Equity Beta And Systematic Market Exposure

Introduction

An equity portfolio can hold many securities, yet its exposure to a broad stock-market movement can often be summarized by one measure: beta. For an insurer, this gives a compact way to estimate how an equity portfolio might respond to a stated movement in a chosen market benchmark. It is an estimate of sensitivity, not a guarantee of the portfolio’s actual return.

This tutorial defines security and portfolio beta, shows how to calculate portfolio beta from market-value weights, and applies it to hypothetical market scenarios. The calculations use a fixed portfolio over the measurement interval and a linear approximation that isolates market-related movement.

Definition: For a security return $R_i$ and a benchmark return $R_m$ measured over the same period, the security’s beta relative to that benchmark is $\beta_i=\operatorname{Cov}(R_i,R_m)/\operatorname{Var}(R_m)$, provided $\operatorname{Var}(R_m)>0$. Beta measures the security’s linear sensitivity to the benchmark. A beta of 1 indicates the same estimated sensitivity as the benchmark; a beta greater than 1 indicates greater sensitivity, and a beta between 0 and 1 indicates lower sensitivity. A negative beta indicates estimated movement in the opposite direction.

Beta is tied to the benchmark and the data or model used to estimate it. A beta estimated relative to one broad equity index is not automatically the beta relative to another index. In applications, the security betas supplied for a calculation are treated as estimates for the benchmark and time horizon relevant to the problem.

From Security Betas to Portfolio Beta

Suppose a portfolio holds $n$ securities. Let $V_i$ be the beginning market value of security $i$, and let $V=\sum_{i=1}^{n} V_i$ be the portfolio’s beginning market value. The beginning market-value weight is $w_i=V_i/V$, so the weights sum to 1. If the portfolio has a cash allocation, include cash as a holding; for this approximation, cash has beta 0 relative to the equity benchmark.

Portfolio beta: For a portfolio with fixed beginning weights over the measurement interval and constituent betas measured against the same benchmark and over the same horizon, portfolio beta is $\beta_P=\sum_{i=1}^{n}w_i\beta_i$. With a cash holding, include its weight and beta 0 in the sum.

The formula requires value weights, not an unweighted average of the listed betas. A security representing half of the portfolio’s value should influence the result more than one representing a small fraction. The formula also presumes that the weights describe the portfolio whose return is being measured; substantial trades or cash flows during the interval can make a starting-weight estimate less representative.

\beta_P=\frac{\sum_{i=1}^{n}V_i\beta_i}{\sum_{i=1}^{n}V_i}

For a specified market return $R_m$, the beta-based approximation to the portfolio’s market-related return is $\beta_P R_m$. If the portfolio’s beginning value is $V$, its approximate market-related dollar change is $V\beta_P R_m$. These equations omit any separate contribution from security-specific movements, active return, fees, or other effects.

\text{Approximate market-related portfolio return}=\beta_P R_m, \qquad \text{Approximate dollar change}=V\beta_P R_m

Why the Weighted-Beta Formula Holds

The portfolio formula follows from the linearity of covariance. For fixed weights, the portfolio return is $R_P=\sum_i w_iR_i$. Its beta relative to the same benchmark is the covariance of $R_P$ with $R_m$, divided by the variance of $R_m$. Expanding that covariance gives the weighted sum of the constituent betas:

\begin{aligned} \beta_P &=\frac{\operatorname{Cov}(R_P,R_m)}{\operatorname{Var}(R_m)}\\ &=\frac{\operatorname{Cov}\left(\sum_i w_iR_i,R_m\right)}{\operatorname{Var}(R_m)}\\ &=\frac{\sum_i w_i\operatorname{Cov}(R_i,R_m)}{\operatorname{Var}(R_m)}\\ &=\sum_i w_i\frac{\operatorname{Cov}(R_i,R_m)}{\operatorname{Var}(R_m)} =\sum_i w_i\beta_i. \end{aligned}

This justification depends on using the same benchmark for all constituent betas and treating the weights as fixed for the return calculation. It does not say that the portfolio’s actual return must equal beta times the benchmark return. Beta captures the linear market relationship; realized returns can differ.

Worked Example: Estimating Beta and a Down-Market Change

A hypothetical insurer holds three equity funds with beginning values of $3,200,000, $2,800,000, and $2,000,000. Their estimated betas to the same broad-market benchmark are 1.20, 0.80, and 1.40, respectively. Assume the holdings remain fixed over the scenario interval and ignore fees and non-market effects. Estimate portfolio beta and the dollar change if the benchmark falls by 6%.

