This Selected Issues paper addresses some of the key policy and economic challenges facing the Canadian economy. The paper presents a new approach to predicting the business cycle in the context of the Canadian economy. This approach uses a range of parametric and nonparametric tests to gauge the ability of various indicators to predict turning points in the business cycle. The paper also presents a model that links the inflation rate to the business cycle and the rates of change in the exchange rate and in unit labor costs.


This Selected Issues paper addresses some of the key policy and economic challenges facing the Canadian economy. The paper presents a new approach to predicting the business cycle in the context of the Canadian economy. This approach uses a range of parametric and nonparametric tests to gauge the ability of various indicators to predict turning points in the business cycle. The paper also presents a model that links the inflation rate to the business cycle and the rates of change in the exchange rate and in unit labor costs.

VI. Unemployment Insurance Reform 1/

1. Introduction

The Canadian unemployment insurance (UI) system was set up in 1940 with the narrow objective of providing temporary income replacement for unemployed individuals. During the 1950s and 1960s, its mandate was broadened to provide for additional longer term income support. 2/ However, the adverse effects of the UI system became apparent with the rise in structural unemployment during the 1970s. In response, the authorities imposed restrictions on the UI program in order to limit its adverse impact on labor market adjustment. In particular, the benefit rate was lowered, entrance requirements became more stringent, and greater emphasis was placed on “active” labor market policies. 3/

In October 1994, the Government issued a policy paper on UI reform that discussed two approaches to reducing the generosity of UI and improving labor market incentives: (1) making the program less generous for all claimants; and (2) distinguishing between short- and long-term users. In December 1995, the Government presented to Parliament a proposed reform entitled the Employment Insurance Act that combined some elements of both approaches.

The Government’s proposals are described and reviewed in the discussion below. Section 2 describes the current UI system and the Government’s proposed reform, and section 3 provides an evaluation of the proposals.

2. The current system and the 1995 proposal

Under the current system, benefits are provided to workers who have recorded between 12 and 20 weeks of insurable employment during the year preceding unemployment; the duration of benefits varies between 14 and 50 weeks (Table VI-1). 4/ The maximum weekly benefit for most claimants is equal to 55 percent of average weekly insured earnings received during the previous employment qualification period. For claimants with low insured earnings, the replacement rate is 60 percent. The maximum level of insurable earnings is adjusted according to the average increase in wages and salaries over an 8-year period.

Table VI-1.

Canada: Unemployment Insurance Reform Proposals

article image
Source: Government of Canada, From Unemployment insurance to Employment Insurance: A Supplementary Paper, (Ottawa, 1994); and A 21st Century Employment System for Canada (Ottawa, 1995).

The 1995 UI reform proposal contained a number of changes to the UI system designed to improve the efficiency of the labor market. For example, the reform would increase incentives to seek part-time work by basing UI benefits on hours worked rather than on the number of weeks worked, and by abolishing the minimum hours requirement. It also would reduce benefits for previous claimants so as to discourage frequent use of the system. Outlays for active labor market programs funded by the UI premium also would increase and focus on five areas: wage subsidies, earnings supplements, self-employment assistance, job creation partnerships, and skills loans and grants. In addition, a reserve would be built up in the UI fund that would permit fixing the UI premium rate over the cycle, thus avoiding the need to raise premiums during an economic downturn. These measures are described in more detail below.

a. Changes of UI benefits

Under the proposed reform, employees would qualify for unemployment insurance coverage on the basis of the number of hours worked (420-700 hours depending on the regional unemployment rate) rather than on the number of weeks worked (previously 12-20 weeks). In addition, the reform would eliminate the disqualification of those working less than 15 hours per week for a single employer. As a result, these measures would extend UI coverage to an estimated 500,000 part-time workers and reduce the disincentive for searching for part-time work while unemployed.

