Equity Risk Premium Explained: Definition, Historical Data, Implied ERP, and How It Affects Valuation

The equity risk premium is one of the most important inputs in all of finance. It sits at the heart of every discounted cash flow model, every cost-of-equity estimate, and every CAPM calculation. Yet most retail investors have never heard of it, let alone understood how it is estimated or why it matters.

This guide explains what the equity risk premium is, how analysts measure it historically and from current market prices, what country risk premium adjustments are, and how changes in the ERP ripple through discount rates and stock valuations.

What Is the Equity Risk Premium?

The equity risk premium (ERP) is the additional return investors expect to earn from holding stocks rather than risk-free assets like government bonds. It represents the compensation the market collectively demands for bearing the uncertainty and volatility that comes with equity ownership.

Put differently: if you could earn 4.5% per year on a 10-year U.S. Treasury note with near-zero risk, you would only choose to hold stocks if you expected them to return something meaningfully higher. That 'something higher' is the equity risk premium.

The formal definition is simple:

Equity Risk Premium = Expected Return on Equities minus Risk-Free Rate

If investors expect the broad stock market to return 9.5% annually and the current risk-free rate is 4.5%, the implied ERP is 5.0%.

The ERP is not a fixed number. It fluctuates with market conditions, interest rates, economic uncertainty, and investor sentiment. A central challenge in finance is estimating it accurately enough to produce useful valuation outputs.

Why the ERP Matters So Much

Understanding the ERP abstractly is one thing. Understanding why a 1-percentage-point change in the ERP can shift a stock's fair value estimate by 15% to 25% is another.

The mechanism works through the discount rate. In a discounted cash flow model, future cash flows are discounted back to the present using the cost of equity (or WACC for enterprise valuation). The cost of equity is built directly from the ERP via the CAPM formula:

Cost of Equity = Risk-Free Rate + Beta x Equity Risk Premium

A higher ERP raises the cost of equity. A higher cost of equity raises the discount rate. A higher discount rate reduces the present value of every future dollar of earnings or free cash flow. The result: valuations fall.

The inverse is also true. When the ERP compresses, valuations expand. This is one reason why the prolonged low-rate, low-ERP environment of the 2010s produced such elevated market multiples: future cash flows were being discounted at historically low rates.

The CAPM Formula and Where ERP Fits

The Capital Asset Pricing Model provides the formal framework for using the ERP in cost-of-equity estimation:

E(R) = Rf + Beta x (Rm minus Rf)

Where:

• E(R) is the expected return on the asset
• Rf is the risk-free rate (typically the 10-year U.S. Treasury yield)
• Beta is the asset's sensitivity to market-wide movements
• (Rm minus Rf) is the equity risk premium

The ERP (Rm minus Rf) represents the expected excess return of the market over the risk-free rate. Beta then scales that premium up or down based on the individual stock's systematic risk.

A stock with a beta of 1.0 earns exactly the market ERP on top of the risk-free rate. A stock with a beta of 1.5 earns 1.5 times the ERP. A stock with a beta of 0.6 earns only 0.6 times the ERP.

Example with numbers:

Assume a risk-free rate of 4.5%, an ERP of 5.0%, and three stocks with different betas:

Now suppose the ERP rises to 6.0% (reflecting greater market fear or uncertainty). The high-beta technology stock's cost of equity jumps from 12.0% to 13.5%. Discounting the same cash flow stream at 13.5% instead of 12.0% produces a materially lower fair value estimate.

The Historical Equity Risk Premium

The oldest and most common method for estimating the ERP is to look at what stocks have actually returned above government bonds over long historical periods.

The most widely cited historical ERP data comes from Aswath Damodaran, a professor of finance at NYU Stern School of Business. Damodaran maintains publicly available datasets going back to 1928 for U.S. markets, updated annually. His data tracks the annual returns on stocks, T-bills, and T-bonds, and computes the excess return of equities on both arithmetic and geometric average bases.

Key Findings from Historical Data

Using U.S. data from 1928 onward:

• Arithmetic average ERP over T-bills: approximately 6.3% to 6.6%
• Geometric average ERP over T-bills: approximately 4.4% to 4.8%
• Arithmetic average ERP over T-bonds: approximately 4.2% to 4.6%
• Geometric average ERP over T-bonds: approximately 2.8% to 3.4%

The range is wide because the answer depends on which starting year you use, whether you compare against T-bills or T-bonds, and whether you average arithmetically or geometrically.

Arithmetic vs. Geometric Averaging

This distinction matters significantly and is a frequent source of confusion.

The arithmetic average is a simple average of annual excess returns. If stocks beat bonds by 8%, -2%, 12%, and 3% over four years, the arithmetic average is 5.25%.

The geometric average compounds the returns. In the same example, the geometric average (computed as the annualized return of the overall investment over the period) would be lower because negative years have a disproportionate drag on compounding.

