Fisher Equation

The identity behind every interest rate

If you put a thousand dollars in a savings account paying 4% nominal interest and prices over the year rise by 3%, you do not gain 4% in purchasing power. You gain roughly 1%. The Fisher equation is the way economists keep that fact straight. Written exactly,

(1 + i) = (1 + r) × (1 + π^e)

where i is the nominal interest rate quoted on a financial contract, r is the real rate — the inflation-adjusted return on goods rather than dollars — and π^e is the inflation rate that lender and borrower expect over the life of the contract. Expanding the right-hand side gives

i = r + π^e + r·π^e

and dropping the cross-term yields the approximation almost everyone uses informally:

i  ≈  r + π^e

The approximation is excellent at the low single-digit inflation rates typical of modern advanced economies — at r = 2% and π^e = 2% the cross-term is 0.04 percentage points — and grows worse the higher inflation runs. In Argentina at 80% inflation and a 5% real rate the cross-term is 4 percentage points, larger than the real rate itself. For most pedagogical and policy work the linear form is the working tool; serious bond-mathematics work uses the multiplicative form.

Fisher himself and the historical context

The decomposition is named for the American economist Irving Fisher, who introduced it in The Rate of Interest (1907) and developed it fully in The Theory of Interest (1930). Fisher was building on a centuries-old observation — debasement of currency was understood to drive up nominal yields well before him — but he was the first to write the identity cleanly and to argue that the real rate is the economically relevant quantity for intertemporal decisions. His timing was poetic: Fisher published the 1930 book just months before he famously declared that stocks had reached a "permanently high plateau," and the deflation of the early 1930s drove ex-post real rates sharply positive even as nominal rates fell — exactly the contractionary mechanism his equation predicts. The Great Depression made the Fisher decomposition unforgettable for the generation that came after.

Fisher's broader contribution was to put time and inflation at the centre of capital theory. The "real rate of interest" in his sense is the rate of intertemporal substitution between consumption today and consumption tomorrow, expressed in goods. Nominal contracts merely denominate that intertemporal trade in a unit of account whose value drifts. Modern asset pricing, from CAPM to consumption-based models, inherits this distinction.

Ex ante versus ex post — the role of expectations

A loan is signed today; inflation is realised tomorrow. The Fisher equation has two cousins that differ only in which inflation appears:

Form	Equation	Use	What is observable
Ex ante (decision)	i = r + πe	How rates get set at origination	i directly; πe only via surveys or TIPS
Ex post (accounting)	rreal = i − π	What return was actually earned	Both i and π directly
Long-run identity	Δi = Δπe, r constant	"Fisher effect" prediction	Tested by co-integration

The ex-ante form is the one that drives behaviour. Lenders demand i based on their best forecast of π^e; borrowers accept it for the same reason. When realised inflation π deviates from the π^e built into the contract, the ex-post real rate i − π differs from the agreed real rate r, and wealth is silently transferred between the parties. A surprise inflation hurts lenders and helps borrowers; a surprise disinflation does the opposite. The 1980s gave the canonical example of the latter: Volcker disinflation broke entrenched 10%+ inflation expectations, but long-dated bonds still carried the high nominal coupons priced in earlier, delivering enormous ex-post real returns to bondholders.

The Fisher effect and what the data say

Fisher's identity makes a strong long-run prediction known as the Fisher effect: a one-percentage-point permanent rise in expected inflation should raise nominal rates by exactly one percentage point, leaving the real rate unchanged. Empirically, this holds approximately on multi-decade horizons. Co-integration tests on US 3-month Treasury bills against ex-post CPI inflation since 1953 find a long-run pass-through coefficient between 0.8 and 1.0, depending on sub-sample. Cross-country panels show similar pass-through. The intuition is clean: real rates are pinned down by the deep parameters of the economy — productivity growth, time preference, demographics — and inflation enters nominal rates simply as a unit-of-account adjustment, not a real-resource shift.

Over short horizons the pass-through is far weaker and noisier. There are at least three reasons. First, π^e is unobservable and changes only slowly; nominal rates can move quickly on liquidity shocks or policy surprises while expectations lag. Second, real rates themselves are not actually constant — they cycle with business conditions, demographic shifts and global saving glut considerations. Third, monetary policy can deliberately drive a wedge between i and r + π^e over the medium term. The point estimates of the Fisher effect over five-year windows since 2000 are noticeably below 1; in the 2010s, with the zero lower bound binding for nominal rates, the relationship broke down completely.

