Skip to content

Fire, Then Ice: summary and key takeaways

The note uses a long UK history of equity and bond returns to argue that the stock-bond correlation depends on the inflation regime and the level of bond yields. It says the negative correlation seen since 1998 is historically unusual and may flip positive if inflation rises.

Funnell, B. · 2017 · Man InstituteSummary updated 20 November 2026Practitioner research
Published2017Firm research
Asset focusMulti-assetInflation regimes
TypeResearchInflation & Macro
Reading time4 minthis summary

Fire, Then Ice at a glance

The question
Does the sign of the stock-bond correlation follow the inflation regime and the level of bond yields, and what would a flip back to positive mean for equity portfolios?
Dataset & period
UK equities and UK consols (perpetual bonds first issued in 1752), ten-year trailing correlation of monthly returns, chart covering 1764-2017. Sample size: Not stated in the paper.
Method
The note shows a rolling ten-year correlation of UK equity and consol returns over centuries, built from spliced datasets. It compares three episodes when the correlation fell towards zero or below, looking at inflation and bond yields. It then uses a dividend discount (Gordon) argument and a four-zone chart of inflation direction and level to explain the pattern.

Key findings of Fire, Then Ice

  • Stocks and bonds have been negatively correlated since 1998, which the note presents as unusual.
  • In UK data, stocks and bonds moved together across the 250 years before this century, with no earlier period of clearly inverse behaviour.
  • The correlation fell close to zero in only two extended earlier periods, 1803-1821 and 1861-1895.
  • In all three low-correlation episodes, inflation slowed and bond yields fell below trend nominal GDP growth. The correlation bottomed at about the same time as yields.
  • The note's four-zone framework says the correlation is positive in disinflation and inflation, and negative in deflation and reflation.
  • The author guesses the correlation could turn positive when US ten-year yields reach roughly 3 to 4%, near trend nominal GDP growth.

Robustness, as reported in the paper

  • The note says the 1974 dip towards zero was short-lived and tied to an oil-price shock and a government order for pension funds to buy equities.
  • The note says the pattern is based on only three episodes, though it calls the trends clear.

Caveats and limits

  • The note is a short firm research piece with no formal statistical test, standard errors or out-of-sample test (inferred).
  • The author admits the theoretical case is weak: the Gordon growth model implies no link between bond yields and the equity risk premium.
  • Long-run UK data are spliced from several sources, and the 3 to 4% threshold for the US is an opinion the author says cannot be known precisely.
  • Most of the support is chart and table based, and no trading costs or portfolio results are reported (inferred).

Items marked (inferred) are our reading of the design, not statements by the authors.

Why it matters

The note frames the stock-bond hedge as regime dependent rather than permanent. A practitioner can use it as a reference for how inflation and yield levels have lined up with correlation in UK history. It is an opinion piece, so its forecasts need independent testing.

This is a firm research note or web article, not a peer-reviewed paper.

Questions about this paper

How long have stocks and bonds been negatively correlated?

The note says the correlation has been negative since 1998, after stocks and bonds moved together for about 250 years in UK data.

What does the note link the correlation to?

It links it to the inflation regime and the level of bond yields. Higher yields went with a positive correlation and lower yields with a lower, zero or negative one.

At what yield might the correlation turn positive?

The author suggests around trend nominal GDP growth, which he puts at roughly 3 to 4% for the US ten-year yield, while saying the exact level cannot be known.

Where can I read the original paper?

The original is at its publisher: https://www.man.com/insights/fire-then-ice. We link to the paper and do not host the PDF.

Follow new quant research

Download MacroFinance market data as CSV or Excel, or pull it by API, and test the ideas in these papers yourself. New summaries are added as we review them.

Create a free account
Summaries are written by MacroFinance from the published papers, in our own words, for information only. They are not investment advice and may omit detail. Papers are not hosted here: each page links to the original at its publisher.