Diversification

“Don’t put all your eggs in one basket” – grandmother’s wisdom is the core of modern portfolio theory. Harry Markowitz won the Nobel Prize in 1990 for working out the mathematics, and diversification is often called the only “free lunch” in finance: the one thing that lowers risk without necessarily lowering expected return.

How diversification works#

Imagine you own a single stock. Your fate hangs on one company: its management, products, competitors, a single scandal. Own two stocks from different industries and their individual bad luck partly cancels out. With twenty stocks, one total loss barely registers.

Try it yourself – see how quickly risk falls as you add stocks:

Two things stand out:

1. The first few stocks do the most. With 10 to 20 stocks from different sectors, volatility already falls a lot.
2. A floor remains that can’t be diversified away, no matter how many stocks you add. That’s the risk of the market as a whole.

Two kinds of risk#

The last row is crucial: markets only reward risk that can’t be diversified away. Choosing to carry unsystematic risk – say, a portfolio of three favourite stocks – means more risk, but no higher expected return for it.

Correlation: the key#

How well diversification works depends on how similarly investments move. That’s measured by correlation, a number between −1 and +1:

• +1: both always move exactly together. Mixing does nothing.
• 0: no relationship. Mixing lowers risk considerably.
• −1: they always move in exactly opposite directions. In theory, risk can be eliminated entirely.

In reality most stocks are positively correlated because they all depend on the economy. Between stocks and high-quality government bonds, correlation has often been low or even negative in recent decades – which is why combining the two is so popular.

Why it matters so much: winners are rare#

A striking finding from finance research: most individual stocks do poorly. Economist Hendrik Bessembinder studied every US stock since 1926. More than half earned less over their lifetime than one-month Treasury bills. The entire net wealth creation of the US stock market over Treasury bills came from only about 4% of stocks.

Diversifying on several levels#

Good diversification has several dimensions:

• Across companies and sectors: not just technology, not just banks.
• Across countries and regions: no country led in every decade. Japan was the star of the 1980s, the US after 2010.
• Across asset classes: stocks, bonds and perhaps real estate or gold respond differently to growth, rates and inflation.
• Across time: investing regularly means buying at many different prices.

Home bias#

Investors almost everywhere over-invest in their home market. The UK is about 3% of world stock market value, Germany about 2%, yet domestic stocks make up a far larger share of many local portfolios. US investors are less affected, since US companies make up roughly 60% of global stock indices – but even they often hold little outside the US. This home bias feels familiar but concentrates risk: job, home and portfolio all depend on the same economy.

Diworsification: too much of a good thing?#

Legendary fund manager Peter Lynch coined “diworsification” for diversification that only adds cost and complexity. Owning ten different global ETFs that all hold the same stocks isn’t diversification, it’s clutter. Good diversification doesn’t mean “as many products as possible”, but “as many different sources of risk as sensible”.

Summary#

Diversification lowers risk by offsetting the individual fortunes of single investments. How well it works depends on correlation. Company risk can be diversified away almost completely, market risk can’t – and only the latter is rewarded with higher expected returns. Because a few stocks produce most of the gains, broad diversification is the best insurance against missing them.