One Policy Rate, Two Real Rates: Why the Inflation Measure Decides What December Tightening Means

Data as of November 13, 2026.

Three central banks set policy rates in December. The Federal Reserve decides on December 8 and 9, and both the European Central Bank and the Bank of England decide on December 17. Their nominal rates are 3.875 percent, 2.50 percent and 4.00 percent. Subtract the inflation measure each bank sets policy against and those settings produce real policy rates of 0.78 percent, minus 1.50 percent and 0.50 percent. That is a range of 2.28 percentage points where the nominal spread is 1.50, and each bank's own real rate moves by more than a point when a different price measure is subtracted. The measure, not the rate, is doing the work.

The three decisions on the calendar

The Federal Reserve has held its target range at 3.75 to 4 percent since October 28, when the vote was 10 to 2 and both dissents favored an immediate quarter-point increase. A quarter point is fully priced for December 8 and 9 after October consumer prices rose 0.4 percent on the month, a repricing traced in the November 11 mechanics analysis.

The European Central Bank has held its deposit facility rate at 2.50 percent since September 16, and its October flash estimate put euro area inflation at 4.0 percent. The Bank of England raised Bank Rate to 4.00 percent on November 5 by a vote of 8 to 1, with one member preferring 4.25 percent.

Three banks, three settings, and in nominal terms a spread of 1.50 percentage points from the ECB to the Bank of England. That spread tells you what an overnight deposit pays. It does not tell you what the rate is worth, because worth is decided by what prices did, and none of the three banks is measuring price change the same way.

One rate, two answers in the same office

Take the Bank of England's 4.00 percent and subtract each of the price measures sitting on the same desk. Against September CPI of 3.5 percent, the real rate is 0.50 percentage points positive. Against the third quarter GDP deflator of 2.3 percent, published with today's growth data, it is 1.70 points positive. Both figures come from the same statistical system, describe the same economy in overlapping months, and differ by 1.20 percentage points.

The reason is coverage. CPI tracks a fixed basket of consumer purchases and includes imports at their full price. The GDP deflator tracks everything produced domestically, including investment goods and government output, and excludes imports. When import prices rise faster than domestic prices, the deflator runs below CPI, and the real rate computed from it looks higher. That is precisely what has happened through the autumn energy move, which pushed euro area energy prices up 20.8 percent in the year to October.

The Bank of England said as much. Its November case rested on energy pushing CPI toward roughly 5 percent by mid-2027 while activity ran stronger than expected, the reasoning set out alongside the third quarter growth composition. A bank tightening against an import price shock is deliberately accepting a real rate gap: high against domestic price growth, lower against the basket households actually buy.

Why the United States cannot produce a headline real rate at all

The American case is harder. There is no published twelve-month CPI for October 2026, because October 2025 prices were never collected after the 2025 shutdown suspended the survey, and the bureau cancelled that release rather than reconstruct it. The aggregate annual comparison therefore has no base month, the mechanics of which are set out in the October CPI analysis.

Whatever annual rate is used this month has to be built by substitution. Compound the October monthlies and the range is wide. Core at 0.3 percent per month annualizes to 3.66 percent, which puts the 3.875 percent midpoint 0.22 points positive. Headline at 0.4 percent annualizes to 4.91 percent, which puts the same midpoint 1.04 points negative. The Federal Reserve's own preferred measure, core PCE, ran near 3.1 percent over twelve months, which puts the real rate near 0.78 points positive, the arithmetic carried through the four central banks comparison.

One policy rate, then, and a 1.82 point spread across three defensible measures. On the compounded headline the Fed is accommodative. On core PCE it is modestly restrictive. Both statements are arithmetically correct and neither can be published as "the" real rate.

The euro area runs negative on the measure it quotes

The ECB is the cleanest case because its governing council sets policy against headline HICP and says so. At 2.50 percent against 4.0 percent headline, the real deposit rate is 1.50 percentage points negative. Against core HICP of 2.5 percent, which excludes energy, food, alcohol and tobacco, it is exactly zero.