Method. Calculate the total value, convert each holding to a beginning market-value weight, and apply the beta-based market-movement approximation. The supplied betas are treated as fixed estimates; no probability distribution or interest-rate assumption is used.

Step 1: Total portfolio value. $V=\$3{,}200{,}000+\$2{,}800{,}000+\$2{,}000{,}000=\$8{,}000{,}000$.

Step 2: Calculate the weights. The weights are $3{,}200{,}000/8{,}000{,}000=0.40$, $2{,}800{,}000/8{,}000{,}000=0.35$, and $2{,}000{,}000/8{,}000{,}000=0.25$. They sum to $0.40+0.35+0.25=1.00$.

Step 3: Calculate portfolio beta. $\beta_P=(0.40)(1.20)+(0.35)(0.80)+(0.25)(1.40)=0.48+0.28+0.35=1.11$.

Step 4: Estimate the return and dollar change. With $R_m=-0.06$, the estimated market-related portfolio return is $(1.11)(-0.06)=-0.0666$, or $-6.66\%$. The dollar change is $\$8{,}000{,}000(-0.0666)=-\$532{,}800$. The corresponding approximate value after this market-related change is $\$8{,}000{,}000-\$532{,}800=\$7{,}467{,}200$.

Check and common mistake. Beta 1.11 implies a market-related decline 1.11 times the benchmark decline: $1.11(6\%)=6.66\%$, consistent with the result. An unweighted average of the three betas would be $1.1333$, which is not the portfolio beta because the three holdings have different values.

Applying Beta to Portfolio Values

Once portfolio beta is known, a scenario calculation has two distinct steps: estimate the percentage return first, then multiply by the portfolio’s beginning value to obtain a dollar amount. Keeping these steps separate helps avoid confusing a percentage movement with a dollar loss or gain. The result is a first-order market-related estimate, not a full forecast of all sources of portfolio return.

Worked Example: Value Weights When Holdings Differ in Size

A hypothetical equity portfolio has $2,400,000 in a fund with beta 0.70, $1,600,000 in a fund with beta 1.25, and $2,000,000 in cash. The betas are relative to the same broad-market benchmark. Assuming the holdings stay fixed and cash beta is 0, estimate the portfolio beta, the dollar change under a 4.5% benchmark increase, and the approximate ending value.

Method. Use beginning values to calculate weights, include cash with beta 0, and apply the linear market-movement approximation. No interest earned on cash or other portfolio return is included.

Step 1: Find the total and weights. Total value is $V=\$2{,}400{,}000+\$1{,}600{,}000+\$2{,}000{,}000=\$6{,}000{,}000$. The weights are $0.40$, $0.2666667$, and $0.3333333$, respectively; their sum is 1.00.

Step 2: Calculate beta. $\beta_P=(0.40)(0.70)+(0.2666667)(1.25)+(0.3333333)(0)=0.28+0.3333334+0=0.6133334$, or approximately 0.6133.

Step 3: Estimate the market-related return and value change. The benchmark return is $R_m=0.045$. Thus the estimated portfolio return is $0.6133334(0.045)=0.0276000$, or 2.76%. The dollar change is $\$6{,}000{,}000(0.0276)=\$165{,}600$. The approximate ending value is $\$6{,}000{,}000+\$165{,}600=\$6{,}165{,}600$.

Check and common mistake. The cash holding reduces portfolio beta because one-third of the portfolio has beta 0. Averaging only the two fund betas would give 0.975, incorrectly treating the cash allocation as if it did not exist. Using full precision for the weights gives the reported 0.6133 beta and $165,600 change.

How a Portfolio Change Affects Beta

Portfolio beta is not a permanent property of an investment account. It changes when market values, holdings, or their beta estimates change. If a position is replaced while the total portfolio value is held constant, the beta effect can be calculated from the change in that position’s weighted contribution. This can help compare the market sensitivity of two allocation choices, while leaving questions of implementation and broader risk measurement aside.

Worked Example: Replacing Part of a Position

A hypothetical $10,000,000 equity portfolio holds $3,000,000 in Stock A with beta 1.40, $5,000,000 in Stock B with beta 0.80, and $2,000,000 in cash with beta 0. The portfolio replaces $1,000,000 of Stock B with cash, leaving total value unchanged. Estimate the beta before and after the trade, and compare the market-related dollar changes if the benchmark falls by 12%.

