Do Higher Interest Rates Cause Inflation? What the Evidence Says About the Fed's September Hike

On September 16, 2026, the Federal Open Market Committee raised the federal funds target range by a quarter point to 3.75% to 4.00%, its first increase since 2023, and said in its statement that “inflation remains elevated.”¹ Eight days later Bill Ackman posted that the Fed “might have just made a mistake.” His argument ran in two steps. Demand for compute and energy will not fall when rates rise, because “winning the race for super intelligence has a near infinite ROI.” And since “interest costs are embedded in everything,” higher rates will raise prices instead of restraining them: “the more the Fed raises rates, the more inflation we will have.”²

The same week the 30-year fixed mortgage rate reached 7.03%, up from 6.30% a year earlier.³ The August CPI report had shown headline inflation at 3.4%, core at 2.4%, and shelter at 3.0%.⁴ So the claim lands on receptive ground. Anyone shopping for a house has watched a rate increase make their housing more expensive, and shelter is more than a third of the CPI.

Several measured mechanisms support parts of the claim, and economists have built models in which it holds. As a statement about aggregate inflation, most of the causal evidence of the last sixty years contradicts it. Sorting it out means separating four things that get run together: what a chart of rates and inflation can and cannot show, what happens when researchers isolate a policy change from the Fed’s reaction to the economy, which specific prices higher rates do push up, and what conditions would have to hold for the aggregate result to flip.

The short answer

Higher policy rates lower aggregate inflation over a horizon of one to three years. That is the finding of every major study that isolates policy changes from the Fed’s systematic response to the economy, and of the central banks’ own model assessments of the 2022 to 2023 tightening cycle. Higher rates also raise particular prices through particular channels: firms’ working-capital costs, and rents, through less construction, mortgage lock-in, and would-be buyers who stay renters. Those channels are real and measured, and in housing they can be large. They coexist with the aggregate result. They do not overturn it.

Theory produces an inflationary rate hike under two conditions: a permanent change in the policy rate, which is the neo-Fisherian case, or a government that pays its higher interest bill with printed money rather than taxes, which is fiscal dominance. Neither describes a quarter-point hike aimed at a 3.4% CPI. And an AI investment boom that raises the neutral real rate changes how restrictive a given policy rate is, without changing the sign of what a hike does.

Why the chart cannot settle it

Plot the ten-year Treasury yield against CPI inflation for the 1970s and both lines climb together. The annual averages:

Read causally, that table says higher rates bring higher inflation. Read the other way, it says inflation dragged rates up behind it, which is what the Fisher equation predicts: a nominal yield is a real yield plus expected inflation, so when expected inflation rises and the real rate does not, the nominal yield has to follow. Mishkin tested this in 1992 and found that inflation and interest rates “trend together in the long run when they exhibit trends,” with no evidence of a short-run relationship at all.⁶ The 1970s co-movement is the long-run Fisher effect showing up in a decade of rising trend inflation. It says nothing about what a rate hike does.

The second problem is that central banks raise rates because inflation is high or forecast to rise. Every observed tightening is therefore paired with the conditions that provoked it. When Volcker’s Fed pushed the effective funds rate to a monthly average of 19.1% in June 1981, CPI inflation had peaked at 14.8% the year before.⁷⁸ A naive regression of inflation on the policy rate over that period would find that hikes accompany high inflation, because they do. The rate was responding to the inflation.

Econometricians ran into exactly this in the early vector autoregression literature. Sims documented in 1992 that a measured tightening was often followed by a rise in the price level, and Eichenbaum’s discussion of that paper named it the “price puzzle.”⁹ The accepted resolution was that the Fed had information about coming inflation that the statistical model lacked, so what looked like a hike causing inflation was a hike anticipating it. The cost channel discussed below is the main alternative explanation, and it has been tested directly.

What isolated policy shocks show

The way around both problems is to find changes in the policy rate that were not responses to the economic outlook, and trace what happened afterward. Romer and Romer did this in 2004 for 1969 to 1996 by reading the FOMC’s intended funds rate off meeting records, regressing its changes on the Fed staff’s own Greenbook forecasts, and keeping the residual: the part of each move the forecasts did not explain.¹⁰

A one-percentage-point shock of that kind pushed industrial production down 4.3% at its trough, roughly two years out. The price level was “virtually unchanged for the first 22 months and then falls steadily,” ending 6% lower after four years. The closest thing to a price puzzle in their estimates is a 0.3% rise over the first eight months that is not close to statistically significant.¹⁰

The modern approach uses the movement in interest-rate futures in a narrow window around FOMC announcements, on the logic that nothing else moves markets in those thirty minutes. Gertler and Karadi found in 2015 that shocks identified this way produce output and inflation responses “typical in monetary VAR analysis,” and that modest moves in short rates translate into large moves in credit costs because term premia and credit spreads move too.¹¹ Jarociński and Karadi then showed in 2020 that an announcement carries two kinds of news at once. A pure tightening raises rates and lowers stock prices; a signal that the Fed sees a stronger economy raises both. Ignoring the second kind “biases the inference on monetary policy nonneutrality,” and separating them restores the conventional response to the first kind.¹²

The Fed’s own account of the mechanism is broader than “people borrow less.” Lower rates “elicit greater spending on goods and services, particularly on durable goods such as electronics, appliances, and automobiles,” make marginal investment projects attractive, and “affect the demand for housing and thus influence house prices.” Higher rates run each of those in reverse.¹³ Credit availability, asset prices, the exchange rate and expectations all move with them.

