Liquid real assets: inflation protection with tradable real assets in a portfolio

In depth11 min read · 13 sources

Liquid real assets are exchange-traded real asset investments meant to make a portfolio more robust against inflation: inflation-linked bonds, commodity futures, gold, listed infrastructure, REITs and shares of resource, agricultural and forestry companies. No single building block protects reliably in every inflation regime. Together, however, they can narrow the gap that equities and nominal bonds leave when inflation surprises. Here you will learn how inflation protection is measured, what research shows for each building block and how to build a portfolio component from them.

What counts as liquid real assets

What is meant is a sub-portfolio, a sleeve, of liquid real assets that can be traded and rebalanced daily. Typical building blocks are:

• Inflation-linked government bonds (linkers) such as US TIPS or euro linkers: the principal and the coupon payments grow with the price index.
• Commodity futures via broad indices of energy, metals and agricultural commodities, with spot, roll and collateral return (see active or passive in bonds and commodities).
• Gold as a precious metal without ongoing income (see gold in a portfolio).
• Listed infrastructure such as utilities, grids, toll roads and pipelines, whose charges are often linked to inflation.
• REITs and real estate stocks, whose rents can follow prices over the medium term.
• Natural resource equities: commodity producers as well as agricultural, forestry and water companies.

They should be distinguished from private real assets: unlisted infrastructure, direct real estate, farmland and forests. These can be more closely tied to inflation: according to Wurstbauer and Schäfers (2015), in the US only direct infrastructure provided a partial short-term hedge against inflation. However, they are illiquid and only valued by estimate; their smoothed returns understate the risk (see measuring private markets). Liquid real assets trade this smoothing for daily prices, with equity risk in the case of REITs and infrastructure.

Inflation beta: how to measure inflation protection

A good inflation hedge delivers a higher nominal return when inflation rises, ideally percentage point for percentage point. Bekaert and Wang (2010) measure this with the inflation beta, the slope of a regression of the nominal return on inflation. A beta of 1 means perfect protection: the real return is unaffected.

It is important to separate expected from unexpected inflation. Expected inflation is already reflected in interest rates and prices; it is mainly the surprise that can do damage. Fama and Schwert (1977) therefore split inflation into both parts and found for the US that equities and bonds protected reliably against neither. For a portfolio, what matters most is the beta to unexpected inflation. The estimates by Bekaert and Wang for developed markets show how weak the protection of traditional assets is:

Source: Bekaert and Wang (2010), Tables 1 and 5: pooled regressions across countries with a separate intercept for each country, data mostly up to January 2010.

The building blocks in inflationary periods

Broadly speaking, research agrees. Over 95 years in the US, the UK and Japan, Neville et al. (2021) found that unexpected inflation is bad news for equities and bonds. Commodities delivered positive returns in inflation surges, but with large differences between commodities; among active strategies, trend following protected most reliably. For real estate stocks the picture is weaker: for US REITs from 1972 to 1992, Yobaccio et al. (1995) found partial protection against expected inflation, but a perverse reaction to unexpected inflation. According to Rödel and Rothballer (2012), listed infrastructure in a large international sample hedged no better than other stocks; only companies with strong pricing power did slightly better.

Worked example: the inflation beta of a sleeve

The inflation beta of a portfolio is the weighted average of the betas of its building blocks. The betas below are illustrative values for the reaction to an inflation surprise, not estimates. The range of actual estimates is wide.

If you weight the building blocks by inverse volatility instead (illustrative values: linkers 6%, commodities 16%, gold 15%, infrastructure 14%, REITs 18%), linkers get 39.4% and commodities only 14.8%. The inflation beta falls to 1.23: the sleeve becomes calmer, but less sensitive to inflation. And in the overall portfolio, it only works in proportion to its weight: a 15% sleeve with a beta of 1.38 and an 85% core with a beta of −1 together give −0.64.

Building an LRA sleeve

• Set the objective: should the sleeve cushion inflation surprises (high beta, i.e. more commodities) or preserve purchasing power over years (rather linkers with a suitable maturity)?
• Weight by inflation sensitivity: building blocks with a high beta per unit of risk get more weight. Because the betas are estimated imprecisely, upper and lower limits help.
• Risk parity: equal risk contributions instead of equal capital shares. Maillard et al. (2010) show that the volatility of such portfolios lies between that of minimum variance and equal weighting. Inverse volatility is the simple approximation without correlations.
• Rebalancing: commodities and gold are highly volatile and often move differently. Regularly resetting to the target weights sells after rises and buys after declines (see rebalancing); according to Erb and Harvey (2006), a rebalanced commodity portfolio can deliver excess returns even though individual futures had hardly any on average.
• Set the size: calculate the beta of the whole portfolio, i.e. share × sleeve beta plus the rest × beta of the core. As the example shows, a small sleeve rarely reverses a negative beta.

Costs and implementation

Almost all building blocks are available as index funds: linker ETFs, commodity ETFs, which replicate futures directly or via swaps, physically backed gold funds and ETCs, and ETFs on infrastructure, real estate and resource equities. Besides ongoing costs, what matters is the roll yield for commodities, the tax treatment and the issuer risk of ETCs. With foreign linkers such as euro linkers without currency hedging, you also bear currency risk, which can swamp the inflation protection in dollars (see currency risk). Funds on North American pipeline partnerships (MLPs) have their own tax peculiarities. This is not tax advice.

