Debt-to-GDP: Why 90% Isn't a Magic Line

Debt-to-GDP is the single most widely quoted number in public finance, including on this site's own rankings page. It's a genuinely useful measure: dividing a government's total debt by the size of its economy gives a rough sense of how large that debt is relative to the country's capacity to eventually service or grow out of it, and it lets you compare a small economy's debt against a giant one on a common scale. But the ratio has also been treated, at times, as if it contained a hidden magic threshold — a specific number past which growth collapses. That claim has a real, well-documented history, and it turned out to be far shakier than it was made to sound.

The Reinhart-Rogoff claim

In 2010, economists Carmen Reinhart and Kenneth Rogoff published influential research finding that average economic growth rates tended to be notably lower in countries and years where gross government debt exceeded roughly 90% of GDP, compared to periods below that threshold. The finding landed at a politically potent moment — in the middle of the post-financial-crisis austerity debates in the US and Europe — and the "90% threshold" was widely cited by policymakers and commentators as evidence that high debt causes growth to slow sharply once that line is crossed.

The spreadsheet error

In 2013, researchers at the University of Massachusetts Amherst — Thomas Herndon, Michael Ash, and Robert Pollin — obtained the original underlying spreadsheet and found significant problems with it, including a coding error that accidentally excluded several countries' data from a key calculation, along with debatable choices about weighting and which country-years to include. When corrected, the sharp growth cliff at 90% mostly disappeared; the relationship between higher debt and lower growth still existed in the data, but as a much gentler, more gradual association rather than a hard threshold effect. The episode became a well-known case study in economics and data-journalism circles about the risks of drawing strong policy conclusions from a single influential dataset without independent replication, and about how an appealingly precise number can outrun the evidence behind it once it enters political discourse.

Why thresholds are context-dependent

Even setting the spreadsheet episode aside, the broader idea of a single universal debt threshold runs into an obvious problem: countries carry very different debt loads without obvious distress, and the level that looks fine in one country looks alarming in another. Japan sits around 230% of GDP by the gross measure this site tracks and has not experienced a debt crisis; various emerging-market economies have come under serious market pressure at debt levels closer to 60% of GDP. The gap isn't a mystery — it comes down to factors the raw ratio doesn't capture at all: who holds the debt (domestic savers versus foreign creditors), what currency it's denominated in (a country's own, freely floating currency versus a foreign currency it can't create), the maturity structure (long-dated versus needing frequent refinancing), the strength and credibility of a country's institutions and central bank, and whether the debt was built up gradually with a track record of being serviced, or accumulated suddenly amid already-fraying confidence.

The direction of causation is also disputed

Even setting aside the data-quality problems with the original 2010 study, economists have long debated a more fundamental question buried in this kind of research: does high debt cause slow growth, or does slow growth cause debt to rise, as a struggling economy generates less tax revenue and spends more on stabilizers like unemployment support? Most researchers now believe the true relationship runs in both directions at once, which makes it very difficult to extract a single clean threshold from historical data in the first place — a country crossing 90% debt-to-GDP during a severe recession tells you something quite different from a country reaching the same ratio through years of expansionary borrowing during otherwise normal growth.

What actually matters more than the ratio

Economists studying debt sustainability tend to focus less on any single debt-to-GDP snapshot and more on the relationship between a government's interest rate and its economy's growth rate — often shorthanded as "r versus g." When the growth rate of the economy exceeds the interest rate the government pays on its debt, a given debt-to-GDP ratio tends to stabilize or shrink on its own over time, even without aggressive austerity, because GDP is growing faster than the debt stock. When the interest rate exceeds growth, the ratio tends to drift upward even with a balanced primary budget, all else equal. This single comparison explains far more about a country's actual trajectory than whether its current ratio sits above or below any round number like 60%, 90%, or 100%.

Institutions and track record

A less quantifiable but widely cited factor is the strength and credibility of a country's institutions — an independent central bank, a functioning tax system, a track record of servicing debt through past downturns, and political systems capable of adjusting course before a crisis rather than only after one. Two countries with an identical debt-to-GDP ratio and similar currency and maturity profiles can still be priced very differently by bond markets if one has decades of consistent, boring debt management behind it and the other has a recent history of restructuring, high inflation, or political instability. Markets are, in effect, pricing in a judgment about future behavior, not just the current stock of debt — which is part of why the same ratio can command very different borrowing costs for different countries.

None of this means debt-to-GDP is a useless number — it remains the standard, most comparable way to size up a government's debt load across countries and over time, which is why it anchors this site's rankings. It just isn't a tripwire. A country's real vulnerability depends on the fuller picture: who holds the debt, in what currency, at what rate, and against what growth trajectory — the same set of questions that explain why Japan and various emerging markets can sit at such different points on the same ratio scale with such different outcomes.