WILL THERE AGAIN BE A REVENUE WINDFALL FOR THE FEDERAL BUDGET?

In recent years federal revenues have persistently come in higher than anticipated, opening up further choices in budget planning. More often than not, the decision has been to ramp up spending. Most of the provinces that have tabled budgets this year have similarly found higher-than-expected revenues and have opted for the same choice. This note asks the question whether the new federal Conservative Government will also strike the revenue mother lode. It looks like a yes.

The starting point for the analysis is last fall’s Economic & Fiscal Update from the previous Liberal Government. The fiscal projections in that document were based upon the average private sector forecasts from September 2005. TD Economics has surveyed the most recent publicly available forecasts. The Department of Finance will do a more comprehensive survey prior to finalizing the budget – which is expected late April or early May – but the results are not likely to differ greatly from what TD Economics has found. We present our survey results in Table 1. We also show the TD Economics projections, which tend to be less optimistic than other forecasts. For this analysis we have only examined Gross Domestic Product (GDP) as other fiscal drivers, such as interest rates, have not changed appreciably since the September 2005 forecasts.

Nominal GDP serves as a very rough proxy for the overall tax base. It can be divided into real and inflation components. Real GDP projections have not changed much since September 2005, being just marginally stronger for 2005 and 2006, but significantly weaker for 2007 as sentiment grows about a slowdown originating from the United States. The bigger change is on the inflation side. The GDP deflator rose much faster than projected in 2005 and is expected to rise at a faster clip again in 2006 before settling back to a similar rate of increase as expected in September for 2007. The main driver for the faster rise in the GDP deflator is higher-than-anticipated commodity prices. Everything else equal, this would raise federal revenues. In the first instance the effect would be through higher corporate income taxes, but as dividends and possibly wages would ultimately rise from the stronger corporate profits, personal income taxes could be expected to benefit as well.

Relative to the September 2005 survey used for the Fall Update, the level of nominal GDP in the current round of private sector forecasts is up $11 billion for 2005, $21 billion for 2006 and $19 billion for 2007, as shown in Table 2. The differences are somewhat lower for the TD Economics projections. Total federal government revenues are about 15 per cent of nominal GDP. Therefore, a starting hypothesis is that federal government revenues should be higher than in the Fall Update (which was used as the point of departure for the Conservative Party’s election platform fiscal plan) by 15 per cent of the change in forecasts of nominal GDP. This would suggest that federal government revenues will be up by $1.65 billion for 2005, $3.15 billion for 2006 and $2.85 billion for 2007. The fiscal impacts would be similar – ie revenues up something like $3 billion for fiscal 2006-07 and a bit less than that increment for 2007-08.

Actual revenue collections are also above forecast

Economists in the Department of Finance are no doubt testing that starting hypothesis at this moment. Most importantly, they will be examining whether the higher nominal GDP is being reflected in the actual revenue being collected. As much of the growth rate difference in nominal GDP between current forecasts and the September 2005 survey was for 2005, they will in particular be examining what has happened to collections since the Fall Update.

In Table 3 we show recent collections data for personal and corporate income taxes. We only examine these two sources because they are large and tend to be the wildcards for budget forecasts in that they typically have the largest forecast errors. As is so often the case, collections from these taxes have been very volatile, making it difficult to ascertain trends. In the Fall Update, personal income tax revenues were expected to rise 1.4 per cent from fiscal 2004-05 to 2005-06. The growth rate was so low in part because of the tax cuts introduced in the Update. Actual revenue collections surged in December and January despite the tax cut (as revenues are now recorded on an accrual basis, the tax cut for the 2005 taxation year is reflected in the December figure) bringing the total for April through January to 4.1 per cent above the figure for the corresponding 9 months of the previous year. If this pace were to hold, then 2005-06 fiscal year personal income tax collections would be $2.4 billion above the level projected in the Fall Update. But will the recent pace hold?