The 2022 to 2023 cycle was tested in real time. The ECB’s model assessment of its tightening through March 2023, published in May 2023, estimated that the rate increases had “lowered inflation by around 50 basis points in 2022,” with the downward impact “expected to average around 2 percentage points over the period 2023-25, with estimates differing substantially across the three models.”¹⁴ On the U.S. side, an IMF working paper asked whether the pre-pandemic relationships still held. Its answer for February through July 2022 was that transmission ran “around 25 percent weaker than normal,” which meant the Fed needed roughly four hikes to do the work of three. Weaker, on that estimate, and still pointed the usual way.¹⁵

The cost channel: where the “embedded in everything” argument shows up in data

The mechanism behind Ackman’s second step has a name and a literature. A firm pays wages, suppliers and inventory before it collects revenue. If it finances that gap with short-term borrowing, the nominal interest rate is part of its marginal cost, and a rate increase is a cost increase it will try to pass on. Barth and Ramey found in 2001 that after monetary contractions many industries showed falling output together with rising price-to-wage ratios, the signature of a supply shock rather than a demand shock. The pattern was much stronger in the 1959 to 1979 data than in 1983 to 2000.¹⁶ Ravenna and Walsh built the channel into a standard New Keynesian model in 2006 and showed that once marginal cost depends on the nominal rate, tightening generates its own cost-push shock and the central bank has to trade inflation against output.¹⁷

The empirical record is mixed. Rabanal estimated a model with the channel for both the U.S. and the euro area and concluded that “cost channel effects are absent in both cases.”¹⁸ The strongest recent evidence in its favor came from the ECB in August 2025. Researchers linked firms’ one-year-ahead selling-price expectations from the ECB’s SAFE survey to each firm’s borrowing in the euro-area credit register and to monetary policy surprises during the 2022 to 2023 tightening. Firms revised their expected selling prices upward after tightening announcements, and the revision grew with the firm’s working-capital exposure: on average a 10 basis point policy surprise raised expected selling prices 60 basis points, against a baseline expectation of about 4.5%.¹⁹

The authors are careful about what that shows. They estimate only the cost channel, and write that “while we find a positive impact on firms’ expected selling prices, our results do not contradict the overall impact of monetary policy tightening on inflation.”¹⁹ The same firms that face higher financing costs face customers with less credit, competitors with less demand, and an exchange rate that has strengthened. What a firm would like to charge and what it can charge are different quantities. The cost channel is a real force pointed the wrong way for the Fed, and the aggregate studies above already include it.

Mortgage interest is not in the CPI

Housing is where the claim is strongest, and it needs a distinction the argument usually skips. The Bureau of Labor Statistics does not measure a homeowner’s shelter cost by what the homeowner pays. It “views owned housing units as capital (or investment) goods distinct from the shelter service they provide,” so “interest costs (such as mortgage interest), property taxes, real estate fees, most maintenance, and all improvement costs are part of the cost of the capital good and are also not treated as consumption items.” What enters the index instead is owners’ equivalent rent: “the implicit rent that owner occupants would have to pay if they were renting their homes.”²⁰

That was a deliberate change. Before 1983 the CPI-U priced homeownership with a user-cost approach that included an interest estimate based on house values, which meant that Volcker’s rate increases fed directly into measured inflation. BLS announced in 1981 that it would switch to rental equivalence effective with the January 1983 data, precisely to stop the index treating a house purchase, and its financing, as current consumption.²¹

Owners’ equivalent rent carries a relative importance of 25.9% in the CPI, rent of primary residence 7.7%, and shelter in total 35.3%, on the July 2026 weights published with the August release.⁴ In August those components rose 3.1%, 2.7% and 3.0% over the year. Meanwhile the monthly payment on a new $400,000 mortgage went from about $2,476 at last September’s 6.30% to about $2,669 at this September’s 7.03%, an increase of nearly 8%, and none of that increase appears in any CPI component. The affordability of buying a home and the inflation rate of shelter are different objects, and they can move in opposite directions for years.

