One of the morals from the classic Hans Christian Andersen story, The Emperor’s New Clothes – in which it takes a naively honest child to risk appearing ‘extraordinarily simple of character’ and point out that the Emperor is in fact wearing nothing at all – is that even generally accepted views should be questioned if contradicted by the facts. In that spirit, we re-examine a commonly held view: that the Canadian economy is significantly underperforming relative to both the U.S. and its longer-run production potential. Taking a step back and looking at recent data in a historical context at the least suggests that this argument is not as one-sided as it may at first seem. This was true even before a run of strong economic data in early 2017 for Canada (with Canadian growth in Q1 2017 tracking, by most estimates, a third consecutive quarterly outperformance relative to the U.S.) as, arguably, underperformance of the Canadian economy in the immediate aftermath of the oil price shock never did fully retrace the outperformance relative to other advanced economies accumulated earlier in the recovery from the 2008/09 recession.
Although there remain legitimate reasons to be concerned about the future of the Canadian economy (the lack of business investment growth, even excluding the energy sector, has created concerns about the pace of future productivity growth, for example), even with the shock to the oil & gas sector over the last two years Canada has retained its position as the G7 GDP growth leader relative to pre-recession levels and recent momentum has improved. There has clearly been an increase in labour market slack in Alberta; however, improvement among most oil importing regions has offset to leave the national unemployment rate (and most other measures of labour market ‘slack’) at levels that historically have been consistent with the national economy operating at capacity. Wage growth has been disappointing according to some measures but less so in others and it remains the case growth in ‘real’ worker incomes in Canada have outpaced the U.S., where the economy is more widely accepted to be on a firmer footing, over the full recession/recovery period. Consumer price inflation trends in Canada, while ostensibly not concerning, are also arguably more similar to the U.S. than is commonly acknowledged and there have clearly been more worrying trends emerging in asset (particularly housing) prices in Canada.
Canada continues to lead the G7 countries in GDP growth from pre-2008/09 recession levels…
Canadian GDP growth slowed dramatically after oil prices started to decline sharply towards the end of 2014; however, that underperformance has not been enough to reverse outperformance in the period leading up to the oil price shock, including relative to the U.S. economy. 2015 marked just the second year since the recession that Canadian growth was weaker than in the U.S. (0.9% in Canada versus 2.6% in the U.S.) with the Canadian economy weighed down by the plummet in oil prices. This factor contributed to the Canadian economy underperforming the U.S. in 2016 again, on an average annual basis, but relatively modestly (1.4% versus 1.6%). The year started out slowly, with growth in the second quarter in particular weighed down by the Fort McMurray wildfires, but finished strongly. It remains the case that Canada has not yet relinquished its spot as the G7 growth leader when measured relative to activity just prior to the beginning the 2008/09 recession (Chart 1).
…and recent momentum has improved
Of course if weak momentum over 2015 and early 2016 were to persist, it would just be a matter of time before Canada’s longer-run outperformance reverses; however, evidence is mounting that the economy is emerging from its slow patch. As shown in Chart 2, weakness in the Canadian economy over the last two years has been largely concentrated in the oil & gas sector. There was also some earlier slowing in nonenergy commodity sectors but non-commodity sectors (the other ~80% of the economy) have continued to grow, on balance, at a steady pace. We continue to expect the level of investment activity in the oil & gas sector to remain subdued relative to earlier 2014 levels because oil prices remain below where they were. Nonetheless, weakness has clearly eased and some preliminary data (including a tick higher in investment intentions and recent increases in oil & gas drilling activity) suggests that the sector will shift back to a modestly positive contributor to growth in the near-term. Outperformance over the second half of 2016 in Canada left the year-over-year rate of growth in GDP in the fourth quarter already equal to that of the U.S. at 1.9% and early data releases for 2017 have been constructive.
The regional composition has changed but the Canadian unemployment rate is still low…
Labour markets have clearly deteriorated in the oil-producing provinces, particularly in Alberta where the unemployment rate has increased sharply from low levels; however, improvement elsewhere – unemployment rates have declined in almost all of the net oil-importing provinces (Manitoba was the one exception although that province still had the second-lowest unemployment rate in the country in February, behind only B.C.) – suggests that the net impact has been more of a geographic shift in the location of labour market strength/weakness than a net impact on the ‘national’ labour market. The Canadian unemployment rate was 6.6% in February, slightly below its level at the end of 2014. The same pattern is importantly true using broader measures of labour market slack that account for ‘hidden unemployment’ in the form of discouraged workers or involuntary part-time employees (Chart 3).
The current national unemployment rate level is also, notably, still well-below its 7% 10-year pre-recession average, a period in which the Bank of Canada’s historical estimates of the ‘output gap’ – which estimates the extent to which the economy is operating below (negative) or above (positive) its longer-run sustainable production capacity – were positive, on average. In other words, current levels of labour market ‘slack’ nationally are below levels that have historically been considered consistent with ‘full-employment’ at the national level.
…even when compared to the U.S.
Adjusting Canadian unemployment rate data to U.S. definitions – differences in the way the rates are calculated, in large part related to the definition of who is considered to be "in the labour force," typically add about a percentage point to the ‘official’ Canadian unemployment rate relative to the U.S. – the Canadian rate was 5.4% (on a seasonally adjusted basis) in February. That is still above the U.S. rate, which has declined to 4.7%, but that is not actually historically unusual. Prior to the 2008/09 recession and recovery period – over which Canada’s labour market broadly outperformed the U.S. – the Canadian rate had not been below its U.S. equivalent since 1982 with the average gap over the pre-recession period (~2 percentage points) more than twice the current difference. Even in the boom-times in the energy sector ahead of 2008/09 the Canadian unemployment rate never dipped below the U.S. rate (Chart 4).