Currently, the unemployment insurance benefit is based on an employee’s average weekly earnings while working, so that claimants in areas of high unemployment may receive benefits based on earnings over 12 weeks of work. The proposed reform would base the UI payment on earnings calculated over a fixed period of 16 to 20 weeks prior to becoming unemployed. Therefore, claimants who work for only a proportion of the fixed period would automatically receive less benefits. The effect of this proposal would be to reduce benefits for those having been unemployed just prior to the previous 12-week period. However, because of its potentially adverse impact in areas of high unemployment, this change would be phased in over six years.

The reform would make a modest distinction between frequent and infrequent users. The basic UI replacement rate would be kept at 55 percent for occasional users of the program. For frequent users, the replacement rate would be cut by 1 percentage point for every 20 weeks of benefits collected in previous years up to a maximum of 5 percentage points. Means-testing of benefits would also be made more stringent by reducing the level of annual earnings at which the income test begins to apply from $63,570 to $48,570 for occasional users and to $39,000 for frequent users. Benefit history for determining the intensity of past use of the UI program would begin in July 1996, implying a relatively slow phase-in of these measures.

The loss of benefits to low-income families as a result of the tightening of eligibility requirements partly would be offset by a new income supplement. This supplement would be provided to UI claimants with children and whose family income is below $26,000 (the current system supplements the replacement rate of low-wage claimants by 5 percentage points).

b. Changes to UI financing

The unemployment insurance system is self-financing and currently operates roughly on a pay-as-you-go basis, requiring increases in payroll taxes during periods of economic weakness. To address concerns that this feature of the system has exacerbated cyclical downturns, the reform would build a reserve in the UI account deemed sufficient to finance benefits during a recession at a stable premium rate. 1/

Another undesirable feature of UI financing is that the formula on which the maximum amount of insurable earnings (MIE) is based has caused the MIE to grow much faster than average wages; by 1995 the MIE was $42,380 per year and exceeded the average industrial annual wage by 40 percent. Because premiums are capped at the MIE, this divergence between MIE and the average wage has penalized high-wage workers who are infrequent UI claimants. To address this concern, the Government has included in its reform package a proposal to reduce the MIE to $39,000 in 1996 and freeze it at this level until the year 2000.

c. Chances to labor market programs

The UI reform package includes a commitment to improve job search assistance and to replace the numerous employment programs that are currently offered with five back-to-work benefit programs. The proposed reform would create a National Employment Service that will provide centralized information on re-employment possibilities for UI recipients. This information will be provided through job centers and kiosks featuring advanced on-line job-matching services. According to the Government, this service is expected to be sufficient to secure re-employment for most recipients; for those who need additional assistance to return to work, the Government is offering the following benefit programs: wage subsidies, earnings supplements, self-employment assistance, job-creation partnerships, and skills loans and grants.

Wage subsidies (up to 50 percent of the wage and lasting up to a year) would encourage employers to hire individuals who have little on-the-job experience. (The program is estimated to cover 60,000 persons at a cost of $3,000 per person.) Employee earnings supplements would provide income supplements to low paid employees. The purpose of the supplement would be to encourage UI recipients facing reduced earnings to leave the UI system before their benefits expire, in order to gain job experience and raise the likelihood of permanent employment.

Self-employment assistance would build on the existing self-employment program by providing increased assistance in planning and setting-up a business. Participants would continue to receive UI benefits for up to one year while establishing their new business. This program is intended to build on the success of the previous self-employment program, in which 40 percent of beneficiaries hired additional employees and, on average, one additional job was created per participant. 1/

Job-creation partnerships, would employ UI recipients on community-based public projects (this program is expected to provide about $500 million for 80,000 recipients). Skills loans and grants would provide UI claimants with resources to attend accredited institutions to develop job skills. To be eligible, the claimant will have to show initiative by presenting a personal career plan to the authorities and be prepared to contribute his or her funds toward the cost of training. This program requires the recipient to initiate the grant process instead of passively accepting training assistance as in the current system.