For forward-looking valuation purposes, there is genuine debate about which to use. Those building a DCF model argue for the arithmetic average because the discount rate in a DCF is applied one period at a time. Others argue for the geometric average because it reflects the actual compounded wealth accumulation of a long-term equity holder.

In practice, many valuation analysts use an ERP in the range of 4.5% to 6.0% for U.S. equities, representing a blend that reflects both the long-run data and some judgment about current conditions.

The Long-Run U.S. Average: 4% to 6%

Across multiple data sources and methodological choices, the long-run historical U.S. ERP converges on a range of roughly 4% to 6%. This is the 'stylized fact' that shows up in finance textbooks, CFA Institute curriculum materials, and most institutional valuation models.

The 5.0% midpoint is a common practical assumption. Analysts who believe markets are currently overvalued might shade the ERP higher (implying a higher required return). Those who believe current valuations are fair relative to rates and growth expectations might shade it lower.

The Implied Equity Risk Premium

The historical approach looks backward. The implied ERP looks forward, deriving the premium that current market prices and earnings expectations imply about investor required returns.

The concept is simple: if you know today's market price level, expected earnings or dividends, and a reasonable assumption about long-term growth, you can solve for the discount rate (and thus the ERP) that makes the math work.

How Damodaran Calculates the Implied ERP

Damodaran publishes monthly implied ERP estimates for the U.S. market. His approach is based on a dividend discount model applied to the S&P 500 index:

1. Start with the current level of the S&P 500
2. Estimate the expected cash flows to equity holders (dividends plus buybacks) for the next several years using analyst consensus earnings estimates
3. Assume a stable long-run growth rate for cash flows in the terminal period (typically anchored to expected nominal GDP growth)
4. Solve for the discount rate that sets the present value of all those cash flows equal to the current index level
5. Subtract the current risk-free rate from that solved discount rate to get the implied ERP

The implied ERP fluctuates month to month with market prices and changes in consensus earnings estimates. It can move quickly: when markets fall sharply without a corresponding decline in earnings estimates, the implied ERP rises (stocks are cheaper relative to fundamentals). When markets rise without earnings keeping pace, the implied ERP compresses.

Historical Range of the Implied ERP

Damodaran's implied ERP estimates show considerable variation over time:

• During periods of market stress (2008, early 2020), the implied ERP spiked above 6% to 8% as prices dropped and fear elevated required returns
• During periods of market euphoria (late 1990s dot-com peak), the implied ERP compressed below 2%, suggesting investors were accepting very little premium for equity risk
• In more 'normal' conditions, the implied ERP has typically ranged between 4% and 6% for U.S. markets

As of early 2026, the implied ERP for U.S. equities has been in the range of 4.0% to 5.5%, reflecting elevated risk-free rates (the 10-year Treasury has been in the 4.0% to 5.0% range) and broadly high market valuations. The higher risk-free rate has been partially offset by higher nominal earnings growth expectations, keeping the implied ERP within its historical norms.

Historical vs. Implied ERP: Which Should You Use?

This is an active debate among practitioners. Each approach has genuine strengths and weaknesses.

Arguments for using the historical ERP:

The long-run average reflects actual investor experience over many market cycles, recessions, bubbles, and crises. It is not influenced by current market sentiment or near-term earnings estimates. It is stable and consistent across time periods.

Arguments for using the implied ERP:

The historical average is backward-looking. It says nothing about what the current market is pricing in. Using a historical ERP of 5.0% when the market's implied ERP is 3.5% or 7.0% can produce systematic valuation errors. The implied ERP reflects current conditions.

The practical resolution:

Most sophisticated practitioners use the implied ERP as the baseline for current valuations while using the historical range as a sanity check. If the implied ERP is well below its historical average, it suggests equities are relatively expensive. If the implied ERP is well above the historical average, it suggests equities are relatively cheap.

Damodaran, whose work is widely referenced in institutional finance, publishes both historical data and monthly implied ERP estimates. His January 2025 estimate for the U.S. implied ERP was approximately 4.4%, reflecting the post-rate-hike interest rate environment and elevated market prices.

Country Risk Premium: Adjusting ERP for Non-U.S. Markets

When valuing companies that operate in or are domiciled in countries with higher political, economic, or currency risk than the U.S., the base ERP must be adjusted upward. This additional adjustment is called the country risk premium (CRP).

The logic is straightforward: holding equities in an emerging market with currency volatility, potential nationalization risk, weak rule of law, and underdeveloped capital markets requires more compensation than holding U.S. equities. A purely U.S.-derived ERP would understate the required return and therefore overstate fair value.

How Country Risk Premium Is Estimated

The most common method uses sovereign credit default swap (CDS) spreads or sovereign bond yield spreads to quantify how much riskier a country's government debt is compared to U.S. Treasuries. That spread is then scaled by an equity-to-bond volatility ratio (typically around 1.5x) to convert the debt risk premium into an equity risk premium.

