Inflation and Stocks: 100 Years of Data

One hundred years of equity returns under different inflation regimes produce a counterintuitive empirical pattern: stocks have historically suffered most precisely when many investors expect them to be a hedge.

The long-run real return on equities is positive — but the path through inflation regimes has been deeply non-linear. The 1970s decade produced cumulative real losses comparable to the worst deflationary episodes, despite continuous nominal price growth.

The empirical literature on equities and inflation is older and richer than the popular narrative. Fama and Schwert’s 1977 paper, Modigliani and Cohn’s 1979 inflation-illusion hypothesis, and the cross-country evidence assembled by Bekaert and Wang in 2010 all converge on a finding that defies the intuitive hedge story. The data, observed across regimes since 1900, is the subject of this article.

The long-run picture: positive real returns, non-linear path

U.S. equities, measured by Robert Shiller’s CAPE-adjusted dataset of S&P composite returns since 1900, have delivered a cumulative real return of approximately 6.5% per year over the full 125-year window. This is the canonical “equities-as-real-asset” data point: over very long horizons, equity ownership translates into real wealth growth that meaningfully exceeds inflation. This pattern is decoded in our analytical primer on inflation. The Dimson-Marsh-Staunton Triumph of the Optimists dataset extended this finding to 21 countries, with similar though somewhat lower long-run real returns (3.5-6.5% per year depending on the country and sample period).

The path through specific decades, however, has been deeply non-linear. The 1970s in the U.S. produced a cumulative real return of approximately -2.5% per year over the decade — sustained negative real wealth accumulation despite continuous nominal price growth. The 1980s, by contrast, produced cumulative real returns above 11% per year, driven by the collapse of inflation expectations and the corresponding multiple expansion. The 2021-2024 episode produced mixed outcomes: U.S. equities delivered modest real losses through 2022 followed by recovery in 2023-2024; European equities underperformed in real terms through the period.

The Fama-Schwert framework: why the hedge intuition fails

Modigliani and Cohn’s 1979 Financial Analysts Journal paper proposed a behavioural explanation for the negative short-horizon correlation: investors suffering from “inflation illusion” discount future nominal earnings using nominal interest rates without correctly adjusting expected nominal earnings growth upward by the inflation rate. The result is a systematic underpricing of equities during high-inflation regimes — even though the underlying real fundamentals (real earnings, real assets, real cash flows) may be relatively unchanged. Campbell and Vuolteenaho’s 2004 update of the framework, using post-1980 data, confirmed that the inflation-illusion mispricing persisted in modern data, though with diminished magnitude as central-bank credibility improved.

The 1970s decade: the cleanest natural experiment

The U.S. 1970s provide the best-documented case of equity underperformance during a sustained inflation regime. Headline CPI averaged approximately 7.4% per year over 1970-1979, while nominal S&P 500 total returns averaged roughly 5.9% per year — a cumulative real loss of approximately -1.4% per year over the decade, or roughly -13% in cumulative real wealth from start to end. The dispersion within the decade was dramatic: 1973-1974 produced a cumulative real loss exceeding 40% as inflation accelerated and equity multiples compressed. The recovery began only after the 1980-1982 Volcker tightening broke inflation expectations.

The mechanism is consistent with both the Fama-Schwert empirical finding and the Modigliani-Cohn behavioural explanation. Earnings nominal growth tracked inflation imperfectly (the corporate margin pressure documented in the analysis of pricing power and margin response); real interest rates rose as Federal Reserve policy tightened; and the equity discount factor expanded sharply as a result. The resulting cumulative real return was negative across the decade — directly contradicting the popular intuition that equities provide automatic inflation protection.

Sector and style dispersion within the equity asset class

The decade-level real return is an aggregate; the within-equity dispersion was substantial. Companies with concentrated pricing power, commodity exposure on the revenue side, or short-duration cash flows held real value better than the average. Companies with long-duration discounted cash flows (high-multiple growth names, high-debt firms with floating-rate exposure) underperformed. The 1970s sector dispersion in U.S. equities was approximately 20 percentage points per year between the best and worst quintiles by inflation sensitivity — a magnitude that recurred in attenuated form in the 2022 European data documented in the analysis of dividend stocks and pricing-power resilience.