Method. Recalculate the value-weighted beta for the original and revised holdings, then apply the same benchmark scenario. The supplied betas are treated as unchanged by the trade, and cash is assigned beta 0.

Step 1: Original beta. The original weights are 0.30, 0.50, and 0.20. Therefore, $\beta_{\text{before}}=(0.30)(1.40)+(0.50)(0.80)+(0.20)(0)=0.42+0.40=0.82$.

Step 2: Revised weights and beta. The revised values are $3,000,000 in Stock A, $4,000,000 in Stock B, and $3,000,000 in cash. The revised weights are 0.30, 0.40, and 0.30. Thus, $\beta_{\text{after}}=(0.30)(1.40)+(0.40)(0.80)+(0.30)(0)=0.42+0.32=0.74$. The beta decreases by $0.82-0.74=0.08$.

Step 3: Compare scenario changes. Before the trade, the estimated market-related return is $0.82(-0.12)=-0.0984$, or $-9.84\%$, giving a dollar change of $\$10{,}000{,}000(-0.0984)=-\$984{,}000$. After the trade, the return estimate is $0.74(-0.12)=-0.0888$, or $-8.88\%$, giving a dollar change of $-\$888{,}000$.

Check and common mistake. The estimated decline is $96,000 smaller after the trade. This matches the change in beta exposure: $(0.08)(12\%)(\$10{,}000{,}000)=\$96{,}000$. A common error is to apply the beta change of 0.08 directly to the portfolio value as an 8% change; the effect under this scenario is $0.08$ times the 12% benchmark decline, or 0.96% of portfolio value.

Interpretation, Limitations, and Exam Traps

Beta is useful when the question is specifically about sensitivity to a broad equity-market movement. It does not measure every source of portfolio risk. Two portfolios can have the same beta but different constituent-specific risk, and a beta-based scenario does not capture every actual security return. In particular, the approximation $\beta_P R_m$ omits the portfolio’s residual return and any other effects not represented by the benchmark movement.

The estimate also depends on its inputs. A change in the benchmark, estimation period, or security beta estimates can change the result. If a question gives betas as fixed inputs, use those values rather than attempting to infer new betas. If it gives constituent values, derive weights from the total portfolio value, including cash or any other stated holdings.

The most common calculation traps are mixing benchmarks, using an unweighted average, omitting a zero-beta cash allocation, and applying a percentage return directly as a dollar amount. Before calculating, check that the weights add to 1, the betas refer to the same benchmark, and the scenario return is expressed as a decimal when used in the multiplication.

Key takeaway: For fixed holdings and a common benchmark, portfolio beta is the beginning market-value-weighted average of constituent betas. Multiplying that beta by a stated benchmark return gives an approximate market-related portfolio return; multiplying again by beginning portfolio value gives the approximate dollar change.

Check Your Understanding

For each original problem, calculate using the stated assumptions. Answers are provided so you can check both the method and arithmetic.

A hypothetical portfolio has weights 50%, 30%, and 20%, with betas 1.30, 0.60, and 1.10 to the same benchmark. Estimate its beta and market-related return if the benchmark falls by 3%. (Answer: beta 1.05; return $-3.15\%$.)
A $5,000,000 portfolio holds $1,500,000 in an asset with beta 0.80, $2,500,000 in an asset with beta 1.20, and $1,000,000 in cash with beta 0. Estimate beta and the dollar change if the benchmark rises by 7%. (Answer: beta 0.84; dollar change $+\$294{,}000$.)
A hypothetical $12,000,000 portfolio has weights 25%, 45%, and 30%, and constituent betas 1.60, 0.90, and 1.10. Estimate beta and the dollar change if the benchmark falls by 4%. (Answer: beta 1.135; dollar change $-\$544{,}800$.)
A $20,000,000 portfolio with beta 0.90 replaces $4,000,000 of a holding with beta 1.40 by a holding with beta 0.50. Assuming the total value is unchanged, estimate the new beta and the reduction in estimated dollar loss under a 5% benchmark decline. (Answer: new beta 0.72; estimated loss reduction $\$180{,}000$.)
A security has estimated covariance 0.012 with a benchmark whose return variance is 0.010. A $3,500,000 portfolio consists entirely of that security. Estimate its beta and the market-related dollar change if the benchmark falls by 8%. (Answer: beta 1.20; dollar change $-\$336{,}000$.)

What You'll Learn