Personal income tax collections in December 2005 were 8.3 per cent above the figure for December 2004 and in January 2006 11.5 per cent more was collected than in January 2005. That means that despite the tax cuts, personal income tax revenues have recently grown at almost twice the pace of personal incomes. Due to the progressivity of the tax system, personal income tax collections should grow 15 to 20 per cent faster than the income tax base (an “elasticity” of 1.15 to 1.20). Some additional revenue growth is coming from solid capital gains that are not reflected in Statistics Canada’s measure of personal income (because they don’t reflect real economic activity). But still, the strength of the recent numbers is likely a puzzle to Finance and there must be some concern that the recent momentum will subside in the coming months.

Corporate income tax collections were projected in the Fall Update to rise 12 per cent in fiscal 2005-06. Actual revenue collections have been running a bit hotter than that, with the year-to-date growth figure through January at 13.7 per cent. If that pace were to hold for the fiscal year, the growth increment over the Fall Update would yield an additional $500 million.

What to do with the money?

The analysis of actual revenue collections to date therefore suggests that the simple approach of multiplying the change in nominal GDP forecasts by the average effective tax rate understates Ottawa’s improving revenue fortunes. However, as much of the gain has been realized only recently on the personal income tax side, there is some question of sustainability. As the budget will not likely be delivered for another 4-5 weeks, Finance will obtain more data on actual revenue collections before they have to make a final call on the forecast. In particular, they will get the full array of collections for February and will know the cash balance as of March 31, 2006, the final day of the fiscal year. They will be able to see the inflows of revenues over the critical first two weeks of March, where in some years as much as one-third of the entire year’s collections of corporate income tax revenues is received (because a lot of corporations have December year-ends and their “settlement period” payments are received in early March to reflect any tax liabilities exceeding the installment payments they made in 2005).

The analyses of the changes in nominal GDP forecasts and recent revenue collections both suggest that Ottawa is yet again receiving a revenue windfall. Prudence may dictate that revenues not be marked up by the full extent of the $3 billion a year the simple approach here suggests. But it would be difficult to credibly present a budget without showing a significant improvement in revenue fortunes relative to both the Fall Update of the Liberals and the Conservative’s own election platform fiscal plan. The question then becomes, what to do with this extra revenue?

The first option is to boost prudence itself. The Liberal Government’s forecasts featured a $1 billion “economic prudence” buffer on top of the contingency reserve. With little fanfare, the economic prudence factor was dropped in the Conservative’s election platform fiscal plan. It could be restored, or given the confusion over the distinction between economic prudence and the contingency reserve, forecast “protection” could be augmented through a new approach.

Beyond prudence, we need to turn to the budget identity. The budget balance is the difference between revenues and spending. So the Government could use the extra revenues to run higher surpluses and hence pay back more debt. That would seem to be a low probability outcome. Alternatively, they could put the extra money into higher spending, thus following the lead of their predecessors and most provinces. There would be no shortage of candidates. The “easiest” spending increase would be to void a spending cut. In the Conservative’s election platform fiscal plan, they incorporated an internal spending savings target of $22.5 billion cumulatively over the 5 fiscal years 2006-07 through 2010-11. TD Economics has estimated that the annual pattern works out to something like $2 billion for 2006-07 and $3 billion in 2007-08. By coincidence, that’s about the difference calculated above for the revenue impact of changed economic projections. This, however, would be an unfortunate choice. As the Conservatives have introduced new spending priorities, by definition the priority attached to some previous spending has slipped down a couple of rungs. Some of that existing spending should be cut to finance the new priorities. The Liberal Opposition, as well as the provinces, would likely argue that the extra revenues should be used to keep the federal-provincial daycare agreements in tact.

The final option of course is to cut taxes above and beyond the promises set out in the Conservative’s election platform. That would be taking the road less traveled. A surprise, yes, but a very pleasant one.

Don Drummond, SVP & Chief Economist
416-982-2556