The equity-style dimension has been studied separately. Value stocks (low P/E, low P/B) historically outperformed growth stocks during inflation episodes, consistent with the Modigliani-Cohn framework: the discount-rate sensitivity of long-duration cash flows is amplified during inflation regimes, compressing growth multiples more than value multiples. The 2022 episode partially confirmed the pattern, with value indices outperforming growth indices by 8-12 percentage points in the U.S. and Europe before the late-2023 reversal as inflation expectations re-anchored.

Cross-country evidence and the developed-versus-emerging split

Bekaert and Wang’s 2010 cross-country study examined the equity-inflation relationship across 30 economies. The finding was nuanced: the short-run negative correlation held across both developed and emerging markets, but the magnitude was larger in countries with less anchored inflation expectations and weaker central-bank credibility. Emerging-market equities have historically shown more violent real return swings during inflation episodes — both deeper drawdowns during accelerations and stronger recoveries during disinflations.

The advanced-economy 2021-2024 episode displayed milder dispersion than the 1970s precedent. Most major equity indices delivered cumulative real returns within a 0 to -10% range across the four-year window, recovering substantially by end-2024 as inflation expectations re-anchored. The interpretation across the literature is that improved central-bank credibility has compressed the magnitude of the inflation-illusion mispricing — the mechanism still operates, but with smaller real-return amplitude than in earlier decades, as the equity valuation framework under real-rate regimes formalises.

The horizon dependence of the hedge property

Boudoukh and Richardson’s 1993 finding — that the equity-inflation correlation flips sign at long holding periods — is the structural reconciliation between the empirical short-run negative correlation and the long-run positive real return. On 20-year and longer holding periods, equity real returns become approximately uncorrelated with realised inflation, with the long-run real return reflecting underlying productivity and real earnings growth more than the inflation regime. On 1-5 year holding periods, the correlation is sharply negative, with inflation surprises producing the largest real losses.

The implication, observed in the data rather than recommended, is that the equity hedge property requires a holding period longer than most investors plan for. Over a 30-year retirement horizon, the inflation-illusion mispricing of any given decade typically washes out; over a 5-year window centred on an inflation episode, the mispricing dominates. This is the empirical horizon dependence that the historical CAPE-versus-real-rate dataset documents continuously since 1900 — a feature of the data, not a tactical observation.

What the next decade implies

The dominant policy consensus expects inflation to revert to target by 2026-2027. If realised, the post-2024 environment becomes a disinflationary regime — historically the period during which equity real returns have been strongest, as multiples expand and inflation-illusion mispricing reverses. The 1980s and 1990s U.S. precedents illustrate the magnitude possible in such regimes; the 2010-2019 European experience illustrates the muted version under structurally low inflation. The dedicated treatment is in the inflation cycle decomposition. The empirical record suggests that the cumulative real return outcome over 2025-2030 will depend more on the path of real interest rates and the speed of expectations re-anchoring than on the inflation rate itself.

The structural exposures documented in the cluster — pricing-power dispersion, sector heterogeneity, the asymmetric incidence on long-duration discounted cash flows — will continue to operate in any future inflation episode. The empirical work since Fama-Schwert has not produced a robust mechanism to neutralise the short-horizon negative correlation. The horizon-dependence of the hedge property remains the most reliable structural feature of the equity-inflation relationship — observed across centuries, regimes, and countries.

The equity asset-class observations sit inside the wider system mapped in the complete guide to inflation mechanics, measurement and effects. They sit alongside the corporate margin response analysed earlier in the cluster and connect downward to the bond-market repricing that affects the equity discount factor through real-rate dynamics. The valuation arithmetic is anchored by the canonical real-versus-nominal returns framework, by the comparison of real returns across investment vehicles under different macro regimes, and by the relationship between money supply growth and equity returns. The cross-asset context is provided by the long-run CPI-adjusted gold price dataset and the real-return calculator for running the same arithmetic on specific portfolio assumptions. For the institutional context, the sub-pillar on equity market valuation under real-rate regimes situates the asset-class observation within the broader macro-financial framework.

Last updated — 7 May 2026