In addition to these initiatives, the federal government plans to withdraw from labor market policies that overlap with the provinces’ responsibilities for education and labor market training. To accomplish this change the federal government will phase-out the practice of pre-purchasing enrollment positions at community colleges and will withdraw from financing workplace-based training. Moreover, the federal government is considering giving control over its proposed employment programs to provincial governments or working with the provinces to harmonize the proposed federal programs with those at the provincial level.

3. A preliminary assessment of the reform proposals

The reform intends to correct policies that have hindered labor market adjustment in the past by tightening eligibility conditions and redirecting programs from passive income support to active labor market support.

a. Changes in UI benefits

The reform is likely to reduce disincentives for part-time work by providing employees with unemployment insurance irrespective of the number of hours worked in a particular week. As a result, employers would be less prone to limit the hours of part-time workers in order to avoid paying UI premiums. In addition, raising the number of weeks of work over which UI benefits are based could reduce the mis-recording of earnings and link insurance benefits more directly to true income earned.

The distinction between frequent and occasional users of the system also is a step forward in returning to the initial intention of the UI program to provide temporary income replacement during periods of unemployment. However, the proposed changes are relatively modest compared to the options that were considered in the 1994 policy paper. These included reducing the benefit rate by 1 percentage point for every two weeks of benefits (rather than for every 20 weeks that was stipulated in the proposal) and eliminating the regional variations in entitlement (which was untouched apart from reducing the maximum duration of benefits from 50 weeks to 45 weeks).

Regional extended benefits have been criticized in a number of studies for perpetuating temporary employment patterns and discouraging inter-provincial job search. 1/ However, recent research has cast some doubt on this conclusion. For example, Prasad and Thomas (1995) have failed to identify any change over time in relative unemployment rates between regions that are net beneficiaries of the system of regional extended benefits and those that are net contributors.

The 1994 discussion paper suggested that differentiating the cost of insurance for high risk employees and employers would improve the efficiency of the system. The reform would apply this criterion to employees by distinguishing between frequent and occasional users. However, no adjustment would be made to employers’ contributions. Feldstein (1973) has argued that the lack of risk-based premium rates is likely to encourage firms to respond to demand shocks through temporary lay-offs rather than through changes in wages and hours. Subsequent empirical analysis has confirmed this view. Topel (1983) estimates that incomplete experience rating in the United States may account for 30 percent of all temporary layoffs. Corak (1994) finds that peaks in the hazard rate (the probability that an individual finds employment) in Canada occur just prior to the exhaustion of UI benefits.

Moreover, the lack of experience rating tends to subsidize cyclical industries. According to Corak and Pyper (1994), the net beneficiaries of the UI system in Canada are concentrated in the construction, forestry, and agricultural sectors. Conversely, firms in the community, business and services sector pay a higher premium than otherwise. The subsidized sectors are heavily concentrated in the Atlantic Provinces, and enterprises in these sectors tend to be smaller than in other sectors. Therefore, the introduction of experience rating would reduce distortions to the economic structure of this region and to the size of businesses.

The proposed changes to the system’s generosity are modest and are therefore unlikely to have a significant effect on employment. However, it should be noted that the system’s generosity has fallen continuously since 1977 as a result of previous reform measures and that the effects of changes in UI generosity on employment are difficult to quantify because of the difficulty in isolating its effects on structural unemployment. 1/

b. Evaluation of active labor market policies

The efficacy of active labor policies in promoting employment is subject to considerable debate. For example, a drawback of wage subsidies, is the potential for deadweight (that is, the expenditure is spent on people who would have found employment anyway) and displacement (when the subsidized workers crowd out other employees). Snower (1994) notes that, even in the case of well-designed programs, about 50 percent of the employment gain represents deadweight and displacement.

The evidence is somewhat more encouraging for self-employment assistance. A study by Benus at al. (1993) of these programs in the United States indicates that they significantly increase unemployed workers’ chances of starting a successful business. Benus et al. also found that these programs increase the total amount of time spent employed and raise earnings. However, Katz (1994) suggests that these programs also are subject to deadweight losses. In particular, only 2-5 percent of UI recipients in the United States choose to avail themselves of this type of program and participants tend to be more highly educated and have higher previous earnings than typical UI recipients.