Two more details matter. The Fed’s 2% objective is defined on the personal consumption expenditures price index, “as measured by the annual change in the price index for personal consumption expenditures,” and shelter is about 16% of that basket against 35% of the CPI.²²²³ So a shelter-specific mechanism moves the CPI more than twice as hard as it moves the index the Fed targets. And BLS collects each sampled unit’s rent only every six months, with most observations being continuing leases at unchanged rents, so the shelter index responds to a turn in market rents slowly and over a long stretch.²⁰ When the Fed hiked through 2022 and CPI shelter kept accelerating into 2023, part of what the index was reporting was the 2021 rent surge still working through the panels. San Francisco Fed researchers used that lag structure in August 2023 to forecast that shelter inflation would slow sharply over the following 18 months, which it did.³⁴

Three ways higher rates can push rents up

Less construction where supply is inelastic

Higher rates raise a developer’s financing cost and lower the value of the finished project. The Fed’s October 2023 Beige Book recorded St. Louis construction contacts saying financing was available “at such high rates that, when combined with higher input costs, ‘the numbers don’t work’ on even the best projects.”²⁴ If supply were elastic, a temporary dip in building would matter little for rents. It is not elastic in the places where rents are highest. Gyourko and Molloy’s review of the regulation literature concludes that land-use regulation “appears to raise house prices, reduce construction, reduce the elasticity of housing supply, and alter urban form.”²⁵ Glaeser, Gyourko and Saks showed that supply elasticity decides whether a demand increase becomes more housing or more expensive housing.²⁶ A rate-driven construction pause in an inelastic market removes units that would have been rented.

Demand-side subsidies belong in this story only with a qualifier. Favara and Imbs found that an exogenous expansion in mortgage credit raised house prices, “but to a lesser extent in areas with elastic housing supply, where the housing stock increases instead.”²⁷ Sommer and Sullivan’s model finds that removing the mortgage interest deduction lowers house prices by 4.2% and raises the homeownership rate from 65% to 70%.²⁸ Subsidized credit and tax preferences are capitalized into prices where supply is constrained. That is a claim about specific programs interacting with specific markets, and it is the supply elasticity that determines the size of the effect.

Mortgage lock-in

The July 2026 Monetary Policy Report states that “the majority of outstanding mortgages still have interest rates below 4 percent, substantially lower than the prevailing 30-year fixed interest rate of 6.4 percent,” and names “rate lock” as a factor holding down home sales.²⁹ Fed economists estimated in a 2025 paper that lock-in explained 44% of the drop in mortgage-borrower mobility from 2021 to 2022, and that in the 2022 market the lock-in shock reduced time on market by 29% and raised house prices by 8%. They also found that in a balanced market like 2019 the same shock “would have had little to no impact on prices or tightness.”³⁰

A May 2026 NBER paper by Fonseca, Liu and Mabille models the general equilibrium. Existing-home sales fell 40% between 2022 and 2024. Lock-in cuts both supply, through owners who do not sell, and demand, through owners who do not buy elsewhere, and the net effect is higher demand because “missing downsizers stay in larger homes, particularly in expensive areas.” A temporary rate hike that causes lock-in raises aggregate house prices 4.4% and rents 1.5% relative to a counterfactual without lock-in.³¹ That is a real channel from higher rates to higher rents, and a large one.

Buyers who stay renters

An IMF working paper by De Stefani, using property-level data from the American Housing Survey, found that the 2021 to 2023 rise in mortgage rates pushed many prospective first-time buyers above FHA payment-to-income limits. Cities with larger shares of constrained buyers saw steeper rent growth, concentrated in smaller units occupied by lower-income renters.³² A February 2026 Bank of Canada paper by Rao and Wang used identified monetary shocks on Canadian data from 1997 to 2023 and found that a 100 basis point rise in mortgage rates lowers house prices by 5% after one year and 10% after two, while CPI rent rises 2% to 3% after one year and 5% to 6% after two, with the rent estimates less precise. The authors attribute the rent increase to higher landlord user costs and greater relative demand for renting.³³ The chain is direct: mortgage rates up, ownership less attainable, rental demand up, rents up.

Why rents still fall when the Fed tightens

Every mechanism above is a partial effect, estimated relative to a counterfactual that holds the rest of the economy fixed. The question for CPI is what happens to rents when everything moves at once.

Liu and Pepper at the San Francisco Fed answered it for U.S. data from 1988 to 2019 using high-frequency policy surprises. A tightening equivalent to a one-percentage-point increase in the funds rate lowers rent inflation about 0.6 percentage point on impact and about 3.2 percentage points over the following ten quarters, which translates to a maximum reduction in headline PCE inflation of about 0.5 point.³⁵ Higher rates reduce employment, income and household formation, and those forces on rental demand outweigh the construction and tenure effects over a two to three year horizon.

