How Many Stocks Do You Need?

The question haunts every investor: am I diversified enough? Five stocks? Twenty? Fifty? One hundred? Portfolio theory offers a precise answer, backed by mathematics. Research consistently shows that roughly 20–40 uncorrelated stocks eliminate approximately 90% of unsystematic risk. Beyond that point, you're capturing diminishing returns—adding stocks helps, but only marginally. This insight separates rational portfolio building from unnecessary complexity and sets clear expectations for how diversified you really need to be.

Quick definition: The number of stocks required for adequate diversification depends on their correlation. Uncorrelated stocks diversify faster than correlated ones; roughly 20–40 holdings capture 90% of the diversification benefit.

Key takeaways

• A single stock captures 100% of unsystematic risk; adding a second uncorrelated stock cuts that roughly in half.
• Diminishing returns set in rapidly: moving from 1 to 10 stocks provides massive risk reduction; 10 to 50 provides less; 50 to 500 provides minimal benefit.
• The optimal number balances diversification with manageability: 20–40 individual stocks, or a single broad index fund with hundreds.
• Correlation matters more than count; 50 tech stocks are less diversified than 10 stocks across five different sectors.
• Passive index funds are "overdiversified" by math but justified by liquidity, rebalancing benefits, and lowest fees.

The math of risk reduction

Start with one stock. Its volatility represents 100% unsystematic risk (company-specific) plus systematic risk (market-wide). Now add a second, uncorrelated stock. If the two stocks move independently—one up while the other sideways—their portfolio volatility drops.

With two perfectly uncorrelated stocks of equal risk, the portfolio's standard deviation (volatility) is:

σ_portfolio = √(σ₁² + σ₂²) / 2

If each stock has 40% annual volatility, the portfolio has √(1600 + 1600) / 2 = √1600 = 40% (before correlation adjustment). But if the stocks are uncorrelated, some of the moves cancel. The actual portfolio volatility might be 28–30%, a meaningful drop despite adding a single holding.

Continue this logic. With five uncorrelated stocks of equal risk, the portfolio volatility falls to roughly 18–20%. With ten, to 12–14%. With twenty, to 8–10%. Notice the deceleration: doubling from 5 to 10 cuts volatility by 30%; doubling from 10 to 20 cuts it by 25% more (less bang for the buck).

The formula shows why: unsystematic risk declines with the square root of the number of holdings. Doubling your stock count doesn't cut risk in half; it cuts it by the square root of 2 (roughly 1.41). This square-root law is the mathematical heart of diversification.

Finding the knee in the curve

Research by scholars like Elton and Gruber (1977) empirically measured the diminishing-returns curve. Their findings:

• 1 stock: 100% of unsystematic risk remains (plus systematic risk).
• 10 stocks: ~50% of unsystematic risk remains.
• 20 stocks: ~30% of unsystematic risk remains.
• 50 stocks: ~15% of unsystematic risk remains.
• 100+ stocks: ~5% of unsystematic risk remains.

The "knee" in the curve—where diminishing returns become pronounced—sits around 20–30 stocks. A portfolio of 20–30 uncorrelated holdings captures roughly 85–90% of the possible diversification benefit. Moving from 30 to 100 stocks cuts unsystematic risk by another 5–10%, but the effort and complexity required grow substantially.

For most individual investors, 20–40 stocks across different sectors, industries, and geographies hit the sweet spot: maximum diversification benefit relative to effort.

The role of correlation

This analysis assumes low correlation. In reality, correlation matters enormously. Fifty utility stocks are less diversified than ten stocks spanning utilities, tech, healthcare, energy, and consumer goods—because utility stocks move together.

During normal times, U.S. stocks average a correlation of roughly 0.3–0.5 (moderate to weak). During crises, correlations spike toward 1.0, and diversification fails. This is why the math above is a baseline; real diversification depends on correlation.

A practical implication: when building a 30-stock portfolio, spread holdings across different sectors, industries, and market caps. Diversification works only when your stocks don't all swing the same way.

Visualizing the frontier

The chart shows the relationship between number of holdings and remaining unsystematic risk. Notice the steep drop from 1 to 10, then gradual flattening. Most practitioners target the "knee," around 20–40 holdings.

Individual stocks vs. index funds

A natural question: should you own 30 individual stocks or a single index fund holding 500?

30 individual stocks:

• Pros: Active monitoring, customization, ability to avoid companies you distrust.
• Cons: Research time required, higher transaction costs, risk of missing the signal in company-specific news.
• Math: Captures ~85% of diversification benefit with meaningful effort.

Index fund (e.g., S&P 500):

• Pros: Automatic diversification, minimal fees (0.03–0.10% annually), no research required, rebalancing handled automatically.
• Cons: No customization, forced to hold companies you might dislike.
• Math: Captures ~99% of diversification benefit with zero effort.

Index funds are "overdiversified" by the math above—they hold far more than 30 stocks. But the extra holdings carry negligible diversification benefit and introduce rebalancing dynamics that often boost returns (as weaker stocks are diluted and stronger stocks gain weight). The overhead is paid for by liquidity and passive management simplicity.

