What is the equity risk premium and how is it measured?

The short answer

The equity risk premium answers a simple but foundational question: how much extra return do investors require to hold stocks instead of Treasury bills? Without this premium, no rational saver would accept the volatility, drawdowns and uncertainty embedded in equity ownership.

Two distinct approaches exist for estimation. Historical ERP measures realized returns over multi-decade windows and subtracts the risk-free rate. Implied ERP works backward from current prices, using dividend discount or earnings-based models to extract the discount rate that justifies today’s valuations.

The two methods often disagree. Historical ERP averaged roughly 5-6% in US data since 1928, but implied ERP estimates today often print closer to 3-4%, reflecting compressed valuations relative to history. The gap between methodologies is itself a regime signal.

→ New to equities? Investing for beginners hub

What the data shows

Damodaran (NYU Stern, 2024) provides one of the most widely cited ERP datasets, drawing on FRED, Shiller and S&P data:

• Historical US ERP (1928-2023) averaged 4.84% over T-bonds and 6.62% over T-bills (geometric)
• Implied ERP for the S&P 500 was estimated at 4.60% in January 2024, near the 25-year median
• The implied ERP fell to a low near 1.5% in August 2000 at the dot-com peak
• It spiked above 6.5% in March 2009 during the GFC trough
• Cross-country ERP estimates (Dimson, Marsh, Staunton, 2024) show realized premiums averaging 3.5-5.5% across 23 markets since 1900

The exception worth noting: ex-post realized ERP and ex-ante required ERP are conceptually different. Realized ERP captures what investors got; required ERP captures what they demanded. Ibbotson and Chen (2003) document that during periods of multiple expansion, realized returns can exceed required returns by 2-3 percentage points annually.

→ Dataset: S&P 500 historical returns

Why it happens — the macro mechanism

The ERP exists because equity ownership exposes investors to risks bondholders avoid. Three structural channels generate and adjust this premium across cycles.

Cash flow uncertainty. Equity holders receive whatever remains after debt obligations are met. In recessions, earnings can collapse 30-50% (S&P 500 EPS fell 49% in 2008-2009). Bondholders contractually receive coupons regardless. The ERP compensates equity holders for this residual claim risk. Equity valuation details this mechanism.

Volatility and drawdown risk. The S&P 500 has experienced multiple peak-to-trough declines exceeding 40% in the modern era — 2000-2002, 2007-2009, 2020. Investors require ex-ante compensation for bearing these shocks. Realized volatility on equities (15-20% annualized) is several times that of Treasuries (4-6%). Market stress signals map this volatility to ERP shifts.

Behavioral and liquidity factors. Mehra and Prescott (1985) noted that historical ERPs are larger than rational models predict, the so-called “equity premium puzzle”. Subsequent work (Barro, 2006; Weitzman, 2007) attributes part of the premium to disaster risk and ambiguity aversion — investors charge an extra premium for tail risks they cannot fully model.

Synthesis by regime: in low-volatility, anchored-inflation environments, the implied ERP tends to compress as investors accept lower required returns; in high-uncertainty regimes, the ERP expands meaningfully.

The equity risk premium is the price of patience — what stocks must pay to compete with the certainty of bonds.

→ Framework: Equity markets pillar

What it means for different economic actors

Savers implicitly receive the ERP whenever they hold equities for the long run, even without computing it. The 4-6% historical premium compounds meaningfully over 30-40 year horizons, which is why retirement allocations have historically tilted toward equity exposure.

Investors use the ERP as a building block for asset allocation. Empirical research (Cochrane, 2011) documents that the implied ERP forecasts subsequent 5-10 year equity returns better than alternative valuation metrics in many sample periods, although the relationship is statistically noisy.

Pension funds and endowments use implied ERP as a regime signal for strategic allocation reviews. A compressed ERP near historical lows can prompt reductions in expected return assumptions — a critical input for funding ratios and contribution policies.

A common error is treating the historical ERP as a stable long-term forecast. The realized premium reflects past valuation expansions that may not repeat. Forward-looking implied ERP estimates often produce more conservative inputs for portfolio planning.

Practical observation

Go deeper

Related questions

• What does the CAPE ratio tell us?
• What drives stock returns in the long run?
• What is the earnings yield gap?
• Is the risk-free rate really risk-free?

Frequently asked questions

How does implied ERP differ from historical ERP?

Historical ERP looks backward, computing the average realized excess return of stocks over Treasuries across past decades. Implied ERP looks forward, using current prices and consensus earnings forecasts to solve for the discount rate that reconciles them. The historical figure tells you what investors received; the implied figure tells you what current valuations require investors to receive going forward. The two often diverge by 1-2 percentage points, with implied ERP typically more responsive to changing market conditions.

Why is there an equity premium puzzle?

Mehra and Prescott (1985) showed that standard utility-based asset pricing models predict an ERP of about 1%, far below the 5-6% observed historically. Resolving the puzzle has spawned a large literature pointing to disaster risk premia, habit-based preferences, ambiguity aversion, and long-run consumption risk. The puzzle remains an active research area because no single mechanism fully accounts for the empirical premium magnitude.

Is the ERP stable across countries?

Cross-country evidence (Dimson, Marsh, Staunton, 2024) documents realized ERPs averaging 3.5-5.5% across 23 developed markets since 1900, but with substantial variation. Markets with deeper financial development, longer reliable price histories and lower historical inflation tend toward the higher end. Markets that experienced major disruptions — wars, hyperinflations, regime changes — show distorted realized premia that may overstate or understate forward-looking required returns.

Last updated — 18 May 2026