Not Like a Household

Public debt is explained with a household analogy so consistently that the analogy has become the argument. It is wrong in one specific and important respect, and understanding exactly where it fails is more useful than either accepting or rejecting it wholesale.

A government that borrows in a currency it issues cannot be forced into nominal default. It can always meet an obligation denominated in money it creates. A household cannot. That removes one failure mode (the forced default) and it removes nothing else.

What binds instead is real. A government cannot command goods, labour and materials that do not exist. If it tries to spend beyond the economy's capacity, the result is inflation and a falling exchange rate rather than a missed payment. The constraint is genuine; it is simply not the household's constraint, and it shows up as a different symptom that arrives on a different timetable.

The arithmetic that actually governs whether a debt is sustainable is the relationship between the interest rate on it and the growth rate of the economy. Debt ratios are debt over output, and both terms move. If the numerator grows at the interest rate and the denominator grows faster, the ratio falls even while borrowing continues. Rich economies spent most of the decade after 2010 in exactly that condition, which is why debt levels that sounded alarming were considerably less so than the headlines suggested, and why the condition reversing matters more than any particular level.

There is also rollover risk, which the household analogy obscures completely. A stock of debt is not a single obligation sitting still. Slices mature constantly and must be refinanced, so the exposure is to continued willingness to lend rather than to the level itself. This is why average maturity matters as much as quantity: a long profile buys time to adjust if sentiment turns, a short one means a change in mood arrives immediately as a financing problem.

The UK gilt episode of 2022 showed the limit without any default. A fiscal announcement met a market, yields moved sharply, pension funds using leveraged hedging strategies were forced to sell into the fall, the central bank intervened to stop the spiral, and the policy was withdrawn within weeks. Nothing about solvency changed mechanically. What bound was the price at which the market would absorb issuance, which is a real constraint that no amount of currency issuance dissolves.

Finally, the cautionary tale about how this material gets used. Reinhart and Rogoff's finding of a ninety per cent debt-to-GDP threshold beyond which growth falls sharply was cited in austerity arguments across several continents. A graduate student attempting to replicate it found an omitted range of spreadsheet cells, together with a weighting choice that gave one country-year the same influence as nearly two decades of another country's experience. Correcting both left a weak relationship with no established direction of causation, and the causation is the whole question, since slow growth raises debt ratios at least as readily as the reverse.

The number was load-bearing in real policy for years before anybody checked it, which is the more durable lesson.