The Liberals’ arithmetic problem

You may have missed the release of the Department of Finance’s latest Update of Long-Term Economic and Fiscal Projections. You were certainly meant to miss it: even though it was produced back in October, the actual release of the update was delayed until the Friday just before the Christmas holidays. The Liberals did the same thing with the previous exercise, although at least this longterm update was accompanied with a press release. One gets the impression that the Liberals would rather not think about — much less talk about — the outlook for federal government finances.

And with reason. The government’s Mandate Letter Tracker says that the Liberals’ commitment to balance the budget in 2019- 20 and continue to reduce the federal debt- to- GDP ratio is “Underway with challenges.” Many have already remarked that this is a very generous assessment: the debt- GDP ratio increased in 2015-16 and again in 2016-17, and the government has yet to produce a budget projection in which the budget is balanced before 2040. We’re at the point where it’s fair to ask whether or not the Liberals even want to meet their fiscal commitments.

It’s hard to keep on top of this file: there are a lot of numbers to keep track of, and a new set of projections is published every six months. But there is one number that has stayed the same since at least 2014, and it’s probably the most important one to keep in mind: f ederal government revenues fluctuate in a narrow range around 14.5 per cent of GDP. This ratio — the lowest it’s been since before the Second World War — is the enduring fiscal legacy of the Conservative government.

If there are no changes in the tax structure — and the Liberals have so far not made or promised significant changes — then government revenues will grow roughly in line with the economy, which means that the ratio of revenues to GDP will remain roughly constant. In both the 2014 (the last long-term projection under the Conservatives) and the most recent exercise under the Liberals, projected federal revenues were assumed to be 14.5 per cent of projected GDP throughout the forecast horizon. Put another way, government revenue projections are a combination of economics and politics: economics provides projections for GDP, and the share of GDP that goes to the government is a political decision.

This logic works on the other side as well. For a given set of publicly- provided services — and again, this is a political decision — we’d expect the level of spending required to sustain them to increase with GDP as well. ( This assumes that costs increase with the population and with inflation, that productivity growth is hard to wrest from the public sector, and that public servants will demand the same increases in real incomes that other workers receive.)

The arithmetic here is simple and unforgiving: if you accept — as the Liberals have — the Conservatives’ choices for revenues and the budget balance, you are also obliged to accept the Conservatives’ choices for the total level of public services provided. ( New programs would have to be financed by cutting existing ones.) The original Liberal plan respected that arithmetic: deficits would finance a temporary burst of infrastructure spending, so that the budget would go back to balance when the extra spending stopped.

But that’s not what happened. The Liberals’ infrastructure program has yet to get off the ground, but Direct Program Expenditures ( DPE) — the budget item that involves paying salaries and buying things — has gone from 5.8 per cent of GDP in 2014-15, the last full year of the Conservative government, to a projected 6.5 per cent in the current 201718 fiscal year. ( Transfers to individuals and other levels of government are rulesbased and generally track GDP automatically.)

Since revenues and transfer payments track GDP, the only way to reduce the deficit is to reduce DPE as a share of GDP. The reason the October fiscal update predicted lower deficits than projected in last March’s budget was that it made more aggressive assumptions about how the Liberal government would control growth in direct spending: it projects that DPE will be reduced back to down to 5.7 per cent of GDP in 2022-23.

This seems unlikely. In a blog post, Randall Bartlett of the Institute of Fiscal Studies and Democracy at the University of Ottawa says at “the IFSD, we don’t believe that the federal government’s DPE forecast is realistic.”

Nor do I. To be sure, it’s a feasible scenario: the Conservatives reduced the DPE share of GDP by a similar amount during the course of their last mandate. But that’s just it: the last Conservative mandate was one of continuous cost- cutting that provoked resentment among the sort of people the Liberals view as natural allies. ( The Public Service Alliance of Canada saw fit to run ads attacking the Conservatives in the run-up to the last election.) It’s doubtful the Liberals have the political will to sustain five years of grinding austerity to reach a goal that they manifestly believe is unimportant, in order to respect a commitment they probably regret making. If I were them, I wouldn’t want to talk or think too much about budget projections, either.