That zero is the number to hold. Euro area core inflation has sat inside a band of 2.4 to 2.6 percent every month since May, while energy contributed 1.88 of the 4.03 percentage points in the October flash, as the October inflation read worked through. A policy rate sitting exactly at core inflation means the bank has fully offset domestic price pressure and is absorbing none of the imported energy cost. The trade account carries the rest, with September's 10.3 billion euro goods deficit showing the transfer happening in the external balance rather than in the interest rate.

What the spread does to household money

The measure question is not academic. Take a saver with $10,000 in an account paying 3.875 percent, which yields $387.50 over a year before tax. Measured the simple way, interest earned minus the price measure, the real return is minus $103.50 against compounded headline inflation of 4.91 percent, plus $21.50 against compounded core at 3.66 percent, and plus $77.50 against core PCE at 3.1 percent. One account, one rate, and three results spanning $181, before tax and without deflating the balance.

The same arithmetic runs through a mortgage, with two extra layers. First, the rate a household pays is not the policy rate but the policy rate plus a spread, and that spread has been unusually wide: the 30-year fixed rate stood 219 basis points above the comparable Treasury yield at the end of October, the gap measured in the mortgage spread analysis. Second, the transmission is slow, running through the fixed-rate stock rather than new lending alone, which is the chain described in the rate hike transmission mechanism.

Those two layers explain why the measure gap matters more to a borrower than to a saver. A saver can switch price measures and reach a different conclusion about the same balance. A borrower cannot: the payment is fixed by the contract, and only the assessment of whether that payment is affordable moves with the index chosen. In all three economies the rate on the contract was set in nominal terms, while the judgment about whether it bites was set against a price series nobody wrote into the contract.

What the mismatch signals structurally

Three structural facts sit behind the arithmetic, and each is durable rather than cyclical.

The first is that inflation targets are not identical, so "positive real rate" cannot mean the same thing across borders. The Fed targets 2 percent on PCE. The ECB targets 2 percent on HICP. The Bank of England targets 2 percent on CPI. Hungary targets 3.0 percent on CPI with a plus or minus 1 point band, moving to 2.5 percent in 2028, and Poland targets 2.5 percent with the same width of band. A central bank with a 3 percent target holds a real rate roughly a point below one with a 2 percent target at identical nominal settings.

The second is indexation. The missing October 2025 base month propagates into every contract, benefit formula and bond that references an annual October comparison, and it will keep doing so for the life of those instruments, not just this month. Where a series cannot be produced, the substitution rule that replaces it becomes the operative policy, and it is usually set by whoever wrote the contract rather than by a statistical agency. The general mechanics of that reach into daily life are covered in why CPI is the number households feel.

The third is communication. A central bank that publishes a short statement on a decision date with no annual inflation figure available has one fewer anchor for its own justification. The Federal Reserve has already moved in this direction, cutting one vote from its 2027 calendar and leaving a 63-day gap between decisions, as the 2027 calendar analysis set out. Fewer decision points and a degraded annual base month both push the same way: more weight per release, and more interpretation left to the reader.

What to watch

Three dates resolve most of this. The October PCE release later in November establishes whether the Fed's preferred measure holds near 3.1 percent, which is what decides whether the December move is a tightening or a normalization. The euro area flash estimate for November, due at the end of the month, shows whether headline moves back toward core at 2.5 percent or holds near 4 percent, which decides whether the ECB's real rate stays at zero or turns negative again. And the December 17 Bank of England decision tests whether an 8 to 1 majority holds once the deflator at 2.3 percent is in hand.

The test that settles the question is narrower than it looks. A real rate only restrains spending if the price measure it is set against is the one driving the decisions households and firms actually make. A firm borrowing at 6 percent against a supplier price index rising 2 percent faces a different constraint from a household buying a basket rising 5 percent, and both are inside the same economy in the same quarter. December commentary that quotes a single real policy rate is quoting one of at least three defensible answers without saying which. The instrument to read instead is the pairing: rate against headline for purchasing power, rate against core for domestic pressure, rate against the deflator for the capacity to absorb an imported shock. In December, the three pairings disagree, and the disagreement is the information.