TIPS and break-even inflation — markets price π^e in real time

Treasury Inflation-Protected Securities (TIPS), issued by the US Treasury since 1997, pay a fixed real coupon on a principal that is indexed to the headline CPI. Their yield is therefore a direct measure of the market-required real rate r. A standard nominal Treasury of identical maturity carries yield i. Inverting the Fisher equation gives

π^e ≈ i − r          (break-even inflation rate, BEI)

The break-even is what the marginal investor must expect inflation to be over the life of the bond, otherwise arbitrage would move money between TIPS and nominals. Both the Federal Reserve and the ECB watch break-evens as a real-time read on inflation expectations. Caveats: TIPS embed a small liquidity premium (TIPS markets are less deep than nominals), and break-evens include an inflation risk premium that compensates nominal holders for inflation uncertainty. The Federal Reserve's preferred adjusted measure, the "common inflation expectations" index, attempts to strip both out.

Worked example: TIPS break-even and a 30-year mortgage

Suppose on a Tuesday morning the 10-year nominal Treasury yields 4.20% and the 10-year TIPS yields 1.80%. The break-even is

BEI = i − r = 4.20% − 1.80% = 2.40%

The market is pricing in roughly 2.4% average annual inflation over the next decade — just above the Fed's 2% target. Now consider a household borrowing $400,000 on a 30-year fixed-rate mortgage at 6.5%. The market's π^e = 2.4% implies a real mortgage cost of

r_mortgage ≈ 6.5% − 2.4% = 4.1%

If inflation surprises to the upside — say it averages 4% over the life of the mortgage — the household's ex-post real rate falls to 6.5% − 4% = 2.5%, and the lender suffers a 1.6-percentage-point per year loss in real return over thirty years. The total real-dollar windfall to the borrower exceeds $100,000 in present-value terms. This is the silent mechanism by which inflation transferred wealth from US bondholders to homeowners over 1965–1980, and again — partially — in 2021–2022.

Negative real rates and financial repression

Throughout the 2010s and into the early 2020s, advanced-economy real rates were persistently negative. The mechanism is straightforward: nominal policy rates near zero (or below, in the euro area, Switzerland, Sweden, Denmark and Japan) combined with positive inflation produced an ex-post real rate below zero. US one-year ex-post real rates averaged roughly −1% from 2010 to 2019 and fell to about −7% during the 2022 inflation surge before policy tightening pushed them back up.

Negative real rates are an explicit lever of what economists call financial repression: by holding nominal rates below the inflation rate, governments transfer wealth from creditors (savers, pension funds, insurance companies) to debtors (themselves, mortgaged households, leveraged firms). The post-WWII era 1945–1980 is the textbook precedent — financial repression eroded WWII debt-to-GDP from over 100% to under 35% even as governments ran modest primary deficits. The Fisher equation is what lets you decompose how much of that erosion was inflation rather than nominal growth.

US 1970s — the great inflation as evidence

The 1970s remain the most dramatic test of the Fisher effect in modern advanced-economy data. US headline CPI inflation rose from 1.6% in 1965 to a peak of 13.5% in 1980 — driven by oil shocks, expansionary fiscal policy, and an accommodative Federal Reserve. The 10-year Treasury yield over the same period rose from 4.3% to 13.9%. The simple Fisher decomposition holds extraordinarily well: i roughly tracked the rise in inflation expectations one-for-one, leaving the real rate within a narrow band around 1.5–2.5% for most of the decade. The breakdown came in 1979–1982, when Paul Volcker pushed the federal funds rate to 19% — well above realised inflation — to break inflation expectations. Real rates spiked to 8%+ ex post, signalling that nominal rates were not just passively tracking inflation but actively setting it.

r-vs-g and sovereign debt sustainability

The Fisher decomposition feeds directly into the most important inequality in modern public finance. A government with debt-to-GDP ratio b, paying nominal yield i, facing inflation π and real growth g, sees its debt ratio evolve approximately as

Δb ≈ (r − g) × b − s

where r = i − π is the average real interest rate on outstanding debt and s is the primary surplus as a share of GDP. If r < g, the debt ratio shrinks automatically even at a balanced primary budget — fiscal grace. If r > g, the ratio explodes unless the primary surplus offsets the gap — fiscal pain. The Fisher identity is what lets you turn observed market nominal yields into the real r that enters this inequality. Olivier Blanchard's 2019 AEA presidential address brought the r-vs-g framing back into the mainstream, arguing that with r < g persisting for a decade, the social cost of public debt was lower than postwar economists had assumed.

