Gold, commodities and alternatives

Assets that produce no income are valued entirely by what someone else will pay, which is a different proposition from assets valued by their cash flows.

Gold

The most discussed.

It produces no income and has storage and insurance costs, meaning the return depends entirely on price change.

Its long-run real return has been positive and modest, considerably below equities over long periods, with very long stretches of poor real performance.

Its claimed roles: inflation hedge, crisis hedge and currency debasement hedge.

The evidence on the inflation hedge is weaker than the claim: gold has performed well in some inflationary periods and poorly in others, and the correlation over shorter horizons is unreliable.

The crisis hedge case is stronger, with gold frequently rising during acute market stress, though not consistently.

Which makes a modest allocation defensible for diversification and makes a large one a substantial bet on a specific outcome.

Commodities more broadly

A different proposition again.

Broad commodity exposure has historically shown low correlation with equities and bonds and has performed well during inflationary periods driven by commodity prices.

The complication is how exposure is obtained: most investors hold futures-based products rather than physical commodities, and the return from rolling futures contracts can differ substantially from the spot price change.

Which means a commodity fund can lose money over a period when the underlying commodity price rose, a feature that surprises holders.

Long-run real returns from broad commodity indices have been modest and volatile.

The general question

What an asset without cash flow is worth.

Equities and bonds have an expected return derived from cash flows: earnings, dividends, coupons.

Assets without cash flow have an expected return derived from expectations about future prices, which is a weaker foundation.

This does not mean they are worthless — scarcity, monetary properties and industrial demand are real — and it does mean that valuation is harder and that the case for a long-run positive real return is less clear.

Cryptocurrency

Which belongs in this discussion.

It produces no cash flow and is valued entirely by expectation.

Volatility has been very high, with multiple declines of substantial magnitude.

Correlation with equities has varied and has been higher than the diversification argument suggests during several stress periods.

Regulatory treatment varies enormously and is evolving.

Consumer protections are limited or absent in many jurisdictions, and losses from exchange failures, fraud and lost keys are substantial and unrecoverable.

Which means that whatever view one takes of the technology, it is not a substitute for savings and should not be held in amounts whose loss would matter.

Other alternatives

Briefly.

Hedge funds and absolute return strategies, where the average net-of-fee record has been unimpressive and where dispersion between managers is large.

Private equity, where reported returns are affected by valuation practices and where access, fees and lock-ups are significant.

Infrastructure, which has some inflation-linkage and is generally accessed through listed vehicles for retail investors.

Collectibles — art, wine, watches, cars — where the returns quoted are generally indices with substantial survivorship bias, costs are high, liquidity is poor and enjoyment is a legitimate part of the return.

And peer-to-peer lending, which has produced substantial losses in several markets and where the risk is credit risk rather than a novel asset class.

The case for a small allocation

For balance.

Genuine diversification requires assets that behave differently, and a portfolio of equities and bonds is exposed to environments in which both perform badly — as recent inflationary periods demonstrated.

A modest allocation to assets with different drivers is defensible on that basis.

The practical caution is that the allocation should be small enough not to matter if it fails and large enough to matter if it works, which is a narrow range.

And that complexity has a cost in fees, understanding and behaviour.

The marketing to be sceptical of

Recognisable claims.

Anything described as uncorrelated, which generally means uncorrelated until it matters.

Anything promising equity-like returns with bond-like risk.

Historical returns from indices with survivorship bias.

Products launched after an asset class has performed well.

Complexity presented as sophistication.

And anything requiring you to accept illiquidity without a clearly quantified premium for doing so.

The reasonable position

Which most evidence supports.

A portfolio of broadly diversified global equities and high-quality bonds, at an allocation matched to circumstances, covers what most investors need.

A small allocation to gold or broad commodities is defensible for diversification and is not necessary.

Everything else should clear a high bar of explanation before entering a portfolio.

And anything you cannot explain simply is generally being sold rather than bought.

General information only, not investment advice. Investments can fall in value and past performance does not indicate future returns. Consult a regulated financial adviser.