Chapter 2: Portfolio Theory and Asset Pricing#

You already know that a rational investor wants more return and dislikes risk. But how do you blend many investments into a single portfolio that gives you the best possible balance? This chapter builds that answer from the ground up — starting with the simple idea of not putting all your eggs in one basket, and ending with the main models that connect an asset’s risk to its fair expected return.

The Big Picture#

Every day, investors face a trade-off: they can aim for higher returns, but only by accepting more uncertainty. Portfolio theory shows us how to mix assets so that we cut unnecessary risk while keeping as much return as possible. From that base, asset pricing models like the CAPM and APT give us a clear way to figure out the return an investor should ask for when taking on a certain kind of risk. Understanding these ideas helps answer a key question: “Given how risky this asset is, what is a fair price for it today?”

Portfolio Selection and the Magic of Diversification#

Imagine you own a single stock. Its price jumps around from day to day — that’s volatility, a simple measure of risk. If you add a second stock that doesn’t move in perfect lockstep with the first, something surprising happens: your total ups and downs can shrink, even if neither stock, on its own, becomes safer. This is the heart of diversification.

Harry Markowitz gave us a math way to describe this. He said investors care about two numbers: the expected return of a portfolio (what they think they’ll earn on average) and the risk, measured by how spread out the returns are — its variance or standard deviation.

For a single asset, risk is just its own volatility. For a portfolio of two assets, though, risk depends not only on each asset’s volatility but also on how they move together. We measure that co-movement with covariance or the more intuitive correlation coefficient, usually written as . Correlation can range from (they move perfectly opposite) to (they move perfectly together). A correlation of zero means there’s no linear link.

If we put weight in asset 1 and weight in asset 2 (with ), the expected return of the portfolio is simply the weighted average:

The variance of the portfolio, however, is not a simple weighted average. It contains a crucial extra term:

Because , we can see that when , the last term is smaller than it would be if the assets were perfectly correlated. In fact, the portfolio’s standard deviation can be less than the weighted average of the individual standard deviations. That’s the magic: by mixing assets that don’t perfectly march together, you cancel out some of the swings.

Think of a vendor selling both ice cream and umbrellas. On a sunny day, ice cream sales soar but umbrellas gather dust; on a rainy day, the reverse happens. The two businesses have a negative correlation, so the vendor’s total income is steadier than either business alone. Diversification works the same way in financial markets.

The more assets you add, the more of this cancelling-out occurs — but only up to a point. Risk can be split into two parts.

Unsystematic risk (also called company-specific or diversifiable risk) is the risk tied to a single firm or industry — a strike, a product flop, a lawsuit. You can get rid of almost all of it by holding a broad basket of stocks from different sectors.

Systematic risk (market risk or non-diversifiable risk) is the risk that hits the whole market — a recession, interest rate changes, a war. No matter how many stocks you own, you can’t diversify away the overall economy’s ups and downs.

So diversification lets you avoid putting all your eggs in one basket for risks unique to single eggs, but you still have to carry the basket through the same storm.

Markowitz showed that for any set of risky assets, there is a whole region of possible portfolios. If we plot them with expected return on the y-axis and standard deviation on the x-axis, they form a bullet-shaped cloud. The upper edge of that cloud is the efficient frontier: the portfolios that give the highest expected return for each level of risk. Any sensible investor should pick a portfolio on that frontier.

Efficient frontier: The set of portfolios that offer the maximum expected return for a given amount of risk (or, equivalently, the minimum risk for a given expected return).

These ideas give you a clear way to build a portfolio. But they leave one big question: how much extra return should you ask for taking on a bit more risk, especially the risk you can’t diversify away?

📝 Section Recap: By combining assets that don’t move in perfect sync, you can slash company-specific risk without giving up expected return. Only systematic risk matters to a well-diversified investor, and the efficient frontier shows the best risk-return trade-offs you can get.

The Capital Asset Pricing Model and the Security Market Line#

Once we accept that investors can — and should — diversify away unsystematic risk, the next step is to figure out how the market prices an asset’s contribution to a diversified portfolio. The Capital Asset Pricing Model (CAPM) was created to answer exactly that.

CAPM makes a few simplifying assumptions: investors are rational and dislike risk, they can borrow and lend at the same risk-free rate, and they all hold a portfolio that covers every risky asset. In such a world, everyone will combine the risk-free asset with the same portfolio of risky assets — the market portfolio, essentially a mix of all available assets weighted by their market value.

