CAPM Assumptions and Intuition: Why Only Beta Gets Paid (And Your Stock Picker Does Not)

Ever built a beautifully diversified portfolio, stared at the efficient frontier, and thought, “Okay, but why does the market pay me for some risks and ghost me for others?” Welcome to CAPM assumptions and intuition — the model that takes our portfolio theory toolkit and says: if everyone did the rational thing (yes, we’re pretending), here’s how prices would line up.

The CAPM is the market’s way of saying: take unique risk if you want, but don’t expect a tip.

We’re building on your work with diversification, constraints, and resampling. You’ve stretched the efficient frontier, stress-tested weights, and cursed transaction costs. Now let’s flip to equilibrium: if everybody optimized like you, what would expected returns look like?

---

What Is the CAPM?

The Capital Asset Pricing Model (CAPM) is a one-factor equilibrium model that links an asset’s expected return to its systematic risk — the part you cannot diversify away because it moves with the market. In CAPM world:

• Investors hold the same tangency portfolio of risky assets — the market portfolio — then mix with the risk-free asset according to their risk tolerance.
• Assets earn higher expected returns only if they increase the portfolio’s exposure to market-wide risk, captured by beta (β).

Mathematically:

E[R_i] = r_f + β_i (E[R_M] - r_f)
β_i = Cov(R_i, R_M) / Var(R_M)

Where:

• E[R_i]: expected return of asset i
• r_f: risk-free rate
• E[R_M]: expected market return
• β_i: asset i’s sensitivity to the market

Graphically, this is the Security Market Line (SML) — expected return on the y-axis, beta on the x-axis. Everything priced right sits on that line. Above? Underpriced (too high a return for given beta). Below? Overpriced.

---

Assumptions of CAPM (Yes, We’re Doing the Dream-World Version First)

CAPM runs on a set of assumptions. Are they perfect? No. Are they useful? Shockingly, yes. Here’s the set:

1. Mean-variance investors: Everyone cares only about expected return and variance (or returns are normally distributed, or utility is quadratic — pick your cocktail). Translation: our Markowitz world is the right world.
2. Single-period horizon: Everyone plans over the same period (e.g., one year), then recalibrates. No long multi-stage chess while you’re playing checkers.
3. Homogeneous expectations: Investors agree on expected returns, variances, and covariances. No spicy disagreements about Tesla’s future. Just consensus.
4. Frictionless markets: No taxes, no transaction costs, unlimited shorting, infinitely divisible assets, price-taking. You can rebalance without crying.
5. Unlimited borrowing and lending at the risk-free rate: You can lever up the market portfolio like it’s leg day. Same rate for everyone.
6. All assets are tradeable and included: The market portfolio contains every risky asset, value-weighted — including global equities, bonds, real estate, commodities, maybe even the Mona Lisa if we’re being literal.
7. Markets clear competitively: Prices adjust so that supply equals demand. Nobody corners GameStop.

If these hold, two-fund separation hits: everyone holds a mix of the risk-free asset and the same risky tangency portfolio — the market.

---

How Does CAPM Work? (From Efficient Frontier to Market Pricing)

You already know how to build an efficient frontier and find a tangency portfolio with a risk-free asset. CAPM just says: if every investor does that using the same inputs, the tangency portfolio must be the value-weighted market. Why?

• Aggregation: If each investor is optimizing with the same covariance matrix and expected returns, their combined holdings must equal the market portfolio.
• Equilibrium: If some asset were underrepresented, its price would drop until expected return rises enough to pull it back onto the SML. Vice versa for overpriced assets.

The punchline: only covariance with the market (beta) affects equilibrium expected returns. Everything else is just noise you can diversify away.

---

Why Does Only Beta Get Paid? (The Intuition)

• Diversification nukes idiosyncratic risk: Remember how international diversification — Position 10, we see you — shredded country-specific drama? That same logic scales up: assemble a broad portfolio and the stock-specific chaos cancels out.
• What survives is systematic risk: recessions, rate shocks, oil spikes. You can’t diversify those away because they hit almost everything.
• Pricing is a team sport: If an asset adds risk that looks different from the market (low covariance), it helps diversification — so it does not need a big expected return. If it amplifies the market’s moods (high covariance), it makes the ride bumpier — so investors demand a higher return.

Analogy time: You’re on a group project (the market). If your contributions rise and fall exactly when the group is stressed (high beta), you make the rollercoaster worse — they’ll only keep you on the team if you promise higher payoff. If you’re the calm one who stabilizes chaos (low or negative beta), you’re prized for your vibes; you can accept a lower expected return.