The lock-in paper says the same thing from inside its own model. The 4.4% price effect is measured against a world with higher rates and no lock-in, and in the authors’ words it offsets “a third of the aggregate house price decline caused by higher rates.”³¹ Two thirds of the decline remains. The Canadian estimates show house prices falling 10% while rents rise 5%, which is a shift in the price-to-rent ratio inside a housing market that higher rates made cheaper to own and dearer to lease.³³

Higher rates can raise rents through construction, lock-in and tenure substitution, and those effects are largest when supply is inelastic and the market starts tight. They partially offset, and can for a time exceed, the downward pressure on rents from weaker demand. Over the horizon monetary policy works on, the demand effect has dominated.

When theory says a rate hike raises inflation

Two bodies of theory produce the result, and each depends on a condition that can be checked against the current hike.

The neo-Fisherian argument starts from the Fisher equation and asks what happens if the central bank moves the nominal rate to a permanently higher level. If the long-run real rate is pinned down by saving and investment, a permanently higher nominal rate can only be consistent with permanently higher inflation. Uribe estimated both an empirical model and a New Keynesian model on postwar data and found that “temporary increases in the nominal interest rate are estimated to cause decreases in inflation and output,” while permanent increases raise inflation and output in the short run.³⁶ The distinction between temporary and permanent is the whole result. Garín, Lester and Sims showed that the textbook model is neo-Fisherian only when the change is persistent and prices are flexible, and that “a modest and empirically realistic fraction of ‘rule of thumb’ price-setters may altogether eliminate” it.³⁷ García-Schmidt and Woodford argued that the result depends on perfect-foresight expectations that people are unlikely to hold.³⁸ A quarter-point hike that markets expect to be reversed within two years is the temporary case in every one of these models.

The fiscal argument is older. Sargent and Wallace showed in 1981 that if the fiscal authority will not adjust, so that debt must eventually be financed by money creation, then “although fighting current inflation with tight monetary policy works temporarily, it eventually leads to higher inflation,” because the bonds sold to fight inflation must later be serviced with seigniorage.³⁹ Cochrane makes the modern version: “each percentage point of higher real interest rates, undertaken to fight inflation, raises interest costs by 1%, and at 100% debt-to-GDP ratio raises interest costs by 1% of GDP. If Congress does not provide those extra funds, that’s an inflationary pressure.”⁴⁰ This is a coherent framework and a live debate at current U.S. debt levels. It is also a statement about fiscal policy, and its conclusion holds only if the Treasury never pays the higher interest bill with taxes or spending cuts.

There is one recent natural experiment in the mirror-image claim, that cutting rates lowers inflation. Turkey’s central bank cut its policy rate to 8.5% by February 2023, four months after annual CPI inflation had peaked at 85.5%.⁴¹⁴² After the May 2023 election the bank reversed course and raised the rate in steps to 50% by March 2024, with annual inflation still running at 67% that February.⁴³ One country with a depreciating currency is not a controlled trial, but it is what the prescription looks like when a government tries it with unanchored expectations.

AI and the neutral rate

The first step of Ackman’s argument, that AI investment is unusually insensitive to rates, is the part with the most support at the Fed. Governor Cook said in February 2026 that “we already see soaring AI-related business investment in data centers and chips, despite interest rates broadly being elevated relative to levels over the past 20 years,” and that “it is possible that the current neutral rate is higher than before the pandemic.”⁴⁴ Vice Chair Jefferson laid out the mechanism in July: “If AI leads to permanently higher levels of productivity growth, it may increase firms’ desire to invest and, hence, their demand for funding. Higher productivity growth may also discourage household savings by increasing expected future income. Under these circumstances, to reconcile the increase in investment with reduced savings, r* would likely rise,” with the caveat that the relationship between productivity growth and real rates is “historically noisy.”⁴⁵

That argument is about the neutral rate, the real rate at which policy neither stimulates nor restrains. If AI raises it from 1% to 2%, a given policy rate restrains less than it used to, and the Fed has to set rates higher to achieve the same effect. What the argument does not do is reverse the sign. A hike still moves policy toward restraint from wherever it starts. The claim that the hike raises inflation needs the cost channel or the fiscal channel to outrun the demand channel, and the estimates above say they do not.

The arithmetic of the current stance is worth stating. The New York Fed’s Holston-Laubach-Williams estimate of the U.S. neutral real rate is 1.01% for the second quarter of 2026.⁴⁶ The FOMC’s September projections put the longer-run nominal funds rate at a median of 3.2%, which with a 2% target implies a neutral real rate near 1.2%.⁴⁷ The new target range midpoint of 3.875% against 2.4% core CPI is a real policy rate of about 1.5%. On those yardsticks the September hike moved policy from close to neutral to modestly restrictive. Ackman’s argument implies the neutral rate is much higher than either estimate, in which case the hike restrains even less, which is an argument for more hikes rather than fewer.