Job creation partnerships have had the least success in generating net employment gains and generally lead to the phenomenon of churning, in which UI recipients alternate between periods of community work and unemployment (see Hellman and Lehmann, 1990 for a study on similar schemes in Germany and Jackman and Lehmann, 1990) for a study on the United Kingdom). However, in areas such as the Atlantic Provinces, which have a large number of seasonal industries, these partnerships can help in stabilizing employment patterns. 2/

A large number of commentators caution that skills loans and grants may not be cost-effective. 1/ In particular, a large proportion of the unskilled are already old and therefore would not continue working for long enough to pay off the investment in training. In addition, the unskilled may have been initially poorly educated, making retraining harder and more costly. An additional argument for education subsidies is that they offset imperfections in credit markets that inhibit poorer segments of the population from financing schooling and training. However, questions have been raised regarding this hypothesis. For example, Murray and Herrnstein (1994) find that after controlling for a score on a combined achievement and ability test, measured family income plays only a small role in explaining school attainment.

Another factor that needs consideration in an assessment of active labor market programs is the fact that these programs are funded by the UI payroll tax. This tax, which is now $2.95 per $100 of insurable earnings, is considered to have a sizable adverse effect on employment. For example, in the OECD’s Interlink model every percentage point increase in the payroll tax “temporarily” increases the structural unemployment rate by the equivalent of 4 percentage points for one year. The Canadian Government’s 1994 discussion paper also estimates that a 15 cent reduction in the UI premium rate would increase employment by 25,000 to 30,000 over a two-year period. 2/


  • Bellman L. and H. Lehmann, “Active Labor Market Policies in Britain and Germany and Long-Term Unemployment: An Evaluation,” mimeo, London School of Economics (1990).

    • Search Google Scholar
    • Export Citation
  • Benus, J. et al., “A Comparative Analysis of the Washington and Massachusetts UI Self-Employment Demonstrations,” Abt Associates (November 1993).

    • Search Google Scholar
    • Export Citation
  • Bloom, H. et al., “The National JPTA Study: Overview of Impacts, Benefits and Costs of Title II-A,” Abt Associates (February 1994).

  • Buckberg, E. and A. Thomas, “Wage Dispersion and Job Growth in the United States,” Finance and Development, International Monetary Fund (Washington: June 1994).

    • Search Google Scholar
    • Export Citation
  • Cameron, S. and J. Heckman, The Nonequivalence of High School Equivalents,Journal of Labor Economics (January 1993), pp. 147.

  • C.D. Howe Institute, Unemployment Insurance: How to Make it Work (Ottawa: 1994).

  • Corak, M., “Unemployment Insurance, Temporary layoffs, and Recall Expectations,” HRDC Evaluation Brief (April 1994).

  • Corak, M., and W. Pyper, “Firms, Industries and Cross-Subsidies: Patterns in the Distribution of UI Benefits and Taxes”, HRDC Evaluation Brief (October 1994).

    • Search Google Scholar
    • Export Citation
  • Feldstein, M., “Lowering the Permanent Rate of Unemployment,” A Study for the Joint Economic Committee of the United States (Washington: September 1973).

    • Search Google Scholar
    • Export Citation
  • Government of Canada, From Unemployment Insurance to Employment Insurance: A Supplementary Paper, (Ottawa: 1994a).

  • Government of Canada, A 21st Century Employment System for Canada, (Ottawa: 1995b).

  • Heckman, J., Summary of Testimony Presented to U.S. Senate Committee on Labor and Human Resources (January 1995).

  • Heckman, J., R. Roselius, and J. Smith, “U.S. Education and Training Policy: A Re-Evaluation of the Underlying Assumptions Behind the New Consensus,” mimeograph, University of Chicago (June 1994).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, Canada: Economic Developments and Policies” SM/94/97 (Washington: April 1994).

