The writer is an FT contributing editor
In July of 1595, the brothers Francisco and Pedro de Maluenda made a loan to Philip II, the Habsburg king of Castile, Aragón, Naples, Sicily, New Spain, Peru and at times England, Ireland, Portugal and the Netherlands. The brothers were financiers, from the Castilian market city of Burgos.
The king had obligations all over Europe, but no administrative way to pay them regularly, so he relied on lenders like the de Maluendas to do it for him. The loan was an asiento, a short-term agreement to make 12 monthly payments to Philip’s army in Lisbon. Philip was to pay the brothers back with the arrival of the silver fleet the next year. Instead of waiting for the fleet, however, they exercised a clause that allowed them to pay themselves back for the short-term asiento by selling juros, long-term loans on which the king paid interest. These included a few perpetual juros — loans on which the king would pay interest forever.
In October of this year, the financier George Soros proposed that Rishi Sunak issue perpetual gilts — bonds that pay interest forever. Sunak is the prime minister of only England, Scotland, Wales and Northern Ireland, but Soros argued that the perpetual would satisfy perplexed institutional investors looking for a stable long-term asset, and he reached back to England’s 1752 consolidated loans, also perpetuals, as a precedent.
There are plenty of other perpetual bonds — including the ones sold by the de Maluenda brothers — that show how historical precedents can be tricky. Traditionally, perpetuals had to be funded. That is, they needed a dedicated, specific source of tax income to guarantee the interest. And they didn’t always predictably behave like long-dated bonds. Forever isn’t just longer than 30 years. It’s a weird length of time.
Crucially, all of the juros the de Maluendas sold were funded — the interest payments were guaranteed by sales taxes, collected by Castile’s cities. As detailed in a 2018 paper by Carlos Álvarez-Nogal and Christophe Chamley, the brothers raised cash by selling juros to 74 Castilian and Genoese investors.
Almost all of them were sold as annuities on heads. The investor picked someone, usually a child, and the juro paid out 14 per cent so long as the child — the head — was still alive. Most investors bought on two heads; they’d pick two children, and the juro would pay out at 12 per cent so long as one was still alive. And only two investors picked the third option, a perpetual juro that paid out only 7 per cent. The yield curve on these annuities was inverted. The longer the duration, the lower the return. And even though forever at 7 per cent seems like a pretty good deal, almost no one took it. The perpetuals weren’t funded any more securely than the other annuities, and forever is a long time to trust anyone — even Philip II, king of everything.
We can trace annuities to medieval monasteries, which would take grants of land and in return provide a living, either in cash or in kind. As the historian James Tracy points out, when French cities began issuing annuities, they distinguished between rentes viagères — an annuity for life — and rentes héritables, heritable or perpetual annuities.
An annuity for life could be issued on the credit of the city, but a perpetual needed a little more security, and was usually funded by the income from a specific, named city property. In 2003, the Beinecke Library at Yale university acquired a 1648 perpetual bond from a Dutch water authority that continues today to pay interest. But according to the historians Jan de Vries and Ad van der Woude, most early-modern Dutch water boards funded that interest with a dedicated land tax, often as high as the property taxes assessed by the state. The Dutch created deep, liquid capital markets in sovereign debt in part because they were particularly inventive with new forms of taxation.
Even the 1752 consols that George Soros mentions as a model for the debt that Sunak should now issue were funded, too. The various debt issues the consols replaced had been funded by specific taxes on things the English couldn’t do without: coaches, windows and in particular wine and spirits. When the Exchequer issued its consols, it kept all those taxes and rolled them into a dedicated sinking fund, a vehicle dedicated to paying interest on the consols and, eventually, buying some of them back.
The prime minister does not seem inclined, so far, to take Soros’s suggestion. As a financier himself, he might believe that contemporary sovereign debt markets have already found a reasonable compromise between rolling over short-term debt and the complexities of forever. If he’s considering perpetuals, though, he’s going to need to think about how to fund them. Perhaps he could ensure comity among his several united kingdoms with a sinking fund and a dedicated new tax on whisky, whiskey and gin.