Asset Allocation
Asset allocation is the division of a portfolio among asset classes — equities, bonds, cash, property, alternatives — and within them by region and type, set to match the investor's goals, horizon and tolerance for loss. Studies since the 1980s attribute most of the variation in portfolio returns to the allocation rather than to the choice of securities within it, which is why wealthtech is built on allocation and low-cost funds rather than on stock-picking.
- Also known as
- Strategic asset allocation, Portfolio allocation, Asset mix
- Used at
- every managed portfolio — pension funds, endowments, wealth managers, robo-advisers, and the individual investor's ISA or 401(k)
- Binding
- a mandate or policy; for regulated advice, the suitability rules that tie the allocation to the client's circumstances
- Typical structure
- a strategic allocation (the long-run mix — 60/40 equities/bonds was the convention for balanced portfolios) with rebalancing back to it, and sometimes a tactical layer that tilts it on a view
- Typical robo-adviser practice
- five to ten model portfolios from cautious to adventurous, each a set of ETFs in fixed weights, assigned by a risk questionnaire and rebalanced automatically; glide paths that shift toward bonds as a goal date approaches
- Parties
- the investor, the adviser or platform that sets the allocation, the funds that implement it, and the regulator that asks whether it was suitable
How asset allocation works in practice
An investor with twenty years to retirement and a tolerance for a 30% fall in a bad year is put in an 80/20 portfolio: 80% equities spread across regions and sizes, 20% bonds, implemented in six index funds. Markets move; a year later equities are 85% of the portfolio; the platform sells equities and buys bonds to return to 80/20 — rebalancing, which sells what rose and buys what fell, mechanically. As retirement nears, the allocation shifts toward bonds on a glide path. Nothing in the process picks a stock; the decisions are the weights.
The evidence behind it: a 1986 study by Brinson, Hood and Beebower found that allocation policy explained over 90% of the variation in pension funds' returns over time — a finding often overstated (it is about variability, not level) and never overturned in its main point: the mix matters more than the picks. The corollary built an industry: if allocation is the decision, the implementation should be cheap, and a robo-adviser that sets the allocation by algorithm and implements it in ETFs at 0.1% is doing the valuable part at a fraction of the cost.
The judgement is in the questionnaire and the models. A risk score that maps a nervous first-time investor to 80% equities is a regulatory problem after the first crash; models built on the last decade's correlations fail when the correlations change, as bonds and equities both falling in 2022 showed.
Example
A robo-adviser's model portfolios run from 1 (20% equities) to 10 (100%). A 35-year-old customer's questionnaire — horizon 25 years, comfortable with a 25% fall, no need for income — maps to portfolio 8: 80% equities (global 55%, emerging 10%, small-cap 10%, property 5%) and 20% bonds (global aggregate 15%, inflation-linked 5%), in seven ETFs costing 0.09% a year. In 2022 the portfolio falls 16% — equities and bonds together — and the customer, who expected bonds to cushion, calls. The adviser's answer is the allocation working as designed in a year the design did not anticipate; the customer stays; in 2024 the portfolio is 14% above its 2021 peak, and the rebalancing in 2022 bought equities at the bottom.
Common mistakes
- A risk questionnaire that measures appetite and not capacity — what the customer can afford to lose. - Allocation models built on correlations from one regime. - Rebalancing so often that transaction costs eat the benefit, or so rarely that the portfolio drifts to a risk the customer did not choose. - A tactical layer that is stock-picking by another name. - Treating 60/40 as a law of nature rather than as a convention of a forty-year bond bull market.