  • International Monetary Fund, Canada: Economic Developments and Policies” SM/95/81 (Washington: April 1995).

  • Katz L., “Active Labor Market Policies To Expand Employment and Opportunity,” mimeo, U.S. Department of Labor (August 1994).

  • Murray, C. and R. Herrnstein, The Bell Curve, (New York: The Free Press, 1994).

  • Organization for Economic Cooperation and Development, 1993-94 Annual Review Canada (Paris: 1994a).

  • Organization for Economic Cooperation and Development, Jobs Study (Paris: 1994b).

  • Organization for Economic Cooperation and Development,1994-95 Annual Review Canada (Paris: 1995).

  • Prasad, E. and A. Thomas, “Aggregate and Provincial Labor Market Adjustment in Canada,” International Monetary Fund, mimeo (June 1995).

    • Search Google Scholar
    • Export Citation
  • Robidoux, B. and C. Dea, “The Natural Rate of Unemployment in Canada: A Statistically-Robust Estimation” mimeo, Department of Finance, (Canada: April 1994).

    • Search Google Scholar
    • Export Citation
  • Royal Commission on the Economic Union and Development Prospects for Canada, (Ottawa, 1985).

  • Snower, D., Converting Unemployment Benefits into Employment Subsidies,American Economic Review, (June 1994) pp. 320341.

  • Topel, R., On Layoffs and Unemployment Insurance,American Economic Review 73, (1983) pp. 54159.

  • United States Government, Economic Report of the President (Washington: 1995).

  • Van Rijckeghem, C.Endogeneity in Structural Unemployment Equations: The Case of Canada”, IMF Working Paper WP/93/94 (Washington: December 1993).

    • Search Google Scholar
    • Export Citation

Prepared by Alun Thomas.


For a discussion, see SM/95/81, pp. 66-77.


The OECD defines active labor market policies as policies that aim at improving access to the labor market and jobs, job-related skills, and labor market functioning.


Both eligibility requirements and the duration of entitlement depend on the unemployment rate in the claimant’s province.


The premium rate for workers also was reduced from $3.00 for every $100 of insurable earnings in 1995 to $2.95 in 1996.


Recent estimates suggest that reducing the replacement rate by 1 percentage point would have the effect of lowering the structural unemployment rate by 0.005-0.01 percentage point (see the estimates in 511/95/81, Van Rijckeghem, 1993 and Robidoux and Dea, 1994). Assuming that frequent users (i.e., those that have received benefits at least 3 times in the past 5 years) represent 40 percent of all users and that their behavior is representative of all claimants, the measures would reduce the average replacement rate by about 2 percentage points over the medium term. On this basis, the proposed reform would lower the structural rate by less than 0.1 percentage point. Of course, the assumption that frequent users are representative is very strong and tends to bias downwards the effect of UI generosity.


In fact, an additional three-year Jobs Fund of $300 million has been created for the Atlantic Provinces to ease the adjustment process in these provinces. The fund will support job creation by encouraging joint partnership initiatives between the federal and provincial governments, the private sector and local communities.


Katz (1994) concedes that it has proven difficult to improve the labor market prospects of youth who drop out of high school.” Evaluations of the major U.S. government programs offering relatively short-term skills training to disadvantaged out-of-school youth--the Comprehensive Employment and Training Act (CETA) and Title II of the Job Training Partnership Act (JTPA)--indicate that they have not succeeded in raising youth employment or earnings. Moreover, the JOBSTART program, while raising the educational attainment of participants, did not raise the earnings of participants in the first four years following entry into the program (Cameron and Heckman, 1993). The job training experience of disadvantaged adults is a little more favorable, in that both CETA and JTPA Title II training programs have succeeded in raising earnings significantly for women. However, the most successful program combined on-the-job training with job search assistance rather than short-term classroom training (Bloom et al., 1994). The difficulty of effectively improving the marketability of dropouts and the apparent high returns to formal schooling serve to underscore the importance of improving education incentives earlier in the life cycle (see Buckberg and Thomas, 1994 for more details).