Damodaran publishes country-level ERPs that combine the U.S. base ERP with country-specific risk adjustments, updated annually. As of early 2025, representative total ERPs including country risk:

• United States: approximately 4.4% to 5.0%
• Germany / Western Europe: approximately 4.8% to 5.5%
• Brazil: approximately 7% to 9%
• India: approximately 7% to 8%
• Argentina: 15% or higher during periods of economic stress
• China: approximately 6% to 7.5%

For a U.S.-listed company with purely domestic operations, the country risk premium adjustment is zero. For a multinational, analysts may blend ERPs proportional to each region's revenue contribution.

ERP During Recessions vs. Economic Expansions

The equity risk premium is not static across the business cycle. It rises during recessions and periods of financial stress, and falls (or compresses) during expansions and periods of investor confidence. Understanding this cyclicality helps explain why valuations behave the way they do across cycles.

During Recessions

Recessions bring multiple forces that push the implied ERP upward:

Falling earnings estimates: Analyst consensus downgrades reduce expected cash flows. With market prices not always falling by the same proportion, the implied ERP can move in complex ways.

Wider credit spreads: As corporate credit markets tighten, the overall cost of capital rises. This elevated risk perception flows through into equity risk assessments.

Increased uncertainty: When economic outcomes are genuinely uncertain, investors demand more compensation for bearing that uncertainty. The 'fear premium' embedded in required returns rises.

Flight to safety: Capital flows toward government bonds, compressing bond yields and (initially) pushing risk-free rates down while equity required returns stay elevated or rise. The spread widens.

During the 2008 to 2009 financial crisis, the implied ERP spiked to 6% to 8% as markets sold off sharply. During the early stages of the COVID-19 shock in March 2020, it briefly spiked above 7%.

During Expansions

Extended economic expansions tend to compress the ERP. Several mechanisms operate simultaneously:

Rising earnings: Strong and predictable earnings growth reduces uncertainty. With less uncertainty, investors accept a lower compensation premium.

Narrowing credit spreads: Easy credit conditions signal low systemic risk. Equity risk perceptions tend to follow credit conditions.

Momentum and overconfidence: Extended bull markets can generate overconfidence. Investors underestimate future volatility and accept lower required returns than fundamentals would justify. This is how ERP can compress to levels that look extreme in hindsight.

The late 1990s is the canonical example. The implied ERP compressed below 2% as investors extrapolated extraordinary technology growth rates indefinitely. The subsequent correction reflected a violent mean reversion in required returns, not just a collapse in earnings expectations.

How Higher ERP Raises Discount Rates and Lowers Valuations

The valuation mathematics of the ERP is worth working through in detail, because the sensitivity is significant.

Consider a simplified example. A company is expected to generate $10 in free cash flow per share next year, growing at 4% per year indefinitely. This is a Gordon Growth Model setup:

Fair Value = FCF next year / (Discount Rate minus Growth Rate)

Scenario A: ERP = 4.5%, Beta = 1.0, Risk-Free Rate = 4.5%

Cost of equity = 4.5% + 1.0 x 4.5% = 9.0%

Fair Value = 10 / (0.09 minus 0.04) = 10 / 0.05 = $200

Scenario B: ERP = 6.0%, Beta = 1.0, Risk-Free Rate = 4.5%

Cost of equity = 4.5% + 1.0 x 6.0% = 10.5%

Fair Value = 10 / (0.105 minus 0.04) = 10 / 0.065 = $154

Simply raising the ERP from 4.5% to 6.0% reduces the model fair value from $200 to $154, a 23% decline, with no change to cash flows, no change to growth, and no change to the risk-free rate.

This amplification effect is why ERP assumptions deserve serious attention. For high-growth companies where the terminal value represents 70% to 90% of total DCF value, the sensitivity is even more extreme. A half-percentage-point change in the discount rate can move fair value by 10% to 15%.

The practical implication: when the ERP is elevated (recessions, crises), fair value estimates under standard models look low relative to market prices that may reflect distressed selling. When the ERP is compressed (late-cycle expansions), models can look optimistic if they use a historically normal ERP while the market is pricing in something lower.

Practical Use of ERP in WACC and DCF Models

For self-directed investors building DCF models or using valuation platforms, the ERP enters the calculation at the WACC level.

The standard WACC formula is:

WACC = (E/V x Re) + (D/V x Rd x (1 minus T))

Where Re (cost of equity) = Rf + Beta x ERP from CAPM.

WACC then serves as the discount rate applied to projected free cash flows to estimate enterprise value. Subtracting net debt from enterprise value gives an equity value estimate, which divided by shares outstanding produces a per-share model fair value.

