S&P 5007,659▼ 1.09%Nasdaq26,140▼ 2.15%Dow Jones52,002▼ 0.34%Nasdaq 10029,188▼ 2.50%VIX19.99▲ 21.45%US Dollar101.84▼ 0.21%Gold (GLD)382.01▼ 0.18%Oil (USO)161.76▲ 2.83%

Diversification explained: what it does and does not protect against

18 October 20264 min readFinancial knowledge · lesson 106
Diversification explained: what it does and does not protect against

Diversification means spreading your money across holdings that do not all react the same way to the same events, so that no single disappointment can dominate the result. It matters because some risks can be reduced cheaply by owning more than a handful of investments, while others cannot be reduced at all. Knowing which is which helps you avoid both needless risk and expecting protection that diversification never promised.

Two kinds of risk

Investment risk is usually split into two parts:

  • Company-specific (idiosyncratic) risk. A product recall, a failed drug trial, an accounting scandal, a lost major customer. These events hit one company or a small group of them.
  • Market (systematic) risk. Recessions, shifts in interest rates, and broad swings in sentiment that move most stocks in the same direction at the same time.

Diversification works on the first kind. When you hold many companies, a disaster at one is averaged against ordinary results at the others. It does very little about the second kind, because every stock is exposed to it. That is why a portfolio of 100 US stocks can still fall sharply in a bear market.

How it works

For a portfolio of two holdings, volatility depends on each holding's volatility, its weight, and the correlation between the two:

Portfolio variance = w1^2 * s1^2 + w2^2 * s2^2 + 2 * w1 * w2 * rho * s1 * s2
Portfolio volatility = square root of variance

w = weight, s = standard deviation of returns, rho = correlation

Only the last term depends on how the holdings move together. When rho is below 1, the portfolio is less volatile than the weighted average of its parts. The lower the correlation, the bigger the benefit. See correlation between stocks for how that number is measured and why it changes.

Example (illustrative numbers)

Two stocks each have annual volatility of 30%, and you hold them 50/50.

Correlation between them Portfolio volatility
1.0 30%
0.5 about 26%
0.0 about 21%
-0.5 about 15%

With a correlation of 0.5, the variance is 0.25 x 0.09 + 0.25 x 0.09 + 2 x 0.25 x 0.5 x 0.09 = 0.0675, and the square root is about 26%. Expected return in this setup is just the weighted average of the two, so the lower volatility does not require giving up expected return. In real life, correlations between two US stocks are rarely below zero, so the realistic benefit is toward the top of the table.

The same math explains diminishing returns. As you add more stocks, the portfolio's volatility approaches a floor set by the average co-movement among them. You can remove the company-specific part, but the common part stays.

How many holdings is enough?

There is no exact number. Studies of US stocks have generally found that most of the reduction in company-specific risk arrives within the first couple of dozen holdings, with smaller gains after that. The result depends on the period and on how the stocks are chosen. Thirty stocks from one industry are not thirty independent bets.

For many investors, a broad fund is a simpler route to wide diversification. An ETF tracking the S&P 500 holds about 500 companies in one purchase, though the index is market-cap weighted, so the largest companies carry the most weight.

What diversification does not do

  • It does not prevent losses in a market-wide decline. In sharp selloffs, many assets that normally move independently tend to fall together, because correlations rise under stress.
  • It does not guarantee a gain. It narrows the range of outcomes. It does not shift the whole range upward.
  • It does not help if the diversification is only on paper. That is the hidden concentration problem.

Hidden concentration

A portfolio can look varied and still depend on one idea:

  • Same sector or theme. Several software, semiconductor, or bank stocks often share the same drivers.
  • Same style or factor. A portfolio of only high-growth names, or only rate-sensitive names, can be hit by one change in the environment.
  • Overlapping funds. Three funds that all hold the same large companies are one position in disguise.
  • Your paycheck. If you work for a company and also own its stock, one event can hit your job and your savings together.

The opposite problem

Spreading money across hundreds of small positions can make each one too small to matter and too numerous to follow. At the extreme you own an expensive copy of an index. Position size, not just count, determines how much any single holding can hurt or help. See position sizing.

Where to see it on StocksXray

Each stock page names the company's sector, and the sector pages (for example Health Care) group companies by GICS sector. Checking which sectors your holdings sit in is a quick test of whether they cluster in one place. The S&P 500 index page shows the broad universe the site covers.

Key takeaways

  • Diversification reduces company-specific risk; it cannot remove market risk.
  • The benefit depends on correlation: the less holdings move together, the more it helps.
  • Most of the gain comes from the first few dozen holdings, with diminishing returns after.
  • Look for hidden overlap in sectors, styles, funds, and your employer.
  • Mix across asset classes too, not only across stocks. See asset allocation.

Related reading

Educational only, not financial advice. Examples use illustrative numbers.