Diversification

Why diversification works

Different assets don't move perfectly in sync. When one falls, others rise (or fall less); when one rallies, others may stagnate. Combining them into a portfolio smooths out the path of returns.

The math: portfolio volatility is less than the weighted average of individual asset volatilities, as long as correlations are below 1.0. The lower the correlations between holdings, the more dramatic the smoothing.

This is one of the few "free lunches" in finance. By combining imperfectly correlated assets, you can lower risk without proportionally sacrificing return — provided the correlations behave as expected.

What good diversification looks like

Across multiple dimensions:

• Asset classes — equities, bonds, real estate, cash, commodities. Different classes respond differently to economic conditions.
• Geography — US, developed international, emerging markets. Domestic markets can underperform for years; international exposure smooths country-specific risk.
• Sectors — within equities, holdings spread across technology, healthcare, financials, energy, consumer staples, etc. Avoids concentration in a single industry's fortunes.
• Market capitalization — large-cap, mid-cap, small-cap. Each behaves somewhat differently across cycles.
• Investment style — growth and value; momentum and quality; etc.
• Currencies — for non-US investors particularly, holdings denominated in different currencies hedge currency risk.

The simplest broad diversification: a low-cost global index fund covering thousands of companies across most sectors and major geographies. From there, additional layers (bonds, international tilts, alternatives) refine the exposure further.

How not to do it

Common mistakes:

• "Diversifying" within a single asset class. Owning ten tech stocks isn't well-diversified — they all share the same sector and macro exposures. Genuine diversification requires actually different assets.
• Holding many funds with overlapping exposures. Five different US large-cap mutual funds often hold the same 100 underlying companies in similar weights. The total exposure is the same as one index fund.
• "Diworsification." Adding so many holdings that performance becomes mediocre across the board. Beyond about 30 well-chosen holdings, additional names add little diversification but dilute conviction.
• Concentration in employer stock. Common for executives — but employer concentration creates correlated risk with employment income. Diversifying out of employer stock is usually wise for risk management.
• Concentration in a single geography. US investors with 100% US equity allocations have done well historically but are exposed to single-country political and economic risk.

How much diversification is enough

The diminishing returns curve is relatively steep:

• One stock — 100% idiosyncratic risk. Single-company failure can wipe out the position.
• 5-10 stocks — meaningful but limited diversification. Most idiosyncratic risk reduced; sector and macro risks remain.
• 20-30 stocks — captures most of the diversification benefit available within an asset class.
• 100+ stocks — diminishing returns; index funds operate at this scale.

For most investors, broad index funds get you to "diversified enough" with minimal effort. Active stock picking can be a layer on top of that, but it's adding additional risk in exchange for the chance of outperformance, not adding diversification.

What diversification doesn't do

The honest limitations:

• It doesn't eliminate risk. Markets can fall together (2008, 2020, 2022). Correlations rise during stress, often when diversification is most needed.
• It doesn't help against mismanaged or tax-inefficient portfolios. Holding 30 funds in a taxable account can be a tax mess.
• It can't compensate for systematically wrong direction. A fully diversified equity portfolio still falls 30%+ in major bear markets.
• It doesn't accelerate returns. A perfectly diversified portfolio is, by definition, not concentrated in the best-performing assets. The trade-off is path smoothing, not return enhancement.

In practice

For most investors, the diversification question reduces to a few choices:

• Pick a target-date fund or a few low-cost index funds covering global equities and bonds.
• Set the equity/bond split based on time horizon and risk tolerance.
• Add specific tilts (international weight, factor exposures, alternatives) only if you have a reason and discipline to maintain them.
• Rebalance periodically (annually is fine) to maintain target weights.

The 80/20 rule applies to portfolio construction. Most of the diversification benefit comes from a few simple decisions; further refinement adds incremental value at increasing operational cost.

Crypto and diversification

Crypto presents specific challenges for diversification:

• Most altcoins are highly correlated with Bitcoin. "Diversifying" across 20 altcoins still leaves the portfolio with most of Bitcoin's macro behavior.
• The asset class is small. Even a "well-diversified" crypto portfolio is concentrated relative to a global equity portfolio.
• Tail risk is high. Many crypto assets eventually go to zero. Diversification across many of them reduces but doesn't eliminate this.

For most investors, crypto exposure is best treated as a single asset class within a larger diversified portfolio, sized to a fraction of total assets that survives a 90% drawdown without consequence.