Key practical notes for using ERP in valuation:

Use a current risk-free rate. The 10-year Treasury yield is the standard proxy, updated daily. Using a stale rate from a year-old model can introduce meaningful errors. As of mid-2026, the 10-year yield has been in the 4.0% to 5.0% range.

Be consistent with your ERP source. If you use Damodaran's implied ERP, use his current estimate. If you use a historical average, use a well-established source and document your choice.

Test sensitivity. Run the DCF at the low and high end of a reasonable ERP range (say, 4.0% to 6.0%) to understand how much the fair value moves. A wide range signals a highly sensitive output.

Match beta estimation to the business. Beta captures the systematic risk component that scales the ERP for an individual stock. A mature consumer staples company may have a beta of 0.5; a speculative technology company may have a beta of 1.8. Using a market-average beta for both would misrepresent their very different risk profiles.

Damodaran's Monthly ERP Estimates

For anyone doing serious valuation work, Damodaran's website (pages.stern.nyu.edu/~adamodar) is an indispensable resource. He publishes:

• Monthly implied ERP estimates for the S&P 500, updated at the start of each month
• Country risk premium data updated annually, covering 150-plus countries
• Historical ERP data going back to 1928
• Spreadsheet models for calculating implied ERP from first principles

The monthly ERP estimates are particularly useful because they capture how the market's collective risk pricing changes in response to economic data, Fed policy, geopolitical events, and earnings revisions. Tracking the implied ERP over time provides a real-time barometer of market risk appetite.

Valuation platforms that incorporate live risk-free rates and current ERP estimates produce discount rate inputs that reflect today's market conditions rather than static assumptions from a prior period.

The Debate: Historical vs. Forward-Looking ERP

The academic and practitioner debate over which ERP estimate to use is long-running and unlikely to be definitively resolved. Both camps have substantive arguments.

Proponents of historical ERP argue that long-run averages smooth out transient noise and reflect a stable underlying investor risk preference. The U.S. 4% to 6% long-run range has been remarkably persistent across different interest rate regimes and economic epochs. Forecasts built on historical averages are robust and not subject to current-period bias.

Proponents of implied ERP argue that valuations should be grounded in what the market is actually pricing today, not what it priced on average over the last century. Using a 5.5% historical ERP in an environment where the implied ERP is 3.5% produces systematic overestimation of fair value. Markets clear at current prices, not average historical prices.

The resolution in practice tends to involve using the implied ERP as the primary estimate, cross-referenced against the historical range to assess relative richness or cheapness. When implied ERP is well below the historical average, it is consistent with an overvalued market. When it is well above, it is consistent with a market offering relatively attractive long-term returns versus risk.

For multi-method valuation platforms, the most defensible approach is to apply the current implied ERP as the central estimate and run sensitivity scenarios using a range that brackets the historical average.

Key Takeaways

• The equity risk premium is the excess return stocks are expected to deliver above the risk-free rate. It compensates investors for bearing systematic market risk.
• The CAPM formula uses the ERP directly: Cost of Equity = Risk-Free Rate + Beta x ERP. Every cost-of-equity estimate is sensitive to the ERP assumption.
• Historical data (primarily from Damodaran's dataset) shows the long-run U.S. ERP has averaged approximately 4% to 6%, depending on the measurement period, averaging method, and comparison benchmark.
• The implied ERP is derived from current market prices and earnings estimates. It fluctuates with market conditions and provides a forward-looking estimate of what investors are currently demanding.
• Country risk premiums adjust the base ERP upward for markets with higher political, economic, or currency risk. Damodaran publishes annual country-level ERP data.
• ERP rises during recessions and periods of financial stress, and compresses during expansions and bull markets. This cyclicality drives much of the valuation multiple expansion and contraction investors observe over market cycles.
• A 1-percentage-point increase in the ERP can reduce a stock's model fair value estimate by 15% to 25% or more, particularly for long-duration growth companies where terminal value is a large share of total value.
• Damodaran publishes monthly implied ERP estimates for the U.S. market, providing a regularly updated benchmark for valuation practitioners.
• The practical debate between historical and implied ERP is real and unresolved. Most serious practitioners use the implied ERP as the primary estimate while cross-referencing against the historical range.

Understanding the equity risk premium is essential for anyone building or interpreting DCF models, WACC calculations, or multi-method valuation frameworks. It is the single input that most directly connects macroeconomic conditions, interest rates, and market sentiment to individual stock fair value estimates.

Analyze Any Stock with Current ERP Inputs

Equity Rank's valuation engine applies live risk-free rates and current ERP assumptions to its WACC and DCF calculations across 800-plus stocks. Every model fair value estimate reflects today's rate environment, not a static spreadsheet from a prior period. Explore how ERP and discount rate assumptions affect the fair value output for any ticker.

7-day free trial · Cancel anytime

Start your free trial at equity-rank.com