The rate-insensitivity premise is also weakening on its own terms. Dallas Fed researchers estimated in February 2026 that roughly $500 to $600 billion of hyperscaler investment since 2023 had been funded internally from retained earnings, and that those firms “have begun turning more recently to public and private debt markets.”⁴⁸ Chair Warsh named that shift at the September press conference as one reason long yields have risen: “the so-called hyperscalers are out in the market raising funding. And so the competition for capital is real.”⁴⁹ Investment financed from cash flow ignores the bond market. Investment financed in the bond market does not.

What this means for DIY investors and households

None of this is a forecast, and the practical lessons hold whichever way the current cycle goes.

• Ask why the rate moved before asking what it means. A nominal yield can rise because expected inflation rose, because the real rate rose, or because the neutral rate itself moved. Those have different consequences for stocks, nominal bonds and TIPS, and only the first is a story about inflation. The real-yield decomposition of the past five years shows nearly all of the ten-year’s rise came from the real leg.
• Your mortgage payment and the CPI are different numbers. If you are buying, a rate increase raises your cost of housing now, whatever the shelter index does. If you already hold a sub-4% fixed-rate mortgage, lock-in is your asset: you are paying yesterday’s rate on a house whose rent equivalent rises with inflation.
• Renters can expect the housing channels to bite first. Construction pauses, lock-in and priced-out buyers all push rents up in the first year or two of a tightening, in tight inelastic markets most of all, before the demand effect brings rent inflation down.
• Do not build a plan on either side of this argument. The disagreement between Ackman and the FOMC is about the size and speed of transmission in one unusual cycle. A portfolio that needs that argument resolved a particular way is taking a bet the forecasting record does not support.
• Real yields near 3% are a price, whichever theory wins. A ten-year TIPS yield above its level for 99% of sessions since 2003 is a fact about what a risk-free real return costs today. It does not require a view on whether the Fed’s model is broken.

What Summitward recommends

Treat “higher rates cause inflation” as a claim with a narrow domain. It holds for rents in the short run in inelastic markets, for firms with heavy working-capital borrowing, and in theory for permanent rate changes or a fiscal authority that refuses to pay its interest bill. It does not hold for aggregate CPI over the horizon the Fed operates on, and the 2022 to 2023 cycle, the one that was supposed to break the old models, ended with inflation falling after the hikes with a lag that the old models predicted.

For a household, separate the affordability of the house you want to buy from the inflation rate the Fed is targeting. Run the purchase at today’s 7% in the rent-versus-buy tool, and run it again at 6% and 8%, because the rate you can get will move more in the next two years than the shelter index will. For a portfolio, the real yield on offer matters more than the direction of the next FOMC decision, and the bond allocation should be sized to the plan rather than to a view on Ackman versus Warsh.

Key Takeaways

• Identified policy shocks lower inflation with a lag of one to three years. Romer and Romer, high-frequency studies, and the ECB and IMF assessments of the 2022 to 2023 cycle all point the same way.
• The 1970s co-movement of rates and inflation is the Fisher effect, plus the Fed responding to inflation. Neither is evidence that hikes raise prices.
• The cost channel is real and measured. In the ECB’s 2025 firm-level study, tightening surprises raised expected selling prices in proportion to working-capital exposure. The authors state this does not contradict the aggregate effect.
• Mortgage interest is not in the CPI. Owners’ equivalent rent is, at 25.9% of the index. A buyer’s payment can rise 8% in a year while measured shelter inflation slows.
• Higher rates can raise rents through three channels (construction, lock-in, priced-out buyers), with lock-in alone worth 4.4% on prices and 1.5% on rents in one 2026 model, relative to a no-lock-in counterfactual. Over 1988 to 2019, tightening still lowered U.S. rent inflation by up to 3.2 points.
• A higher neutral rate from AI changes how restrictive a given rate is. It does not change the sign of what a hike does. At 3.75% to 4.00% against a roughly 1% neutral real rate, policy is modestly restrictive by the Fed’s own estimates.

Frequently Asked Questions

Do higher interest rates cause inflation?

Over the horizon monetary policy works on, one to three years, higher policy rates lower aggregate inflation. Studies that isolate policy changes from the Fed’s reaction to the economy find that a one-point unexpected tightening lowers the price level by several percent after two to four years. Higher rates do raise some specific prices, most clearly rents in tight housing markets and the prices of firms that borrow heavily for working capital, but those effects are smaller than the demand effect in the aggregate data.

Why did interest rates and inflation rise together in the 1970s?

Because nominal interest rates contain expected inflation. When trend inflation rose through the decade, lenders demanded higher nominal yields to keep their real return, and the Fed raised its policy rate in response to the inflation it saw. Both are inflation moving rates, which is the long-run Fisher effect. Mishkin’s 1992 tests found the long-run relationship and no short-run one.

Is mortgage interest included in the CPI?

No. Since January 1983 the CPI-U has measured owner-occupied shelter with owners’ equivalent rent, an estimate of what the home would rent for. BLS treats the house itself as an investment good, so mortgage interest, property taxes, transaction fees and most maintenance are excluded. Before 1983 the index used a user-cost approach that did include an interest estimate, which made measured inflation move with mortgage rates.

