Diversification is not the number of funds you own

There is a short answer about the urge to concentrate a portfolio and a useful one, and they are not the same. What follows is the useful one.

Counting funds measures nothing

Holding several funds that track overlapping indices produces one exposure at multiple fee layers. The relevant question is what the combined holdings actually own, not how many products are involved. Looking through to underlying holdings usually reveals substantial duplication.

Few platforms do that look-through for you, so the practical method is listing the index and top holdings of each fund side by side once, which takes an evening and rarely needs repeating.

Real diversification is across drivers

Different geographies, company sizes, sectors and asset classes respond differently to the same conditions. A global equity fund is well diversified across companies and remains a single asset class. Adding bonds changes the risk profile far more than adding a fourth equity fund does.

The useful part is this: currency is a driver in its own right, and an unhedged global fund carries an exchange-rate exposure that can dominate a few years of returns while the underlying companies do nothing unusual.

Home bias is the usual concentration

Investors across almost every country hold far more of their domestic market than its share of global markets would suggest. That concentrates the portfolio in the same economy that pays your salary and, often, owns your house. Some home bias is defensible for currency reasons; a large amount is a concentration decision worth making deliberately.

Your own earnings point the same way, since someone whose employer, pension and property all sit in one economy already holds a large undiversified position in it before buying a single fund.

Correlation moves in a crisis

Assets that appear uncorrelated in normal conditions frequently fall together in a severe sell-off. Diversification reduces risk over ordinary periods and offers less protection in the worst weeks.

This is an argument for holding some genuinely defensive assets rather than assuming spread equals safety. Correlations are estimated from past data and are not fixed properties, so a portfolio built on a historical relationship between two assets depends on something that has shifted before and can shift again.

Simplicity usually wins

A single global tracker plus a bond allocation is diversified by any reasonable measure and needs almost no maintenance. Every additional holding is another thing to rebalance, monitor and eventually rationalise.

Complexity is easy to add and difficult to unwind, particularly in a taxable account. Where holdings have already accumulated, unwinding them gradually using new contributions and whatever annual tax allowance applies is usually cheaper than one tidy-up trade.

The diversification you cannot buy

Everything held in listed markets shares some exposure to the same global conditions, so no combination of funds removes the risk that markets in general do badly for a decade. Holding cash and short-dated bonds for near-term spending addresses that in a way spreading equities cannot, because it removes the need to sell into the decline at all.

Put simply, adding exotic holdings to chase a low correlation usually buys cost and complexity in exchange for a relationship the evidence supports only weakly. The remaining lever is time: a horizon long enough to let a bad decade be a bad decade rather than a forced sale does more than any allocation refinement, which is why the horizon question comes before the fund question.

The takeaway

Look through to what you own, not at how many products you own.

Pick the one that costs you least, and let the rest wait.