What about wage growth?
Tighter labour markets – in other words, rising demand relative to supply of labour, as evidenced by a smaller unemployment rate – should, so the theory goes, push wages higher and, admittedly, some indicators of wage growth have failed to show any significant upward pressure. One closely watched measure, hourly wages of permanent employees derived from Statistics Canada’s monthly Labour Force Survey has been running just 1% above year-ago levels in January and February this year, without adjusting for inflation. The wage data, like most other Canadian labour force data, is also extremely volatile, however, and other measures have not been so disappointing. Hourly wage growth in the alternative labour market survey, the Survey of Payrolls, Employment, and Hours, has still been trending at slightly above a 2% rate, despite year-overyear declines in Alberta. Overall per-hour labour compensation costs rose 2.1% year-over-year in Q4/16 according to the quarterly productivity accounts data. Combining the employment and wage data, year-over-year growth in total aggregate real (inflation-adjusted) wage earnings was 2.1% in Q4 2016, slightly below the 2.5% U.S. increase but almost double the 1.2% rate a year earlier. Over the full economic cycle Canadian wage growth continues to run above comparable U.S. measures. Although the gap has been narrowing (in part because of weaker Canadian growth recently but also reflecting U.S. labour markets ‘catching up’ to those in Canada where the initial recession recovery was faster), total real wages and salaries in Canada are 17.2% above their Q4 2007 level (the last quarter before the U.S. entered recession) compared to 13.7% in the U.S. (Chart 5).
Consumer price growth still modest, but similar to the U.S….
Weak inflation trends are also often pointed to as evidence that the economy continues to operate well-below its potential (a symptom of an economy with insufficient demand, theoretically, should be weak price growth); however, here too the data is perhaps not as clearly one-sided as is often thought. Energy prices continue to be buffeted by oil price volatility and food price growth has been historically weak, reflecting a combination of earlier declines in agricultural commodity prices as well as industry-specific competitiveness factors. Excluding those, often volatile components, Canadian CPI rose 2.0% from a year ago in February. That was down just slightly from an almost 10-year high 2.2% in January. The latest reading was above the1.8% 10-year pre-recession average and was just slightly below the 2.2% February year-over-year gain in the equivalent, and closely-watched, U.S. measure (Chart 6).
….and clearly evidence of upward pressure on asset prices in some regions
Price pressure resulting from a stronger economy also does not necessarily have to manifest exclusively through prices for current consumption as measured by the CPI. Clearly some upward pressure on asset prices has emerged, particularly in some of the hotter residential real-estate markets. The normal counterpoint is that housing market heat has been largely contained to regions in and around Vancouver and Toronto, so housing market ‘tightness’ is not really a ‘national’ phenomenon requiring a ‘national’ policy response. The Canadian Real Estate Association (CREA) nonetheless noted in the release of their February statistics that almost 60% of local markets they track nationally were in ‘sellers -market’ territory. By our count, that share was about 40% a year ago and just a quarter the year before that. Most of those markets are admittedly still located in Ontario and B.C., but Ontario and B.C. together account for about half the Canadian economy. Over the last three months (ending in February), 75% of markets in Ontario and almost 60% of markets in B.C. saw average house price increases of 10% or more on a year-over-year basis. Those shares are both up from about 25% a year ago and closer to 10% in the same period in 2015. The question is at what point do these ‘local’ trends impact enough markets that they generate ‘national’ concerns for policymakers. Clearly, there has already been a significant impact on national-level measures of house prices. Canadian house prices were up 12.3% in 2016 according to CREA’s benchmark house price index. In February, year-over-year growth in the measure hit a whopping 16%.
So how much ‘slack’ is there in the Canadian economy – and what should be done about it?
It is not possible to directly observe economic slack. Rather the extent to which the economy is utilizing its underlying productive resources needs to be inferred based on a mix of economic data, comparisons to historical performance over multiple business cycles, and economic theory. While there has clearly been an industrial and geographical reorientation of economic strength in recent years away from oil-producing regions and there have been disappointments – for example, weakness in business investment even excluding the oil & gas sector – we would argue that the balance of economic data at the least suggests that the argument that the Canadian economy as a whole is significantly underperforming both the United States and its own long-run production potential is not as one-sided as it may at first seem.
To be sure, saying that the economy nationally is, perhaps, closer to its longer-run production potential than is commonly thought does not mean that regional divergence in growth prospects should be ignored. It obviously doesn’t make sense, for example, to argue that households in Alberta should be ‘okay’ with a 9% unemployment rate because the rate is now substantially lower in Ontario and B.C. The appropriate policy response is also, however, different in the case of a ‘national’ versus a ‘regional’ economic shock. Monetary policy stimulus (ie. low interest rates) has likely provided some benefit to activity in regions where economic activity is more subdued but the persistence of those rates at historically low levels also runs the risk of contributing to overheating in regions (particularly Ontario and B.C.) where activity has been significantly stronger and where, in some markets, worrying trends in house prices have clearly emerged. The closer the economy nationally is judged to be to its long-run production capacity the less appropriate low interest rates become as a policy response with targeted government fiscal policy measures (for example through re-training programs, investments in new industries/technologies to reduce reliance on the oil & gas sector, etc) better suited to deal with regional divergence in performance.