But the Reason Long Rates Rose Matters
\[\Delta y_{10} = \Delta E\left( r_{short} \right) + \Delta E(\pi) + \Delta TP\]
If yields rise because stronger expected growth raises the expected path of future short rates, the movement provides restraint but also signals economic strength. If yields rise because Treasury supply or fiscal risk raises the term premium, the movement is more directly a tightening of financial conditions.
If yields rise because oil prices surge and inflation expectations deteriorate, however, the higher yield tightens financial conditions while simultaneously telling the Fed that inflation risk has increased. The same market move can therefore carry two opposing policy implications.
\[\boxed{\underset{\text{More Market-Imposed Restraint}}{\underbrace{y_{long} \uparrow}}\quad\quad\text{versus}\quad\quad\underset{\text{Greater Inflation Risk}}{\underbrace{\pi^{e} \uparrow}}}\]
This is the central policy tension going into September. Higher long-term rates are already doing part of the Fed’s work through financial conditions, but the information embedded in those rates may simultaneously signal that the inflation problem has become more serious.
A Better Way to Think About Neutrality
The neutral rate should therefore not disappear from monetary economics. It should be expanded. Instead of asking only whether the federal funds rate is above or below r*, policymakers should also ask where the entire yield curve lies relative to a neutral configuration; what portion of long yields reflects expected policy, inflation expectations, and term premium; what the Fed balance sheet is doing to duration and reserve conditions; and what the combined configuration is doing to credit and aggregate demand.
\[\boxed{\text{Required Fed Action} = f\left( \text{Inflation Gap},\text{Employment Gap},\text{Existing Financial Restraint},\text{Inflation Expectations},\text{Supply Shocks} \right)}\]
The higher the restraint already being imposed through long-term rates and credit conditions, the less incremental tightening may be required from the federal funds rate, all else equal. Those last four words are essential.
A Numerical Illustration: How Market Tightening Could Change the Fed Decision
Consider a deliberately simplified example. Suppose that, before the rise in long rates, the FOMC’s assessment implied that an additional 25-basis-point increase in the federal funds rate was needed to generate sufficient restraint. Now suppose the 10-year Treasury yield rises 50 basis points and the move persists.
It would be incorrect to assume a one-for-one equivalence and simply subtract 50 basis points from the policy rate. Instead, policymakers could translate the market move into an estimated equivalent tightening in a financial-conditions or macroeconomic model. For illustration, suppose only 40% of the 50-basis-point long-rate increase is judged to represent persistent real-rate/term-premium tightening relevant to demand, while the rest reflects inflation compensation or transitory factors:
\[50\text{ bp} \times 40\% = 20\text{ bp of effective persistent long-rate tightening}\]
Suppose further that credit spreads and mortgage rates have tightened enough to add an estimated 10 basis points of policy-rate-equivalent restraint, while balance-sheet/reserve conditions are judged neutral at the margin:
\[\text{Market-Imposed Equivalent Tightening} = 20 + 10 = 30\text{ bp}\]
Against an initial judgment that 25 basis points of additional restraint was required:
\[\text{Incremental Fed Tightening Needed} = 25 - 30 = - 5\text{ bp}\]
The policy implication would change from 'raise 25 basis points' to 'hold the policy rate and observe the transmission of existing restraint.' This is an illustration, not an empirical estimate of the current September decision. The conversion coefficients must come from a defensible financial-conditions or structural model, and the Fed must separately assess whether the rise in yields reflects worsening inflation expectations.
If, alternatively, half of the long-rate increase reflected a persistent rise in inflation expectations, the policy conclusion could reverse. That is why the decomposition of the yield move is as important as its magnitude.
Conclusion and Required Policy Action
Monetary policy is often summarized by one number: the federal funds rate. The economy is not. Households, corporations, financial markets, and governments operate across an entire structure of interest rates, credit spreads, asset prices, liquidity conditions, and money creation.
A modern concept of monetary neutrality should therefore move beyond a single overnight rate.
\[\boxed{\text{The Fed sets the short rate. The economy lives on the yield curve.}}\]
The required policy action follows directly from this framework. At the September meeting, the FOMC should explicitly incorporate the material rise in long-term yields and broader credit conditions into its estimate of existing monetary restraint before deciding that another increase in the federal funds rate is necessary.
Given the substantial market-driven rise in long rates since July, the burden of proof for an additional September rate increase should be higher than it was at the previous meeting. If the increase in long yields is primarily a persistent rise in real yields or term premium, and credit conditions remain restrictive, a policy hold would be economically defensible because markets have already supplied part of the required tightening.
If, however, the rise is primarily the consequence of unanchored inflation expectations, a persistent energy shock, or evidence that underlying inflation is accelerating, the Fed should not treat higher long rates as a substitute for policy action. In that case, the inflation signal can outweigh the financial-conditions restraint.
The practical recommendation is therefore conditional but clear: measure the total effective restraint first; decompose the rise in long yields second; then set the short rate. The federal funds rate should be the residual policy instrument needed after accounting for the restraint already delivered by the yield curve, credit conditions, and the balance-sheet environment.
Selected References
Crawley, Edmund, Etienne Gagnon, James Hebden, and James Trevino (2022), “Substitutability between Balance Sheet Reductions and Policy Rate Hikes: Some Illustrations and a Discussion,” FEDS Notes, Board of Governors of the Federal Reserve System, June 3, 2022. DOI: 10.17016/2380-7172.3147.
Doniger, Cynthia, James Hebden, Luke Pettit, and Arsenios Skaperdas (2019), “Substitutability of Monetary Policy Instruments,” FEDS Notes, Board of Governors of the Federal Reserve System, July 19, 2019.
Eyal, Benjamin, Dave Na, and Arsenios Skaperdas (2026), “A Decomposition of Balance Sheet Reduction,” FEDS Notes, Board of Governors of the Federal Reserve System, February 2, 2026.
Anderson, Alyssa G., Alessandro Barbarino, Anthony M. Diercks, and Stephen Miran (2026), “A User’s Guide to Reducing the Federal Reserve’s Balance Sheet,” Finance and Economics Discussion Series 2026-019, Board of Governors of the Federal Reserve System.
Board of Governors of the Federal Reserve System, H.4.1, “Factors Affecting Reserve Balances of Depository Institutions and Condition Statement of Federal Reserve Banks,” September 10, 2026.
Board of Governors of the Federal Reserve System, “An Estimate of the Long-Term Neutral Rate of Interest,” FEDS Notes, September 5, 2018.
Board of Governors of the Federal Reserve System, “Why Have Far-Forward Nominal Treasury Rates Increased So Much in the Past Few Years?” FEDS Notes, February 12, 2026.
Federal Reserve Economic Data (FRED), U.S. Treasury constant-maturity interest-rate series.