Kevin Milligan is Associate Professor of Economics at the University of British Columbia
Thursday's budget will let loose a torrent of numbers. While economists are well known to disagree on many things, we are united in our love for number crunching. Frances Woolley in her recent post provided a few pointers on how to steer around some typical budget number tricks.
In my post today, I build on her advice with a few more tips on how economists read a budget.
The first question to ask of any budget announcement is whether the dollars are recurring or one-time only. If we change a tax that brings in $1-billion a year, the budget changes not just this year but in future years as well. This recurring change is clearly different from a one-time purchase or sale of an asset, but we always struggle to compare recurring to one-time changes. In the United States, the Congressional Budget Office often uses a ten-year budget window in evaluating tax and spending initiatives. We don't have a similar convention in Canada, so it seems that anything goes. Politicians and commentators often choose the time frame that suits their current argument. Confusion results. A good economist keeps her eye open to these tricks and tries to ensure we compare numbers on similar time-frames.
Next up is properly adjusting future dollars to account for inflation and our ability to pay. Dollars spent in the future are different than dollars spent now. Imagine that inflation averages 2 per cent a year, and inflation-adjusted economic growth is 1.5 per cent a year on top of that. In just 20 years, prices will increase by 50 per cent and the size of our economy – and our ability to pay for programs priced in nominal dollars – will double. The pitfalls of across-time comparisons move to the forefront for budget items like Old Age Security, when we are thinking about the affordability of our system in the 2030s. The $108-billion cost of Old Age Security in 2030 is not the same as $108-billion today.
Third, we must always beware of 'inverted hockey stick' spending projections. A government looking to tame deficits will often make unlikely projections for a spending line-item that has increased year after year. Suddenly, after years of growth (the upward-pointing shaft of the inverted hockey stick), the spending line is projected to switch to flat growth (the flat blade of the hockey stick). This problem is pervasive for provincial health spending, but also shows up in the federal budget. Current projections for the federal government show an intention to shrink the average 5 per cent a year spending growth pace we've seen since 2006 down closer to zero. Commitments to contain growth have occasionally been realized; take as examples the Alberta budgets of the early 1990s, the federal budgets of the mid-1990s, or the Ontario budgets of the later 1990s. Experience suggests that successfully enforcing budget cuts requires a substantial and credible plan rather than vague platitudes about cutting waste, hopes for attrition, or saving paper clips.
As a final note, it is always useful when crunching the numbers to keep in mind what the Government of Canada actually does with our tax dollars. Transfers to individuals for insurance programs (such as Employment Insurance and Old Age Security) are 25 per cent of spending. Transfers to provinces and territories (health and other transfers) are another 20 per cent. Interest takes a further 11 per cent. The best way to think of the Government of Canada is a big national insurance company with a side business as a tax collector for the provinces. (This is only slightly different from the US Government, which has been called by Ezra Klein an insurance company with a standing army.) Everything else the Government of Canada does – from fisheries management to culture to the military – takes the remaining 44 per cent. Making any change to the trajectory of total spending when insurance and inter-government transfers are both projected to grow rapidly requires very large changes to that residual 44 per cent.
Follow Economy Lab on twitter