Inflation-linked gilt funds lost far more than inflation would suggest when UK RPI peaked at 14.2% in 2022. Short-dated linkers recovered within 15 months, while the longest-dated funds remain more than 40% below their peak. In our Fixed Income Research Update, Laurentius van den Worm, CFA, Head of Investment Strategy at Timeline, took UK financial advisers through two new research papers on inflation-linked gilts and global corporate bonds.
Here is what the evidence says about whether the inflation link survives inside a fund, why duration explained most of what happened in 2022, and how short-dated credit has behaved as a second source of return in defensive allocations.
The research started with what happened in 2022, and a simple question: what is each part of the bond market actually there to do, and how does it behave in unusual conditions? Inflation-linked gilts were the obvious place to begin.
An individual inflation-linked gilt adjusts both its principal and its coupon in line with inflation. Hold it to maturity and the investor receives the contractual, inflation-adjusted payments. Before maturity, however, its price moves with the market like any other bond.
A linker fund works differently. It holds a spread of maturities, buys bonds after issue, often sells them before maturity and continually reinvests. It has no maturity date of its own, so there is no point at which the inflation link is locked in. Changes in real yields can outweigh the inflation adjustment, particularly for longer-dated bonds.
Because Timeline portfolios invest through bond funds, the research focuses on how funds have behaved rather than the contractual payments of individual bonds.
The team split the UK inflation-linked market into maturity segments and tracked each from December 2017 to September 2026. The full index had a duration of around 23 years going into 2022, among the longest of any major bond market. Duration measures sensitivity to interest rates: a duration of 10 implies roughly a 10% fall in price for a 1% rise in yields.
The pattern is consistent: the longer the duration, the larger the loss. In calendar year 2022 alone, linkers over 15 years fell 48.3%, while linkers up to 10 years fell 3.2%. The shortest segment, up to five years, was excluded as there is too little liquidity to support an investable fund.
Over the full period from December 2017 to September 2026, linkers up to 10 years returned 2.2% a year with volatility of 4.0%, ahead of global short-dated bonds (1.5%) and global short-dated corporate bonds (1.9%). Linkers over 15 years returned -6.8% a year. No bond segment kept pace with RPI in real terms over the period.
The drawdowns were not caused by inflation itself but by the policy response to it. The team compared three recent periods of accelerating RPI. Between September 2016 and January 2018, Bank Rate rose just 0.25 percentage points and every bond segment fell modestly. Between April 2021 and March 2023, Bank Rate rose 4.15 percentage points: linkers over 15 years lost 39.5% and UK gilts 23.5%, while linkers up to 10 years gained 4.9%.
In the third period, April to September 2025, Bank Rate fell by 0.5 percentage points, yet long-dated linkers still lost money. Long bonds are priced on expected future rates, and markets did not treat the cut as the start of a sustained trend.
correlation between all-maturities and 5 to 15 year linkers and UK gilts over their full history, rising to 0.91 and 0.90 over the three years to September 2026.
Correlations between monthly linker returns and UK RPI ranged from just -0.21 to -0.04 across the four series, effectively zero. Over rolling 12-month periods, linker returns only turned sharply negative against inflation when Bank Rate responded aggressively. In other words, the inflation-linking mechanism does not stop linker funds behaving much like conventional gilts in the short term.
To test the longer view, the team identified nine episodes since 2000 in which trailing 12-month RPI accelerated by at least one percentage point. The UK inflation-linked index delivered a positive return in seven of the nine and beat RPI in five. The two exceptions were September 2020 to October 2022, when linkers fell 28.8% as RPI rose 21.4%, and October 2024 to July 2025, when they fell 7.0% as RPI rose 4.5%.
Longer licensed data, which cannot be published, told the same story. Inflation-linked does not mean inflation-proof, and the protection was conditional on how sharply rates moved. Over the long term, equities remain the main driver of real growth in a multi-asset portfolio, with linkers playing a defensive and diversifying role alongside them.
Past performance is not a reliable indicator of future results.
Bond returns are usually explained by two risk premiums. The term premium rewards investors for holding longer-dated bonds. The credit premium rewards them for lending to companies rather than governments, since companies cannot print money to repay their debts and can default.
