What is the equity risk premium (ERP)?
– The equity risk premium is the additional return investors expect to earn from holding equities instead of a “risk‑free” asset (typically U.S. Treasury securities). It compensates for the higher uncertainty, volatility, and potential loss associated with stocks.
– ERP is a forward‑looking concept but is most often estimated from historical returns or from models and surveys. Estimates vary by method, market, time frame and economic environment. (Source: Investopedia / Hilary Allison)
Key takeaways
– ERP = expected return on equities − risk‑free rate.
– It can be measured historically (realized ERP), implied from prices (forward‑looking), or obtained from surveys and component‑based (“building block”) methods.
– ERP is not a fixed number; it changes with market conditions, investor sentiment, interest rates, and country/market risk.
– Use multiple methods and sensitivity analysis when applying ERP in valuation or portfolio decisions. (Source: Investopedia)
Deep dive: why ERP matters
– Valuation: CAPM and many corporate valuation models use ERP to derive the expected return (discount rate) for equities.
– Asset allocation: ERP affects the relative attractiveness of stocks vs. bonds.
– Risk pricing: a higher ERP signals that investors demand greater compensation for equity risk (either because expected returns must be higher or because perceived risk is higher).
Common ways to calculate or estimate ERP (methods and examples)
1) Historical (realized) ERP
– Idea: use past stock returns minus past risk‑free returns over a chosen period.
– Strengths: observable and simple.
– Weaknesses: backward‑looking, sensitive to time period, subject to survivorship bias.
– Example (from Investopedia data): S&P 500 total return averaged 11.91% (2014–2023) vs. 3‑month T‑bill average 1.27% → realized historical ERP ≈ 10.64%.
2) Capital Asset Pricing Model (CAPM) formulation
– CAPM: Ra = Rf + βa (Rm − Rf).
– ERP for the market = (Rm − Rf). For an individual security, expected equity premium = βa × (Rm − Rf).
– Practical steps:
1. Choose Rf (e.g., current nominal yield on a Treasury or TIPS for a real rate).
2. Choose market ERP (see methods below).
3. Estimate beta (β) for the security vs. the chosen market index.
4. Compute expected return Ra = Rf + β (market ERP).
– Note: setting β = 1 gives the market’s expected excess return above Rf.
3) Dividend discount / Gordon Growth approach (implied expected return)
– Gordon Growth Model rearranged: expected return k = D1 / P0 + g, where D1 = next dividend, P0 = current price, g = expected long‑term growth.
– ERP = k − Rf.
– Strength: forward‑looking and price‑based. Weakness: relies on growth estimate and assumes constant growth.
4) Earnings‑yield approach
– Expected return ≈ E / P (inverse of P/E).
– ERP = Earnings yield − Rf.
– Simple and forward‑looking but ignores payout policy and valuations can be distorted by accounting differences.
5) Survey method
– Collect forecasts from investors, analysts, academics about expected average annual equity returns over a chosen horizon (e.g., 5–10 years).
– Compute survey average expected equity return and subtract current Rf to obtain ERP.
– Pros: captures professional expectations; cons: susceptible to sentiment and sampling bias.
6) Building block (risk component) approach
– Sum a risk‑free rate + premiums for specific risk types (business risk, financial leverage, liquidity, country risk, inflation/term risk, etc.).
– Example: Rf 3% + business risk 4% + financial risk 1% + liquidity premium 1% = expected equity return 9% → ERP = 6%.
– Useful for granular analysis of non‑U.S. markets or firms with distinctive risks.
7) Multifactor model adjustment: Fama‑French three‑factor (conceptual)
– Fama‑French expands CAPM by adding factors for size (SMB) and value (HML).
– Expected return is influenced by exposures to these factors; the market ERP remains foundational but additional premiums for size and value explain cross‑sectional returns.
– For valuation, these models adjust expected returns for systematic exposures beyond market beta.
Can ERP be negative?
– Yes. A negative ERP can occur in realized returns if stocks underperform risk‑free assets over the measured period (e.g., during major bear markets). Forward‑looking implied ERPs can be negative if investors expect low/negative future excess returns (rare but possible in extreme scenarios).
– Practical note: a persistently negative ERP would call into question whether the “risk‑free” comparator or market proxy is appropriate, or indicate extreme risk aversion or mispricing.
What does a high ERP mean?
– Investors are demanding larger compensation for equity risk. Interpretations:
– Market perceives higher risk (recession, geopolitical shock, high volatility).
– Stocks are cheaper relative to fundamentals (higher expected returns implied by lower prices).
– Interest rates or inflation expectations may have fallen (raising ERP for given equity expected returns).
– For corporate valuation, a higher ERP increases discount rates and lowers fair values.
What is the current equity risk premium?
