Risk and Diversification

Ask someone to define risk and they usually reach for a synonym — danger, uncertainty, something bad happening. That's not what the word means to anyone actually pricing an investment. Risk is the spread of outcomes an investment could produce: how far the good years and the bad years sit from each other, not how alarming the word sounds in a sentence.

A savings account paying 4% is low risk because its outcome is almost always 4%, full stop — the spread is nearly zero. A single stock is high risk not because it's “dangerous” in some vague sense, but because its one-year return could concretely land anywhere from an 80% loss to a 120% gain, and you don't know which until the year is already over. Same underlying idea, a wildly different band of possible outcomes.

That band has a name — volatility — but you don't need the formula to use the idea. Wider band, riskier holding. Narrower band, safer one. Everything else in this chapter is about what does and doesn't narrow that band.

Holding one stock means your return is that stock's return — whatever a single company's management, competitors, lawsuits, or bad quarter do to its price happens directly to your entire investment. A wonderful year is wonderful. A catastrophic one is catastrophic, with nothing else in the portfolio to absorb it.

Put $10,000 into one company and a rough year that costs it 80% of its value leaves you with $2,000. There is no averaging, no other holding quietly having a decent year alongside it — the company's outcome and your outcome are the same number, because they are the same bet.

This is also why holding a large amount of your employer's stock is riskier than it feels. Your paycheck already depends on that company doing well. Loading your investments with the same company's stock means a bad year for it can cost you your income and your savings at once, for the same underlying reason.

Before diversification can make sense, one more idea has to click: not all combinations of stocks are equally diversified, even when they're different companies. Correlation is just whether two things tend to move together — no algebra required, only the pattern.

Twenty stocks that are all positively correlated — say, twenty companies in the same industry, riding the same trend — behave a lot more like one big stock than twenty small ones. Real diversification needs holdings that don't rise and fall for the same reasons, which is a different requirement from simply owning “a lot of companies.”

Diversification means holding many different investments so no single one can do that much damage. If one of twenty holdings has a disastrous year, it's one-twentieth of the portfolio, not the entire thing — and historically, on average, some other holding having a strong year has offset a meaningful part of that loss.

The chart below runs the same idea thirty separate times. Each trial draws twenty companies from an identical, volatile world — one-year returns anywhere from an 80% loss to a 120% gain are on the table for any of them. Toggle between holding just the first company from each trial and holding the average of all twenty, and watch what happens to the range of outcomes.

There are two different kinds of risk hiding inside “risk,” and only one of them responds to diversification. Idiosyncratic risk is specific to one company — a lawsuit, a product recall, a fraud, a bad management decision. Because these events are largely independent of each other, spreading across many companies cancels a lot of it out, which is exactly what the chart above shows.

Systematic risk — also called market risk — is the part that hits nearly everything at once: a recession, a spike in interest rates, a war, a pandemic. In the 2007–2009 financial crisis, broad stock indices fell by roughly half from peak to trough, across nearly every industry and nearly every country at once. Owning twenty companies instead of one did nothing to soften that, because the source of the risk was never any individual company to begin with.

Toggle between the two views above and notice what actually changes: the best and worst single-year outcomes both move dramatically closer to the average once the outcomes are spread across twenty stocks instead of concentrated in one. That's the trade — diversification gives up the chance of an extraordinary single-stock win in exchange for taking the catastrophic single-stock loss off the table.

It doesn't give up the market-wide swings underneath it. Spreading across many holdings tightens the band that idiosyncratic risk carves out, but the whole diversified basket can still fall 30% or 40% in a bad year for the market as a whole — it's just no longer at risk of falling to zero because one company failed.