You’ve never had it so bad

Unlock the Editor’s Digest for free

Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.

As we approach 2024, the humungous rally of recent weeks has taken the edge off the bond bear market. Not long ago, we compared the drawdowns suffered by today’s bondholders to those experienced in the 1970s. Today we’re venturing further back into the British and American archives to see how 2023 stacks up against the long long run.

The Bank of England publishes macro time series of a frankly ridiculous length. If thirteenth century wool export volumes are your thing, you probably already knew this. Magically, they include monthly bond yields stretching back to 1753.

We wanted to produce a chart of bondholder drawdowns, but there’s a problem. Before 1935, yields on the mix of short and medium-term bonds that the Bank splices into a 10yr bond yield series are not readily available. What is available is a yield series for 3 per cent Consols — perpetual bonds with a much higher price sensitivity to yield movement.

We could either show a chart of the wild ride enjoyed by perpetual bond holders, or normalise the yield data to show what would’ve happened to a holder of hypothetical ten-year gilts. We chose the latter, understanding that our approach may make some people upset. If this is you, there is a comments sections at the bottom of this post to help you deal with the emotional fallout.

To the data! It turns out that it really has been an exceptional time to hold gilts. Loss-averse holders of (hypothetical) ten-year bonds have never had it so bad, even taking account of the recent Santa Rally. Use the filter to see how these drawdowns (total loss from peak, taking income into account) relate to yields.

Looking across the water to the United States, 2023’s nominal drawdown is pretty impressive in a 230-year context. But something even nastier happened to risk-averse bondholders in 1842.

Federal debt had actually been paid off by 1835 under President Jackson. But we were keen to show a long-run drawdown series, and so used a data-series published in November by the Financial Analysts Journal that reflects, prior to 1857, a par-weighted total return series of federal, municipal and corporate bonds (with an equal-weighting between 1857-1871).

As Sidney Homer writes in his History of Interest Rates:

During the years 1835-1841, when there were no treasury issues outstanding, the yields on local issues rose. … This seven-year debtless interval was in fact one of great financial disturbance. The second Bank of the United States had lost its charter, and federal deposits had been transferred to state banks. A period of wild speculation had ended in the collapse of 1837, which had been succeeded by a heavy depression.

Great financial disturbance sounds like bondholder manna. But sinking muni prices overwhelming drive the dog years of 1842-43, and this note from the Atlanta Fed helps explain why:

…American states embarked on ambitious plans of infrastructure investment in canals and banks financed by extensive public borrowing, much of it from foreign investors. Apparently ambition exceeded ability. By 1842, eight states and the Territory of Florida were in default on their loans. Four states would ultimately repudiate all or part of their debts. It was a debt and default crisis not unlike late 20th century emerging-market crises.

Oops.

Drilling down into the FAJ spreadsheets, 1842 returns are driven by abysmal returns from bonds issued by Pennsylvania and Maryland, which collectively accounted for around half their ‘index’. As the Atlanta Fed notes:

Northeastern states raised property taxes to cover debt service. Pennsylvania and Maryland, however, temporarily defaulted because of delays in implementing tax increases.

Oops again.

The US current and historical data is not an apples-to-apples comparison. After all, debts from state governments leveraging-up on the back of potential future land taxes are not ‘safe assets’ if the central government is not willing to stand behind them. Just ask Chinese lenders. And we are not comparing constant maturity bond records (as we have contorted ourselves to do for the UK). But the fact that the drawdown on US bonds in 2022-3 got pretty close to levels surpassed only during widespread (if temporary) issuer defaults 180 years ago is quite a thing.