The only thing worse than a legal borrowing limit is no legal borrowing limit

The end of another debt-ceiling debacle, with the United States just days away from defaulting on its debt before an agreement was reached authorizing new borrowing, has elicited the usual relieved sighs of “let’s never do that again.” Alas they surely will do that again – the current deal expires in 2025 – not just once but over and over.

That’s only partly the fault of the debt ceiling. The total amount of debt the U.S. government can issue has been restricted by law since 1917. For most of that time no crisis ensued. Only with the radicalization of the Republican Party in recent years has the procedural formality of lifting the ceiling become the semi-annual occasion for playing Russian roulette with the country’s credit we know it as today.

Still, leave a loaded gun around long enough somebody’s bound to pick it up. The potential for brinksmanship in such a provision – an absolute dollar limit on borrowing, without a corresponding limit on deficits – was always present. On the other hand, before the era of mutually assured default the ceiling wasn’t really much of a ceiling, was it?

All of which might suggest that any such legislated limit on borrowing is, by its nature, a bad idea. Isn’t this the sort of thing best left to the discretion of those in government? Why tie their hands? We elect them. If we don’t like their decisions, about debt or anything else, we can always replace them.

But governments’ discretion is constrained in all sorts of other ways. Every law binds the executive, for starters: that is the point. Even legislatures cannot just pass any law they like. They are obliged to work within certain limits, most of them self-imposed: the Constitution, international treaties, federal-provincial agreements.

So the question becomes: should deficits and debt be included on the list of things that are subject to legislated rules? And if so, what sort of rules?

The case for some kind of legal limit on public borrowing makes much the same appeal to democratic principle that might have been marshalled against it. But it applies a broader definition of democracy than simply “whatever the legislature happens to vote at any given moment.”

First, a legislature that takes on debt is deciding on behalf of not just its own but future generations, who will have to bear much of the fiscal consequences. That might be defensible where the money is borrowed for capital investments, of a kind that benefit those same future generations. Not so much for borrowing for current consumption.

Second, a legislature might chronically spend more than it taxes because of a phenomenon known as the public choice dilemma: while the costs of a given spending program are spread across the general population, the benefits are often concentrated on a narrow minority. But that minority might hold disproportionate sway, precisely by virtue of its narrowness. Its members have, individually, a great deal to gain from the program, and thus every incentive to lobby for it – whereas members of the general population do not stand to gain or lose enough individually to care much what happens either way.

Third: provinces that take on debt are making decisions that may have consequences for other provinces, whether via interest rates, or the exchange rate, or in perceptions of their creditworthiness. Yet these other provinces have no say in their decisions.

For all these reasons and more, there is a case to be made for some sort of legislated limits on the amount a government can borrow. It needn’t be as crude as a debt ceiling. It might be expressed relative to GDP, rather than in absolute terms. There could be exceptions for war, disaster or depression. It might apply over the business cycle, rather than a single year. It might even distinguish borrowing for investment from borrowing for consumption. But something.

I know what you’re thinking. With so many caveats and conditions, what’s the point? How could it be enforced, at any rate? What’s to stop a legislature, having voted for a borrowing limit one year, from voting to repeal it the next? Short of a constitutional amendment, the answer is: nothing. Rules are no substitute for will. If a legislature fundamentally doesn’t want to restrain itself, it will find a way around any rule.

But rules can buttress will. A borrowing rule doesn’t have to work perfectly to be of some benefit. Even the necessity of repealing it takes time, and political capital, and as such can serve as a deterrent. That’s particularly relevant to the provinces, which are under enormous fiscal pressure due to the exploding costs of health care. The risks that one or more of them might buckle under the strain in coming decades are real.

An ill-conceived borrowing rule nearly brought the U.S. to default. But Canada may find in time that the absence of one can do the same.