Can rate hikes raise rents?

Yes, for a time and in some markets. Higher rates cut construction in supply-constrained cities, lock existing owners into low-rate mortgages so fewer homes come to market, and keep would-be first-time buyers renting. A 2026 NBER model puts the lock-in effect at 1.5% on rents relative to a no-lock-in counterfactual, and a Bank of Canada study finds CPI rent rising after identified mortgage-rate shocks. Over the full 1988 to 2019 U.S. record, though, a one-point tightening lowered rent inflation by up to 3.2 points within ten quarters.

What is the neo-Fisher effect?

The prediction that a permanent increase in the nominal policy rate raises inflation, because the long-run real rate is fixed by saving and investment and the Fisher equation then requires inflation to absorb the change. Uribe’s estimates support it for permanent changes and find the conventional disinflationary effect for temporary ones. A single hike that markets expect to reverse is a temporary change.

Does AI change how monetary policy works?

It may raise the neutral rate, if higher expected productivity lifts investment demand and lowers saving, which Fed officials have said is possible. A higher neutral rate means a given policy rate is less restrictive, so the Fed would need to set rates higher for the same effect. That changes the level of rates needed. It does not make a hike inflationary.

Related Guides

• The Ten-Year Rose 392 Basis Points Since 2021. The Real Yield Rose 395. for the decomposition of the current yield into real rate and inflation compensation, and why it matters which leg moved.
• Does the Fed Really Set Interest Rates? for what the Fed controls, what markets set, and where r-star fits.
• Will AI Lower Interest Rates? for the productivity and neutral-rate arguments in full, including the model that puts AI’s effect on r-star near a point.
• Is a 7% Mortgage High? for where this week’s rate sits in fifty-five years of Freddie Mac data.
• Is a 30-Year Fixed Mortgage an Inflation Hedge? for the household side of mortgage lock-in.
• Why Macroeconomic Forecasts Are Not an Investment Strategy for the record of forecasters, and why the Ackman-versus-FOMC question should not size your portfolio.
• Rent vs. Buy: A Financial Analysis Beyond the Monthly Payment for the decision that a 7% mortgage actually changes.