For most investors, an index fund is the rational choice. For those willing to research and monitor 30–40 stocks, individual holdings work but require discipline.

Sector and geographic diversification

Diversification isn't just a count; it's a composition. Consider two investors:

Investor A: Holds 50 U.S. large-cap tech stocks. Investor B: Holds 10 U.S. stocks across five sectors, 10 international stocks, and 5 bond holdings.

By count, A is more diversified. By correlation and sector exposure, B is far more diversified. B faces lower unsystematic risk because the holdings don't move together.

This teaches an important lesson: after reaching 20–30 holdings, additional diversification comes from asset classes and geographies, not additional stocks. A 60% stock / 40% bond portfolio is more diversified than a 100% stock portfolio with twice as many holdings.

Real-world examples

The Berkshire Hathaway exception: Warren Buffett holds roughly 5–10 major positions, making Berkshire a concentrated portfolio. Buffett argues he has deep knowledge of those businesses, giving him an edge. For most investors, this is not replicable; Buffett's decades of experience and capital allow him to evaluate 10 companies better than you can. His model is exceptional, not instructive.

The Vanguard Total Stock Market fund: Holds nearly 3,500 stocks. It's massively "overdiversified" by the math, but the cost is low (0.03% annually), so overdiversification is free. Every investor gets the exact market return minus tiny fees.

A typical DIY investor portfolio: 30–40 stocks across 8–10 sectors, 15–20% in international, 20–30% in bonds. This captures >90% of diversification benefit with manageable complexity. Research time: 5–10 hours monthly for updates and rebalancing.

A 401(k) holder's portfolio: 4 mutual funds (U.S. large-cap index, international index, bond index, real estate index) totaling 2,000+ holdings across all asset classes. Overdiversified by stock count; perfectly diversified by asset class. Zero research required.

Common mistakes

1. Assuming 500 stocks are 10x more diversified than 50: Due to the square-root law, 500 stocks reduce unsystematic risk only ~10% more than 50. The added holdings provide minimal marginal benefit.

2. Holding correlated positions and claiming diversification: Owning 20 financial stocks is not diversified; you're just betting bigger on one sector. Correlation kills diversification.

3. Believing you're underdiversified without calculating it: Many investors worry they need 100+ stocks. In reality, 25 uncorrelated stocks achieve most diversification. Worry instead about correlation—do your holdings move together?

4. Adding holdings without understanding their correlation: Before adding a new stock, ask: does this move with my current holdings? If yes, it's not diversifying anything. If no, it is.

5. Confusing diversification math with actual outcomes: The theory assumes uncorrelated returns. In crises, everything correlates. Diversification works best in normal times; understand its limits during panic.

FAQ

Q: If 30 stocks give me 90% of the benefit, why do index funds hold 500+? A: Liquidity and rebalancing efficiency. Large funds can hold more names with minimal cost increase. The overdiversification is free and doesn't hurt.

Q: How do I know if my stocks are uncorrelated? A: Calculate historical correlation (available on most finance websites). Aim for average correlations of 0.3–0.5 within equities. Add bonds (correlation ~-0.2 to equities) or alternatives (often very low correlation) to genuinely diversify.

Q: Is 20 stocks reasonable for a part-time investor? A: Yes. Allocate 2–3 hours monthly to research and rebalancing. If that's too much, use one or two broad index funds instead.

Q: Does sector diversification matter if I own an S&P 500 index? A: No. The index already includes all sectors in market-weighted proportions. You're automatically diversified by sector.

Q: Can I diversify with just bonds and stocks, or do I need alternatives? A: Bonds and stocks work well together (0.3–0.5 correlation typically). Alternatives (gold, real estate, commodities) can help in crises but add complexity. A simple 60/40 stock-bond portfolio is adequately diversified for most investors.

Q: How many stocks do I need if I rebalance quarterly? A: Rebalancing increases the benefit of each holding, so 15–20 stocks might suffice. Without rebalancing, aim for 25–30 to be safe.

Related concepts

• Efficient frontier: The set of portfolios offering maximum return for a given risk level; portfolios below it are suboptimal.
• Correlation coefficient: Ranges from -1 (perfect negative correlation) to +1 (perfect positive correlation); near 0 is ideal for diversification.
• Standard deviation: Total volatility; the metric used to measure portfolio risk.
• Rebalancing: The process of selling winners and buying losers to maintain target allocations; increases diversification benefit.
• Unsystematic risk: Company-specific danger; reduced by increasing number of uncorrelated holdings.

Summary

The mathematics of diversification is unforgiving: you need roughly 20–40 uncorrelated stocks to eliminate 85–90% of unsystematic risk. Beyond that, marginal benefit declines steeply. The practical upshot: a diversified portfolio need not be enormous. Twenty carefully chosen stocks across different sectors beat fifty correlated ones. An index fund (overdiversified by math) beats both because its cost is negligible. Choose your path based on the time you can commit to research, but understand the numbers: diversification yields diminishing returns, and the first 30 holdings capture the bulk of the benefit.

Next

Understand the danger of owning too many stocks—when diversification becomes di-worsification, and how mediocre holdings dilute portfolio returns.