For advanced readers: horizon, estimation error and premiums

Horizon: in Bekaert and Wang (2010), the inflation beta of bonds rises with the horizon and, at five years, is no longer significantly different from 1 in developed markets; that of equities remains well below even then. Against unexpected inflation, bonds in developed markets remain clearly below 1 even over five years. Short-term and long-term protection are therefore different questions.

Estimation error: the standard errors are large. For real estate in North America, Bekaert and Wang estimate a beta of 2.04 with a standard error of 4.58. That is practically meaningless. Inflation surges are rare and have had different causes, such as oil price shocks or overheated demand; betas from one episode do not carry over reliably to the next. Bekaert and Wang therefore also test whether the betas change with the monetary policy regime, for example after the introduction of inflation targets.

Premiums in the breakeven rate: the difference between the nominal and the real yield is not a pure inflation forecast. It contains an inflation risk premium, which according to Bekaert and Wang (2010) has mostly been positive, often substantial, and varies over time, and a liquidity premium on linkers (D'Amico et al. 2018). The risk premium makes nominal bonds slightly more rewarding on average (with a linker, you pay for the insurance), while the liquidity premium works the other way.

What the research shows

• Bekaert and Wang (2010): in more than 45 countries, standard bonds and broad equity indices are poor inflation hedges. Money market instruments, foreign bonds, real estate and gold improve the picture, but inflation risk remains hard to hedge. That is why linkers play an important role.
• Attié and Roache (2009) used an error correction model to examine how well traditional asset classes hedge inflation over different horizons in a diversified portfolio. According to the summary by Bekaert and Wang (2010), bonds, equities and real estate have negative betas to unexpected inflation there, while gold and commodities have positive ones; even in the long run, a portfolio of traditional assets is hard to protect against unexpected inflation.
• Gorton and Rouwenhorst (2006): from 1959 to 2004, fully collateralized commodity futures were positively correlated with inflation, unexpected inflation and changes in expected inflation, and negatively correlated with equities and bonds.
• Neville et al. (2021): in inflation surges, passive equity and bond strategies performed poorly, commodities well but unevenly; trend following and some factor strategies such as quality offered partial protection.
• Erb and Harvey (2013) take a critical look at the claim “gold protects against inflation”: the real gold price was historically high in 2013 and could revert to the mean.
• REITs and infrastructure: no reliable short-term protection, rather a long-term co-movement with the price level (Yobaccio et al. 1995; Chatrath and Liang 1998; Wurstbauer and Schäfers 2015).

Limitations and common mistakes

• In the short term, everything works differently: over months, interest rates, the economy and sentiment decide; an inflation hedge can lose money even in an inflationary year.
• Real interest rate risk of linkers: long linkers react strongly to the real interest rate. In 2022, real interest rates rose so sharply that long linkers suffered price losses despite high inflation; short linkers track current inflation more directly.
• Equity beta: REITs and infrastructure are equities. In crises they fall with the market, and as interest-rate-sensitive sectors they suffer from rising rates, often precisely when inflation is rising.
• Roll yield: commodity futures do not deliver the spot price; in contango, rolling costs return (see roll yield).
• Inflation has causes: our assessment: an energy shock lifts commodities, a wage-price spiral less so; the building blocks react differently depending on the cause.
• Currency: you do not automatically protect against inflation at home with assets in a foreign currency. Bekaert and Wang point out that gold betas outside North America may also contain exchange rate effects.
• Few episodes: a backtest over one inflationary period is a case study, not statistics.

This article is not investment advice.

How Portfolio Lab calculates it

Under Asset classes you will find the building blocks as long monthly series in euros, chained from fund and index data: “Inflation-linked euro government bonds” (@EURO_LINKER), “Inflation-linked US Treasuries (TIPS)” (@TIPS) and UK index-linked gilts (@UK_LINKER), “Commodities broad” (@ROHSTOFFE) with sub-baskets such as energy, industrial metals and agriculture, “Gold” (@GOLD), “Listed infrastructure” (@INFRASTRUKTUR), “Listed real estate (REITs World)” (@REITS), “Timber and forestry (listed)” (@TIMBER), “Water” (@WASSER), “Gold miners (World)” (@GOLDMINEN) and “Energy infrastructure (MLPs)” (@INFRA_PIPELINES). Series in foreign currencies are converted into euros without hedging. Asset classes work like tickers everywhere; with them, the Backtest calculates monthly. How the series are built is described in the asset classes methodology.

In the Backtest you can switch on real values: the app deflates with the price index of the chosen region and also shows the final value and CAGR in real terms. The regime analysis under Macro sorts the months into inflation bands of below 0%, 0–2%, 2–4% and above 4% year-on-year inflation, or into phases of growth and inflation (Goldilocks, reflation, stagflation, cooling). For each state, it shows the return p.a., volatility and worst month of your portfolios, optionally with a lag (see macro methodology). This lets you see how a sleeve performed in months with inflation above 4%, though with great uncertainty because there are few such months. Optimization offers the objective “Risk parity (equal risk contributions)”. The app does not estimate an inflation beta for each building block directly.

This article explains a method; it is not investment or tax advice. Worked examples are simplified and rounded. All terms in the glossary.