---

Examples of CAPM in Everyday Investing

• Cost of equity in valuation: Analysts use CAPM to estimate discount rates: r_e = r_f + β × equity risk premium. That rate drives DCF valuations, hurdle rates, and boardroom debates.
• Portfolio tilt decisions: Two stocks, same expected alpha, but one with higher beta. CAPM says the higher-beta one already bakes in a higher expected return; you’re not smarter for chasing it.
• International assets: In an integrated global market, the market portfolio is global. A country ETF’s cost of equity depends on covariance with the global market, not just its own volatility.
• Hedging overlay: Holding a negative-beta asset? Congrats — it’s an insurance-like position. Expect a lower return, just like how insurance costs you money but saves you during chaos.

---

Common Mistakes in CAPM (Don’t Do These; Your Future Self Will Thank You)

1. Using the wrong “market” proxy: The S&P 500 is not the world. If markets are integrated, your proxy should be broad and global. Home bias = model sin.
2. Beta as destiny: Beta is a conditional, time-varying statistic. Estimation windows, regimes, and leverage matter. Resampling helped stabilize portfolio weights; apply similar humility to betas.
3. Horizon mismatch: One-month beta + five-year project discount rate = chaos. Align the measurement horizon with the decision horizon.
4. Ignoring currencies: For global assets, pick a base currency. Market, rf, and returns must match the same currency/deflator.
5. Treating the equity risk premium as a constant: It breathes! Cycle-aware estimates, implied measures, and scenario ranges are your friends.
6. Forgetting constraints and costs: Short-sale constraints, taxes, and trading costs push reality away from the SML. We covered this in Position 8 — they’re not footnotes; they’re gravity.

---

Assumptions vs. Reality: What Breaks, What Bends

CAPM Assumption	Translation	If Violated (Link to Prior Topics)
Homogeneous expectations	Everyone agrees on inputs	Estimation error and disagreement lead to dispersion; resampling (Position 9) tries to stabilize inputs
Frictionless markets	No costs/constraints	With constraints/transaction costs (Position 8), some assets can’t reach SML; expect wedges and persistent mispricings
Global market portfolio	All assets included, integrated	Market segmentation/home bias reduce international diversification benefits (Position 10) and change relevant beta
Risk-free borrowing/lending	Same rf for all	In practice, borrowing costs differ; constrained borrowing flattens the SML for many investors

Reality check: CAPM is a compass, not a GPS. It points toward equilibrium logic; it doesn’t micro-navigate potholes.

---

Quick Derivation Sketch (For the Math-Curious)

1. Start with mean-variance optimization with a risk-free asset → the Capital Market Line (CML) is tangent to the efficient frontier.
2. If investors share the same inputs, they all choose the same tangency portfolio → call it M.
3. In equilibrium, aggregate holdings equal the market capitalization weights → M is the market portfolio.
4. First-order conditions imply expected excess returns are proportional to covariances with M:

E[R_i] - r_f = λ · Cov(R_i, R_M)
=> E[R_i] = r_f + β_i (E[R_M] - r_f)

Where λ = (E[R_M] - r_f) / Var(R_M).

---

How to Use CAPM Without Getting Played

• Use it as a baseline: Start with CAPM to set a coherent cost of equity. Then test sensitivity to ERP and beta.
• Choose a market proxy that matches your investable universe: Global investors → global market. Domestic mandates → domestic market (with eyes open about segmentation).
• Check for non-market risks that matter: If a strategy’s payoff depends on size, value, profitability, or momentum, you’re outside pure CAPM territory. That’s where multifactor models enter.
• Be honest about frictions: Add spreads, taxes, and constraints to your implementation plan. Theory says “on the SML,” execution says “mind the slippage.”

---

Closing: The Takeaways You Can Trade On

• CAPM assumptions and intuition give you a clean rule: expected returns rise only with market-related risk — beta. Unique risk? Diversify it and stop asking for a reward.
• The model extends your portfolio theory work: from building efficient portfolios to explaining why the market’s pricing looks the way it does when everyone behaves like a mean-variance optimizer.
• When assumptions crack (constraints, costs, segmentation, messy expectations), deviations from the Security Market Line show up — and that’s the on-ramp to multifactor models.

Big insight: Diversification is not just a nice-to-have; it’s the reason markets only pay for what you can’t diversify away.

Use CAPM as your baseline map. Annotate it with real-world potholes. And when the single factor can’t carry the drama, bring in the multifactor cast. But that’s the next episode.