In its November Fall Economic Statement, the federal government presented a long-term projection that shows its debt ratio — that is, federal debt divided by GDP — declining smoothly over the next 30 years. But this outcome follows from overly optimistic assumptions about interest rates and program spending, and a decision to ignore the impact of recessions, which are certain to happen in any 30-year period. Taking such a rosy approach to debt sustainability allows the government to avoid making the hard choices on spending and taxes that no government likes.
Ottawa’s analysis assumes the effective interest rate on federal debt remains below the growth rate of the economy from now all the way to 2055-56. This sunny projection means debt interest falls as a share of GDP with no need for action to improve the government’s fiscal situation. (The parallel with personal borrowing is this: If your income is rising at seven per cent a year, you can take out a loan at five per cent, borrow to pay the interest, and watch your debt decline as a share of your income.)
The historical record provides no support for this optimistic assumption that the interest rate will be less than the growth rate. Quite the opposite: looking back over 35 years the interest rate has exceeded the growth rate by 0.8 percentage points on average, and by 0.4 percentage points over the past 45 years.
It’s true that extending the averaging period to 55 years does result in a negative gap. But only because this longer horizon captures the unusually high inflation of the mid-1960s to the early 1980s, which raised nominal GDP growth on a one-for-one basis but only affected interest rates with a lag. For effective interest rates to remain below GDP growth over the next 33 years would require a similar unexpected acceleration in inflation. But this is ruled out in the federal projection, which shows GDP inflation quickly returning to its target value of two per cent and remaining there.
A second controversial assumption is that the government will allow program spending to fall as a share of GDP while revenues rise broadly in line with GDP. The resulting increase in the government’s operating balance — its revenues minus its program spending — also helps drive the debt ratio down. The slower growth in program spending occurs because increases in many transfer programs are limited by legislated escalators and the growth in target populations. Judging from experience, however, it is very unlikely that any party in power will allow the real value of these programs to decline substantially. Three decades of fiscal austerity is not in the cards.
Our approach to forecasting the debt ratio is to assume instead that program spending stays constant as a share of GDP over the projection period. We then assess debt sustainability by holding the operating balance at 1.2 per cent of GDP, its value in 2028-29, the last year of the government’s medium-term forecast and the starting point for our sustainability analysis.
The third and most important limitation of the federal debt sustainability projections is that they don’t include a provision for the discretionary stimulus spending that inevitably accompanies economic downturns. Building in recessions and the typical government response provides a better idea of the likely fiscal impact of recessions. Over the past 60 years there have been five significant downturns, all of which have prompted governments to respond with temporary stimulus averaging 1.1 per cent of GDP to support incomes. In contrast to automatic stabilizers like Employment Insurance, which increase debt during downturns and reduce it during upswings, these discretionary measures ratchet up a permanent increase in debt.
In our own alternative debt projections, we assume program spending grows at the same rate as GDP and that on average over the projection interval interest rates are the same as GDP growth. In our own simulations, we capture the effects of economic downturns on the debt ratio by playing out 1,000 randomly generated scenarios in which future recessions and government responses to them are similar to those over the past 60 years.
This more realistic analysis reveals a more turbulent path ahead, with quite a different denouement than the government’s smooth slide to a low debt-to-GDP ratio. In our simulations, there is a 50-50 chance that late in the projection period the debt ratio will in fact exceed its 2028-29 value. Using the IMF’s fiscal yardstick, that would be classified as “unsustainable.”
Even more ominously, there is a non-negligible risk the debt ratio will start its upward trend earlier and will end up well above its starting value. In fact, there is a one-in-five chance the debt ratio will be above its 2028-29 value by 2032-33 and will exceed its starting level by about 10 percentage points by the end of the projection period.
For Canada’s debt to be considered highly sustainable in the IMF classification scheme, the chance of systematically exceeding the 2028-29 debt ratio must be less than 20 per cent. To achieve this goal, the federal government would have to permanently increase its operating balance by $13 billion in 2028-29, which would mean making difficult policy choices it has avoided until now. It could do that by raising the GST by almost one percentage point or reducing program spending by more than two per cent of its level in 2028-29. But without such measures the debt is not sustainable — which you’d think would trouble a government whose watchword is “sustainable.”
John Lester is a fellow-in-residence at the C.D. Howe Institute, where Alexandre Laurin is director of research.