Sources

1. Federal Open Market Committee, statement of September 16, 2026. Source of the increase in the target range to 3-3/4 to 4 percent and the “inflation remains elevated” language. federalreserve.gov
2. Samuel O’Brient, “Billionaire investor Bill Ackman says the AI race may have broken the Fed’s inflation playbook,” Business Insider, September 25, 2026, quoting Ackman’s X post of September 24, 2026. finance.yahoo.com
3. Freddie Mac, Primary Mortgage Market Survey, week of September 24, 2026: 30-year fixed 7.03%, 6.95% the prior week, 6.30% a year earlier. freddiemac.com
4. Bureau of Labor Statistics, Consumer Price Index, August 2026, released September 11, 2026, and Table 2. Source of the 3.4% all items, 2.4% core, 3.0% shelter, 2.7% rent and 3.1% owners’ equivalent rent twelve-month changes, and the July 2026 relative importance figures of 35.343 (shelter), 25.918 (owners’ equivalent rent) and 7.735 (rent of primary residence). bls.gov
5. Federal Reserve Board, H.15 Selected Interest Rates, ten-year Treasury constant maturity (monthly averages), and BLS CPI-U, all items, not seasonally adjusted, retrieved through DBnomics. Annual averages computed from the monthly series. db.nomics.world
6. Frederic S. Mishkin, “Is the Fisher Effect for Real? A Reexamination of the Relationship Between Inflation and Interest Rates,” Journal of Monetary Economics 30(2), 1992, pp. 195–215; NBER Working Paper 3632. nber.org
7. Federal Reserve History, “Volcker’s Announcement of Anti-Inflation Measures” and “The Great Inflation.” Source of the 10.8% unemployment peak in late 1982 and the return of inflation to under 5% by the end of the 1981–82 recession. The 19.1% June 1981 effective funds rate is the monthly average from the H.15 series above. federalreservehistory.org
8. Stephen B. Reed, “One hundred years of price change: the Consumer Price Index and the American inflation experience,” Monthly Labor Review, April 2014. Source of the 14.8% twelve-month increase for March 1979 to March 1980. bls.gov
9. Christopher A. Sims, “Interpreting the macroeconomic time series facts: The effects of monetary policy,” European Economic Review 36(5), 1992, pp. 975–1000, and Martin Eichenbaum’s discussion in the same issue, pp. 1001–1011. Attribution per Michael S. Hanson, “The ‘Price Puzzle’ Reconsidered,” Journal of Monetary Economics, 2004. econpapers.repec.org
10. Christina D. Romer and David H. Romer, “A New Measure of Monetary Shocks: Derivation and Implications,” American Economic Review 94(4), September 2004, pp. 1055–1084. Source of the 1969–1996 sample, the 4.3% industrial production trough at months 22 to 27, the price-level path (unchanged for 22 months, 6% lower at 48 months), and the insignificant 0.3% early rise. aeaweb.org
11. Mark Gertler and Peter Karadi, “Monetary Policy Surprises, Credit Costs, and Economic Activity,” American Economic Journal: Macroeconomics 7(1), January 2015, pp. 44–76. aeaweb.org
12. Marek Jarociński and Peter Karadi, “Deconstructing Monetary Policy Surprises: The Role of Information Shocks,” American Economic Journal: Macroeconomics 12(2), April 2020, pp. 1–43. aeaweb.org
13. Board of Governors of the Federal Reserve System, “Monetary Policy: What Are Its Goals? How Does It Work?” Source of the quoted description of the transmission mechanism. federalreserve.gov
14. Matthieu Darracq-Pariès et al., “A model-based assessment of the macroeconomic impact of the ECB’s monetary policy tightening since December 2021,” ECB Economic Bulletin, Issue 3/2023. Covers the 350 basis points of increases through March 2023. ecb.europa.eu
15. Philip Barrett and Josef Platzer, “Has the Transmission of US Monetary Policy Changed Since 2022?” IMF Working Paper WP/24/129, June 2024. The 25% figure applies to February through July 2022; the paper carries the IMF’s standard disclaimer that it represents the authors’ views. imf.org
16. Marvin J. Barth III and Valerie A. Ramey, “The Cost Channel of Monetary Transmission,” NBER Macroeconomics Annual 2001, vol. 16; NBER Working Paper 7675. Samples are January 1959 to September 1979 and January 1983 to March 2000. nber.org
17. Federico Ravenna and Carl E. Walsh, “Optimal monetary policy with the cost channel,” Journal of Monetary Economics 53(2), March 2006, pp. 199–216. ideas.repec.org
18. Pau Rabanal, “The Cost Channel of Monetary Policy: Further Evidence for the United States and the Euro Area,” IMF Working Paper 03/149, 2003; published as “Does inflation increase after a monetary policy tightening? Answers based on an estimated DSGE model,” Journal of Economic Dynamics and Control 31(3), 2007, pp. 906–937. elibrary.imf.org
19. Ugo Albertazzi, Annalisa Ferrando, Sofia Gori and Judit Rariga, “The cost channel of monetary policy: evidence from euro area firm-level survey data,” ECB Working Paper No. 3097, August 2025. Source of the 10 basis point to 60 basis point average effect, the 4.5% baseline price expectation, the working-capital exposure result, and the “do not contradict” sentence in the non-technical summary. ecb.europa.eu
20. Bureau of Labor Statistics, “Measuring Price Change in the CPI: Rent and Rental Equivalence.” Source of the capital-good treatment, the excluded cost items, the definition of owners’ equivalent rent, and the six-month rent collection panels. bls.gov
21. Frank Ptacek and Darren Rippy, “Owners’ equivalent rent and the Consumer Price Index: 30 years and counting,” BLS Beyond the Numbers 2(14), May 2013. Source of the October 1981 announcement, the January 1983 effective date for CPI-U, and the description of the prior user-cost approach. bls.gov