The second paper asked how much credit exposure a global aggregate bond fund actually provides. Using the Vanguard Global Bond Index Fund as a proxy at 30 September 2026, the answer is less than many expect.
Most of the remainder is securitised debt, such as mortgage-backed bonds, which sits between the two. The sovereign share has grown over two decades of quantitative easing. Global aggregate bonds (GBP hedged) fell 13.5% in 2022, their worst year since 2000.
From October 2018 to September 2026, corporate bonds beat their aggregate equivalents at both ends of the maturity range. Short-dated global credit was the strongest bond series in the study, and the only one to come close to cash.
The comparison with cash says more about the period than about the long term. Cash has zero duration, so it inevitably does well while rates are rising quickly.
Unlike short-dated linkers, short-dated credit earned more but also had a deeper maximum drawdown than its aggregate equivalent: 9.2% against 7.3%. A corporate bond is tied to the same company as its shares, so it shares some of that company's risk. In the COVID-19 sell-off of early 2020, short-dated credit fell 4.2% against 1.5% for global aggregate bonds, and corporate bonds also lagged during the 2022 rate shock. Historically, credit has tended to fall alongside equities in severe bear markets, while government bonds have been more stable.
Past performance is not a reliable indicator of future results.
To isolate the credit premium from interest rate risk, the team compared short-dated global credit (1 to 5 years), with an effective duration of 2.7 years, against all-maturity global corporate bonds at 5.7 years. The 3-year hit rate is the share of rolling three-year periods in which each beat its aggregate equivalent.
Short-dated credit captured a smaller premium in absolute terms but more efficiently, with an information ratio of 0.28 against 0.23, around half the volatility and half the drawdown.
The team then simulated a hypothetical blend of 85% global aggregate bonds and 15% short-dated global credit, rebalanced monthly. It does not represent any Timeline portfolio.
The blend also had lower volatility (4.29% against 4.58%) and had fully recovered within 40 months, while the aggregate index has not yet recovered. By September 2026 it was 1.5% ahead cumulatively, but only after lagging through 2019 and 2020.
Future returns from any credit allocation will also depend on starting credit spreads and implementation costs.
Simulated past performance is not a reliable indicator of future results. Past performance is not a reliable indicator of future results.
Within linkers, shorter duration meant lower volatility, smaller drawdowns and smaller 2022 losses.
Linker outcomes depended on the Bank of England's response, not inflation alone.
All-maturities linkers correlated 0.84 with gilts over their history and 0.91 over three years.
A credit premium with less duration risk, though it could lag when rates fall or spreads widen.
This research was carried out for Timeline's Investment Committee and has informed the current review of the Classic portfolios. Laurentius did not comment on specific portfolio changes, and the outcome will be shared with advisers separately.
Want to go through the research in more detail? Our investment team is always happy to talk it through, or book time with your regional rep for a personalised demo.
Sources and notes: Inflation-linked gilt research: Timeline analysis using Morningstar monthly GBP total returns to 30 September 2026, with UK RPI from the Office for National Statistics (series CHAW and CZBH) and Bank Rate from the Bank of England. Maturity segments use fund proxies: Legal & General over-15-year index-linked gilt index fund, Vanguard UK inflation-linked gilt index fund, Dimensional sterling inflation-linked intermediate duration fund and iShares up-to-10-years index-linked gilt index fund. Long-run episodes use the Morningstar UK Treasury Inflation-Linked Securities Index from 2000. Credit research: Timeline analysis using Morningstar data, GBP hedged, October 2018 to September 2026, with SONIA from the Bank of England. Global aggregate bonds use the Morningstar Global Core Bond Index; short-dated aggregate, short-dated corporate and global corporate bonds use Vanguard index funds. Sector weights: Vanguard Global Bond Index Fund, 30 September 2026. Effective duration: Morningstar, 30 September 2026. Fund proxies are net of charges and indices gross of fees. The 85/15 blend is hypothetical and does not represent any Timeline portfolio. Funds and indices are named only to represent asset classes and are not a recommendation.