– There is no single “current” ERP—estimates vary by method, time horizon and data source. Different practitioners use:
– Historical averages (long‑term realized ERP).
– Implied ERP from current prices (e.g., Gordon model or models published by practitioners such as Aswath Damodaran).
– Survey averages from market participants.
– Practical step: assess several ERP estimates (historical, implied, and survey) and run sensitivity analysis around a plausible range (e.g., ±1–3 percentage points) when valuing or allocating.
Practical step‑by‑step: how to select and apply an ERP for valuation
1. Define scope and horizon:
– Which market (U.S., emerging market)? What horizon (short/long term)?
2. Choose your risk‑free rate:
– Nominal: use Treasury yields matching the cash‑flow horizon (e.g., 10‑yr for long term).
– Real: use TIPS for real discounting.
3. Select ERP estimation method(s):
– For conservative/standard practice: combine implied ERP and long‑run historical ERP.
– For firm‑specific work: consider building block adjustments and country‑risk premiums.
4. Calculate expected market return:
– Implied: use D1/P0 + g or forecasted earnings yield.
5. Compute ERP = expected market return − Rf.
6. Apply to CAPM: expected equity return = Rf + β × ERP.
7. Sensitivity and scenario testing:
– Run valuations using a low, base, and high ERP to show range of outcomes.
8. Document assumptions and compare with external benchmarks (e.g., academic estimates, Damodaran’s implied ERPs, consensus surveys).
Important factors to consider in ERP calculations
– Time horizon: short windows give noisy estimates; long windows risk survivorship bias.
– Choice of risk‑free rate: matching maturities and nominal vs. real matters.
– Market proxy: which index (e.g., S&P 500) you use affects returns.
– Inflation and interest rates: higher rates often compress equity valuations and change implied ERP.
– Country and currency risk: add country premium for non‑U.S. markets.
– Liquidity and credit environment: illiquid markets or stressed credit spreads raise required premiums.
– Valuation adjustments: price levels matter — high valuations imply lower implied future returns and a lower implied ERP (all else equal).
Using surveys to determine ERP (practical tips)
– Use large, representative samples of professionals.
– Ask for horizon and market index specifics.
– Be cautious of sentiment bias during extreme market conditions.
– Combine survey results with model‑based and historical estimates.
Exploring the building block approach (practical steps)
– Identify relevant risk components for the market or firm: business, financial leverage, liquidity, country, currency, policy/regulatory, etc.
– Estimate a reasonable premium for each component (use historical decompositions, credit spreads, country CDS, or academic studies).
– Sum components + chosen Rf to form expected equity return; subtract Rf to get ERP.
– Document rationale for each component and test sensitivity.
Understanding the Fama‑French three‑factor model (brief practical note)
– Use when you want to explain cross‑sectional differences in stock returns beyond market beta.
– Expected return = Rf + βmarket × ERP + βSMB × SMBpremium + βHML × HMLpremium.
– SMB (small minus big) and HML (value minus growth) represent additional priced risks; include them when a firm’s exposures to size/value are material.
Real‑world applications of ERP
– Cost of equity in discounted cash flow (DCF) models.
– Strategic asset allocation and mean‑variance optimization.
– Setting hurdle rates for capital projects and M&A decisions.
– Country risk assessments and investment screening.
Example calculations (simple numerical examples)
– Historical realized ERP (example): 10‑yr realized market return 9.5% minus 10‑yr Treasury yield 2.0% → ERP ≈ 7.5%.
– Gordon implied ERP (example): S&P expected D1/P0 + g = 7.0%, Rf = 3.0% → ERP = 4.0%.
– CAPM for a stock (example): Rf = 3.0%, market ERP = 5.0%, β = 1.2 → cost of equity = 3% + 1.2×5% = 9.0%.
Can ERP estimates be relied on precisely?
– No single estimate is definitive. ERP is model‑dependent and sensitive to inputs. Best practice is to:
– Use multiple methods (historical, implied, survey, building block).
– Run sensitivity analyses.
– Be transparent about assumptions and horizon.
The bottom line
– The equity risk premium is central to pricing equity risk and making investment and valuation decisions. It is estimated using historical data, market prices (implied), surveys, or component models. Because it depends on assumptions (time‑horizon, risk‑free rate, market proxy, and method), practitioners should use multiple approaches, document assumptions, and perform sensitivity testing rather than relying on a single “magic” number. (Source: Investopedia / Hilary Allison)
Sources
– Investopedia, “Equity Risk Premium,” Hilary Allison. https://www.investopedia.com/terms/e/equityriskpremium.asp
If you want, I can:
– Compute an implied ERP for a specific market using current index dividends/earnings and a Treasury yield you provide.
– Show a sensitivity table for how valuation changes with different ERP values.