22. Federal Open Market Committee, “Statement on Longer-Run Goals and Monetary Policy Strategy,” adopted January 24, 2012, reaffirmed January 27, 2026. federalreserve.gov
23. Adriana D. Kugler, “A View of the Housing Market and U.S. Economic Outlook,” speech at the Housing Partnership Network Symposium, July 17, 2025. Source of the 16% shelter share of the PCE basket and the 35% CPI-U figure in the footnote. federalreserve.gov
24. Federal Reserve Board, Beige Book, October 18, 2023, Eighth District (St. Louis) summary. federalreserve.gov
25. Joseph Gyourko and Raven Molloy, “Regulation and Housing Supply,” Handbook of Regional and Urban Economics, vol. 5, 2015, pp. 1289–1337; NBER Working Paper 20536. nber.org
26. Edward L. Glaeser, Joseph Gyourko and Raven E. Saks, “Urban growth and housing supply,” Journal of Economic Geography 6(1), 2006, pp. 71–89. nber.org
27. Giovanni Favara and Jean Imbs, “Credit Supply and the Price of Housing,” American Economic Review 105(3), 2015, pp. 958–992. aeaweb.org
28. Kamila Sommer and Paul Sullivan, “Implications of US Tax Policy for House Prices, Rents, and Homeownership,” American Economic Review 108(2), February 2018, pp. 241–274. The 4.2% price decline and 65% to 70% homeownership figures are from the Table 6 discussion. aeaweb.org
29. Board of Governors of the Federal Reserve System, Monetary Policy Report, July 10, 2026, Part 1. federalreserve.gov
30. Aditya Aladangady, Jacob Krimmel and Tess Scharlemann, “Locked In: Mobility, Market Tightness, and House Prices,” Finance and Economics Discussion Series 2024-088r1, Federal Reserve Board, November 2024, revised May 2025. federalreserve.gov
31. Julia Fonseca, Lu Liu and Pierre Mabille, “Unlocking Mortgage Lock-In: Equilibrium Effects in a Spatial Housing Ladder Model,” NBER Working Paper 35237, May 2026. nber.org
32. Alessia De Stefani, “Missing Home-Buyers and Rent Inflation: The Role of Interest Rates and Mortgage Underwriting Standards,” IMF Working Paper 2025/090, May 2025. imf.org
33. Nishaad Rao and Tao Wang, “Channels of Transmission: How Mortgage Rates Affect House Prices and Rents in Canada,” Bank of Canada Staff Analytical Paper 2026-2, February 2026. Magnitudes are from the instrumented VAR on 1997 to 2023 data. bankofcanada.ca
34. Augustus Kmetz, Schuyler Louie and John Mondragon, “Where Is Shelter Inflation Headed?” FRBSF Economic Letter 2023-19, August 7, 2023. frbsf.org
35. Zheng Liu and Mollie Pepper, “Can Monetary Policy Tame Rent Inflation?” FRBSF Economic Letter 2023-04, February 13, 2023. Local projections on Bauer-Swanson policy surprises, 1988 to 2019. frbsf.org
36. Martín Uribe, “The Neo-Fisher Effect: Econometric Evidence from Empirical and Optimizing Models,” American Economic Journal: Macroeconomics 14(3), July 2022, pp. 133–162; NBER Working Paper 25089. aeaweb.org
37. Julio Garín, Robert Lester and Eric Sims, “Raise Rates to Raise Inflation? Neo-Fisherianism in the New Keynesian Model,” Journal of Money, Credit and Banking 50(1), 2018, pp. 243–259; NBER Working Paper 22177. nber.org
38. Mariana García-Schmidt and Michael Woodford, “Are Low Interest Rates Deflationary? A Paradox of Perfect-Foresight Analysis,” American Economic Review 109(1), January 2019, pp. 86–120. aeaweb.org
39. Thomas J. Sargent and Neil Wallace, “Some Unpleasant Monetarist Arithmetic,” Federal Reserve Bank of Minneapolis Quarterly Review 5(3), Fall 1981. minneapolisfed.org
40. John H. Cochrane, “Monetary-Fiscal Interactions,” NBER Working Paper 34257, September 2025. nber.org
41. Republic of Türkiye Ministry of Treasury and Finance, reporting TurkStat, “Inflation: October 2022 Figures,” November 3, 2022: consumer prices up 85.51% year over year. hmb.gov.tr
42. Central Bank of the Republic of Türkiye, one-week repo rate history: 8.50% from February 24, 2023; 50.00% from March 22, 2024. tcmb.gov.tr
43. Fortune, “Turkey’s central bank hikes rates to 50%,” March 21, 2024: 500 basis point increase to 50%, consumer prices up 67% from a year earlier in February 2024. fortune.com
44. Lisa D. Cook, “Opening Remarks for the ‘AI and Productivity across the Economy’ Panel,” NABE Economic Policy Conference, February 24, 2026. federalreserve.gov
45. Philip N. Jefferson, “Navigating Economic Shocks: A Monetary Policymaker’s Perspective,” Stanford Institute for Economic Policy Research, July 16, 2026. federalreserve.gov
46. Federal Reserve Bank of New York, “Measuring the Natural Rate of Interest,” Holston-Laubach-Williams estimates, release of August 27, 2026, final data point 2026Q2, one-sided U.S. r* of 1.01%. The Laubach-Williams model in the same release reads 1.65%; press reports have mislabeled that figure as HLW. newyorkfed.org
47. Federal Open Market Committee, Summary of Economic Projections, September 16, 2026, Table 1: longer-run federal funds rate median 3.2%, central tendency 3.0 to 3.6. federalreserve.gov
48. Hugo De Vere, Srini Ramaswamy and Seth Searls, “How AI debt financing impacts duration supply and interest rates,” Federal Reserve Bank of Dallas, February 10, 2026. dallasfed.org
49. Kevin Warsh, press conference transcript, September 16, 2026. federalreserve.gov

Author disclosure

I have no relationship with Pershing Square, the Federal Reserve, or any institution cited. I hold equities and Treasury inflation-protected securities in my own accounts. Every figure above is taken from the cited source and was checked against it on September 26, 2026; the mortgage payment comparison is a straight amortization at the Freddie Mac rates for a $400,000 loan and would differ for any other loan. The FOMC target range, the mortgage rate, and the CPI figures are as of that date and will be stale within weeks. Nothing here is investment advice.