The recent showdown between the parliament and the executive in the US began when a faction in the Republican Party tried to stop the implementation of the healthcare law of President Obama. They refused to raise the legislatively determined ceiling on the federal public debt – a ceiling that has to be raised with the growth of the economy.
The showdown had developed into a game of chicken that was played – by both sides – on the dire consequences of missing payments on US bonds (Drezner 2013). This time the Republicans have lost, but some of them promise to renew the fight in a few months, when the newly established ceiling will be reached.
Our recent research highlights a strikingly similar showdown that occurred in 1575 Castile, with catastrophic consequences (Álvarez-Nogal and Chamley 2013).
Sovereign debt in 16th-century Spain
Under Philip II, when the Spanish empire was the dominant superpower, the domestic public debt reached a 'modern level' of about 60% of GDP for the first time in history. The main part of the debt was domestic and was in long-term bonds (mostly redeemable perpetuals) called 'juros', which paid a rate of around 7% (when inflation was about 1%). Given the small administrative capability of the central government – with very few tax collectors – and its uncertain commitment to service the debt, the juros were funded, as later in 18th-century England (Chamley 2012) – the service of the debt was financed by earmarked revenues (e.g. sales taxes, monopolies, customs farms). Recall the protests that prevented the Bush administration from including the Social Security tax (that finances retirement benefits in the US) into the general revenues.
Juros were prized for their stable income, and the juros of the highest grade were administered at the city level. The 18 main cities of the realm were committed to pay a fixed annual payment (encabezamiento) to the Crown. For each city, the service of the local juros – which were bonds of the Crown and tradable through the realm – had first claim on that annual payment. Such a setting contributed to an alignment between the tax enforcers and the bond holders in city governments, enhancing the security of the debt payments and therefore lowering the interest rate.
The amounts of the encabezamiento for each city were set, for a number of years, by simple majority in the Cortes, where each city had two delegates. These contributions were the dominant part of the ordinary revenues that could be used for the service of the juros. These ordinary revenues imposed a de facto ceiling on the service of the juros, which – given the stable interest rate – was tantamount to a ceiling on the domestic debt. When the king wanted to raise the encabezamiento or other taxes managed by the cities, he had to get the approval of the Cortes. The cities could not oppose indefinitely the tax increases, but they could moderate or delay them through protracted negotiations. Philip II was not an absolute monarch, and the decentralisation of the revenues through cities that enjoyed some independence lowered his borrowing costs.
The other part of the debt, which was smaller than the juros, was unfunded and had a different purpose and nature – it was in financial contracts called asientos, signed with Genoese bankers, that combined transfers abroad for the military ventures of the Crown (especially in Flanders) and loans of a maturity between one and three years or more at a rate of about 12%. A large fraction of the asientos was purely domestic however, with both disbursements and repayments in Castile. These loans could absorb transitory shocks, and when shocks turned out to be permanent, they were converted into the funded juros.
The system functioned smoothly as long as there were sufficient ordinary revenues to service the juros. Each of the three financial crises under Philip II occurred when ordinary revenues were about to fall short of servicing the juros (Álvarez-Nogal and Chamley 2013). The 1575 crisis is the clearest case.
Budget wars in Castile, 1575
In the early 1570s, whe