If we add the risk-free asset, the set of best risk-return combinations becomes a straight line from the risk-free rate through the market portfolio. That line is the Capital Market Line (CML), and it shows the trade-off for well-diversified portfolios. But what about a single stock? Its risk is not its total volatility; it’s how much that stock adds to the volatility of the market portfolio. That contribution is measured by beta.

Beta (): A measure of an asset’s systematic risk. It tells you how much the asset’s return tends to move when the market’s return moves. Mathematically, .

A beta of 1 means the asset moves with the market. A beta of 0.5 means it’s about half as jumpy as the market. A beta of 1.5 means it’s one-and-a-half times as jumpy.

CAPM says that the expected return on any asset is set by its beta and the market risk premium — the extra return investors want for taking on average market risk:

Here, is the risk-free rate, is the expected return on the market portfolio, and is the market risk premium.

This relationship is drawn as the Security Market Line (SML) — a straight line starting at and sloping up with the market risk premium. In the CAPM world, all assets should lie on the SML. If an asset sat above the line, it would be offering too high a return for its beta — it would be undervalued, and investors would buy it, pushing its price up and its expected return down until it lands on the line. If it’s below the line, it’s overpriced.

Let’s make it concrete. Suppose the risk-free rate is , the expected market return is , so the market risk premium is . A defensive utility stock with would need to return . An aggressive tech stock with would need to return to compensate for its higher systematic risk.

CAPM’s big insight is simple: only systematic risk gets rewarded with a higher expected return. Company-specific risk, which you can diversify away for free, earns no extra reward. So beta is the only measure of risk that matters to a well-diversified investor.

📝 Section Recap: CAPM predicts a straight-line link between an asset’s beta and its expected return, shown by the Security Market Line. Only risk you can’t diversify away — systematic risk — is rewarded by the market.

Arbitrage Pricing Theory: Multiple Factors at Play#

CAPM is neat, but its dependence on the market portfolio and strong assumptions pushed researchers to find a more flexible model. The Arbitrage Pricing Theory (APT), developed by Stephen Ross, does just that.

APT doesn’t assume everyone holds the market portfolio. Instead, it builds on the idea of arbitrage: if two assets promise exactly the same payoffs no matter what happens, they must sell for the same price. If they didn’t, you could buy the cheaper one and sell the expensive one short, locking in a risk-free profit. In a well-functioning market, such chances disappear almost instantly as prices adjust.

APT extends this idea by allowing an asset’s return to be driven by several different sources of systematic risk — called factors — instead of just one market factor. Examples might include changes in GDP growth, surprise inflation, shifts in interest rates, or oil price shocks. Each factor has its own factor risk premium (), the extra return investors want for bearing one unit of exposure to that factor.

The expected return on an asset becomes a combination of its exposures to each factor and the associated risk premiums:

Here, is the sensitivity of asset to factor 1, and so on.

Think of a large airline. Its stock is sensitive to the overall stock market, but it’s also especially sensitive to oil prices (big fuel cost) and to consumer confidence (people’s willingness to travel). A two-factor APT model might include a market factor and an oil-price factor. If the market risk premium is and the oil-price risk premium is , and the stock has a market beta of and an oil beta of , then the required return would be . This gives a much richer picture than a single beta from CAPM.

APT doesn’t tell you which factors to use — that’s both its strength (flexibility) and its weakness (no built-in list). In practice, researchers have proposed many factor sets, like the Fama–French three-factor model that adds size and value factors. The core logic stays the same: assets that move together with the same underlying risks should earn similar expected returns; if not, arbitrageurs step in and force prices into line.

Arbitrage: Buying and selling the same or very similar assets at the same time in different places or forms to profit from a price difference without taking any risk.

The key difference from CAPM: CAPM says “the market portfolio contains all systematic risk,” while APT says “there can be several independent sources of systematic risk, and the law of one price ensures no free lunches.” APT fits better with real-world data, where stocks with similar market betas often have quite different returns — because they are exposed to other risks that the market index doesn’t capture.

📝 Section Recap: APT extends the risk-return relationship to multiple factors, each with its own risk premium. It relies on the powerful no-arbitrage logic rather than a single market portfolio, giving a flexible way to understand why different assets earn different returns.

Summary#

We began with the simple idea that mixing assets that don’t move together can greatly cut risk, leading to Markowitz’s efficient frontier and the key split between risk you can diversify away and risk you cannot. That insight gave us the CAPM, where beta measures how much a stock adds to a diversified portfolio’s risk, and the Security Market Line tells you what return it should offer. Then we saw that APT breaks free from the single-factor mold by allowing multiple economic forces, all held together by the iron rule of no arbitrage. These tools are the foundation for how finance professionals think about risk, build portfolios, and price assets.