Applied Research Bulletin - Volume 3, Number 1 (Winter-Spring 1997) - April 1997

The Canada-U.S. Unemployment Rate Gap

Previous

Contents

Next

The divergence of the Canadian and American unemployment rates is one of the most important labour market developments of the past 15 years. From 1948 to 1981, the unemployment rate in Canada was, on average, the same as that in the United States. That changed during the 1980s when the Canadian unemployment rate averaged more than two percentage points higher than its American counterpart. In the 1990s, the gap rose to almost four points. A comprehensive overview of this issue is provided in The Canada-U.S. Unemployment Rate Gap: An Assessment of Possible Causes, a report prepared by Andrew Sharpe. The report was presented at a conference on the Canada - U.S. unemployment rate gap sponsored by Human Resources Development Canada and the Canadian Employment Research Forum.

Text version

Some Facts

There is a fundamental difference in the labour market dynamics behind the development of the unemployment gap of the 1980s and the increase in that gap in the 1990s.

Between the 1981 and 1989 cyclical peaks, the Canadian labour market performed relatively well. The participation rate and the employment-population ratio rose. Labour force and employment levels grew too. After rising during the 1982 recession, the unemployment rate returned to its pre-recession level by 1989. But the American market performed even better. A Canada-U.S. unemployment rate gap emerged partly because the U.S. unemployment rate fell two percentage points from its pre-recession level.

Between 1989 and 1994, the Canadian labour market performed poorly and the unemployment rate rose significantly. The unemployment rate gap grew two additional percentage points because American labour markets recovered more rapidly from the 1991 recession. Average annual employment growth in the United States was 1.0 percent over that period compared to 0.3 percent in Canada.

Text version

A noteworthy feature of the unemployment gap - the increase in the unemployment rate in Canada relative to the United States between 1981 and 1994 - is a generalized phenomenon, affecting all groups more or less equally, including both genders, all age groups, regions, industries, occupations and skill groups. But again, the 1981-89 and 1989-94 periods are somewhat different from each other. In the 1980s, the unemployment gap for females experienced a disproportionate increase while in the 1990s, the gap for males increased strongly. From a regional perspective, the gap increased much more in Western Canada in the 1980s and in Ontario during the 1990s.

 

Text version

Another interesting feature is the importance of increased duration of unemployment spells in accounting for the emergence of the gap. Between 1981 and 1989, average unemployment duration in Canada rose from 15.1 to 17.9 weeks, while in the United States both incidence and duration fell with the overall decline in the unemployment rate. In the 1990s, average unemployment duration in Canada rose further, hitting 25.7 weeks in 1994, with the incidence of unemployment actually lower than in 1989. In the United States, the small increase in the unemployment rate between 1989 and 1994 was associated with a very steep fall in incidence. The increased length of unemployment in Canada further explains the emergence of the unemployment rate gap between the two countries.

Other Factors

Canada and the United States employ household surveys (the Labour Force Survey and the Current Population Survey respectively) to measure the unemployment rate. These surveys are comparable but minor differences do exist. Persons whose only job search activity is looking at help-wanted ads are counted as unemployed in Canada, but excluded from that category in the United States. The author reports that this definitional difference in the measurement of unemployment accounted for about one-fifth of the Canada-U.S. unemployment rate gap in 1993.

The report looks at the evolution of the "non-accelerating-inflation rate of unemployment" (NAIRU). The NAIRU is an important concept, indicating the rate of unemployment that can be sustained over time without causing inflation to rise. As a result, the NAIRU is a measure of the rate of unemployment that provides an indication of the relative importance of cyclical and structural unemployment in the two countries. Cyclical or demand-deficient unemployment is defined as the difference between the actual unemployment rate and the NAIRU.

According to the author, a "reasonable" estimate of the NAIRU in 1994 was around 7.5 percent in Canada and around six percent in the United States. The actual unemployment rate was 10.4 percent in Canada and 6.1 percent in the U.S. - a 4.3 percentage point gap. The implication? Cyclical causes accounted for two-thirds of that gap (calculated as the difference between the Canadian actual unemployment rate and the NAIRU). The remaining one-third of the gap (the difference between the Canadian and the American NAIRU) was structural.

To explain the structural differences, the author examines unemployment insurance in both countries. In Canada, the UI system is more generous than in the United States. Not only does it offer a higher benefit replacement rate and more extensive coverage but it also requires less prior work than the typical UI program in the United States. In the United States, unemployment insurance is a state responsibility, with the result that conditions vary from one state to the next. Research indicates that the greater generosity of the Canadian system raises labour force participation and increases the duration of unemployment in Canada relative to the United States.

Differences in generosity of UI, union density, and labour compensation partly account for the unemployment rate gap.

Increases in the generosity of Canada's UI may account for the emergence of the unemployment rate gap in the 1980s but not for the growth in the gap in the 1990s. There are timing issues to explain, however. The relative generosity of UI in Canada increased with the 1971 reform and the 1978 introduction of regionally extended benefits. But the sustained Canada-U.S. unemployment rate gap emerged only in the 1980s. This casts doubts about the role of UI on the emergence of the gap in the 1980s. According to one argument, the lag can be explained by a stronger macroeconomic environment in Canada during the 1970s that served to mask the increase in structural unemployment arising from UI changes. Another argument is the possible impact on unemployment of the introduction of regionally extended benefits in 1978.

The 1990s present another timing issue. The Canada-U.S. unemployment gap hit record levels in 1993-94, when the generosity of Canada's UI system was falling significantly. Behavioural lags in the response of individuals to the UI changes could partly account for this paradox.

A growing gap between the density of unions in Canada and the United States, arising from a falling unionization rate in the United States and a roughly stable rate in Canada, has been suggested as a possible explanation of the unemployment gap. Empirical importance of this factor is unclear, however. No major studies of the Canada-U.S. unemployment gap have explicitly looked at the effect of unionization on unemployment. The lack of attention to this issue may be due to the fact that the exact nature of the unionization-unemployment linkages are difficult to specify. In any case, unions have not increased unemployment in Canada - since the unionization rate has not risen - but the decline of density in unions in the United States may have reduced unemployment in that country.

Real labour compensation increased at a faster rate in Canada than in the United States in both the 1980s and 1990s. It is possible that this led to less substitution of capital for labour in the United States and, therefore, faster employment growth. Poorly paid workers in particular have experienced a more severe deterioration of their wages in the United States. This may have led to greater employment opportunities relative to comparable Canadian workers.

Finally, the report finds little evidence in either country that the differences in demographic structures, the pace of structural change or the extent of labour market mismatch has contributed to the sustained increase in the unemployment gap between Canada and the United States in either the 1980s or 1990s.

In summary, the most compelling explanation of the emergence of the Canada-U.S. unemployment rate gap in the 1980s is that it is rooted in a number of structural factors leading to an increase in the NAIRU in the 1970s. However, the much stronger Canadian economy during that decade masked this increase in structural unemployment in Canada and delayed the emergence of a gap until the late 1970s. In the 1980s, Canada's macroeconomic performance was no longer superior to that of the United States and Canada's higher NAIRU led to a higher actual unemployment rate. In the 1990s, Canada's relative economic performance deteriorated significantly, resulting in a further widening of the unemployment rate gap even though there was no major change in the NAIRU in either country.

Previous

Contents

Next