Monetary policy transmission — and the expectations problem

Central banks set i, but the economy responds to r. Raising the policy rate by 25 basis points only tightens conditions if π^e does not also rise by 25 basis points. This is why modern central banks devote so much resource to anchoring expectations: independent central banks, numerical inflation targets, dot plots, forward guidance, press conferences. If π^e moves with i — what economists call "neo-Fisherian" dynamics in some models — then nominal policy is impotent. If π^e is well-anchored, every nominal move translates into a real-rate move, and policy bites.

The flip side is the zero lower bound. Nominal rates cannot fall much below zero (negative deposit rates do exist but face limits from cash hoarding). At the ZLB with low inflation expectations, the achievable real rate is also bounded below, even if a deeply negative r would be required to clear markets — Larry Summers's secular stagnation. The standard policy responses (forward guidance, quantitative easing, raising the inflation target, average-inflation targeting) are all attempts to manipulate π^e so that the achievable r at the ZLB is lower.

Variants and extensions

• International Fisher effect. Across countries, the difference in nominal rates between two currencies should equal the expected change in the exchange rate. Together with purchasing power parity, this gives the uncovered interest parity (UIP) condition. Empirically, UIP fails dramatically over short horizons — the "forward premium puzzle" of Hansen-Hodrick and Fama — but holds better over longer horizons.
• Mundell-Tobin effect. A theoretical refinement: higher expected inflation reduces real money balances, raising savings and depressing the real rate. The Fisher effect is then less than one-for-one. Empirically modest in modern data.
• Neo-Fisherian view. A heterodox claim, advanced by Cochrane, Schmitt-Grohé and Uribe, that the standard monetary-policy interpretation has the sign wrong at the ZLB: persistently low nominal rates may pin down low π^e, contributing to disinflation rather than fighting it. Controversial; partly drives the policy debate over how long to hold rates near zero.
• Term-structure decomposition. Modern affine-yield models (Cochrane-Piazzesi, ACM, Adrian-Crump-Moench) decompose the nominal yield curve into expected real rates, expected inflation, term premia, and inflation risk premia — a high-dimensional version of the Fisher decomposition.

Where the Fisher equation shows up

• Bond markets. The pricing of every nominal sovereign bond rests on the Fisher decomposition. Real yield curves built from TIPS, OATi (France), gilts linkers (UK) and JGBi (Japan) give traders a direct view on real rates.
• Mortgages and household finance. Fixed-rate mortgages are an implicit short position in inflation; floating-rate mortgages remove that hedge. Inflation surprises redistribute wealth between household borrowers and bondholders on a massive scale.
• Pension and insurance liabilities. Defined-benefit pension promises are partly real (CPI-linked) and partly nominal. The Fisher decomposition is what lets actuaries value the two streams consistently.
• Sovereign debt analysis. The IMF's debt sustainability assessments, the European Stability Mechanism's analyses, and every rating-agency model use the r-vs-g framework, with r derived through Fisher from observed nominal yields.
• Monetary policy decisions. Every modern Taylor-rule formulation, including the Fed's policy rule analyses, prescribes a nominal rate as the equilibrium real rate plus expected inflation plus reaction terms. The Fisher equation is the starting line.

Common pitfalls

• Confusing ex-ante with ex-post real rates. The decision-relevant rate is the one priced in at origination, not the one realised after the fact. Many policy errors come from imputing inflation surprises to deliberate policy choices.
• Using the linear approximation at high inflation. Above roughly 10% inflation the cross-term r·π^e becomes non-trivial. Argentina, Turkey and Venezuelan analyses require the multiplicative form.
• Reading TIPS break-evens as pure expectations. Break-evens include a liquidity premium (which lowers them) and an inflation risk premium (which raises them); they are not a clean read on π^e.
• Assuming the Fisher effect holds at all horizons. Long-run pass-through near one does not mean short-run pass-through near one. The 2010s, the ZLB, financial crises and hyperinflations are all known failure modes.
• Ignoring the term premium in long yields. The Fisher equation as written applies cleanest to short-dated debt. Long-dated yields embed term premia and convexity adjustments that are not captured by r + π^e alone.