Directorates
·	Director General's Office
·	Data Development
·	Strategic Evaluations
·	Partnership and Continuous Evaluations
Publications
·	Evaluation Reports
·	Evaluation Briefs
·	Labour Market Development Agreement Series
·	Lessons Learned Series
·	Evaluation Tool Kit Series
·	Evaluation Forum Newsletter
·	From Theory To Practice

• Introduction
• Background
• Methodology
• Key Empirical Findings
• Conclusion

Introduction

A recently completed evaluation of the Self-Employment Assistance (SEA) program provides evidence concerning the program’s short-term performance and the experiences of the first group of participants to complete the program.

Background

The SEA program was developed with the objective of “encouraging labour market self-sufficiency through self-employment”. The program was modelled on the predecessor Self-Employment Incentives (SEI) but with significant changes: SEA has been extended beyond Community Futures areas; it provides improved income support for UI claimants to the level of their regular UI benefit; it includes a mandatory training component; and it targets equity group clients.

The program is available to UI claimants, individuals in receipt of or eligible for social assistance, and participants in the Atlantic Groundfish Strategy (TAGS). Participants receive 52 weeks of income support - for UI claimants, their UI entitlement is extended by 52 weeks when they are accepted into the program. Delivery agents assist participants in developing a business plan and in obtaining financing, provide small-business related training, counsel participants on problems that develop as they establish their business, and monitor participants’ progress.

Methodology

Almost 1,500 SEA participants were surveyed about their program experiences, about the businesses they had established and about their post-program labour market experiences. A comparison group was constructed of UI claimants, matched to the participants by date of claim start, region and self-rated interest in participating in a self-employment program. No comparison group was possible for social assistance recipients (SARs).

This evaluation was conducted early in the life of the program. It covers program participants who entered the program during fiscal year 1992/93. These participants were interviewed in July 1994, on average, eight months after program completion. Thus, labour-market outcomes have been measured over a relatively short time-frame.

Key Empirical Findings

The SEA program has been successful in promoting labour market self-sufficiency among participants. SEA participants were significantly less likely to rely on UI during the post-program period. Only 7.2 percent of SEA participants claimed UI in the 12 months after the program, compared to 29.7 percent of comparison group members. However, approximately 4 years of reduced UI use will be required to recover the additional UI participants drew during SEA.

To some extent this is a mechanical effect of the program - participants do not accumulate insurable weeks while working in their own business. However, among comparison group members who established a business, 30.3 percent had established a UI claim in the 12 months following their reference claim.

SEA participants were also significantly less likely than comparisons to draw social assistance in the post-program period (4 percent vs. 10 percent) and, when they did resort to social assistance, received benefits for a shorter period of time (14 weeks vs. 29 weeks).

The majority (74 percent) of SEA participants were engaged solely in self-employment at the time of the survey, compared with 56 p¢ of the comparison group members who had started a business. Of those SEA participants who were not solely in self-employment, 6 percent combined self-and paid employment, 13 percent were working only in paid employment and 8 percent were not working. For comparison group members who had started a business, the comparable proportions were 15, 23 and 7 percent.

SEA participants reported higher total weekly earnings from all sources after program participation than before they entered the program. Their weekly earnings after SEA ($680) were also significantly higher than those of the comparison group ($457).

Participants worked an average of 56 hours at their SEA-sponsored business, compared to 42 hours per week in their pre-program job. Comparison group members experienced no change in hours worked from their pre-unemployment job to their post-unemployment job. Participants experienced a significant negative change in assets (-$2,818) while the comparison group increased their assets slightly.

When participants are compared with only those comparison group members who started a business, SEA-sponsored businesses were as profitable as the comparison businesses, with no significant differences found in monthly self-employment earnings or in business profits.

Eighty-three percent of SEA-sponsored businesses were still in operation at the time of the interview, a rate that is comparable with business survival of the comparison group businesses and with international experience in self-employment programming.

Business survival rates can be expected to decline in the future. In Ireland, 88 percent of business were still operating one year after the business start date, but only 60 percent after two years. In France, one year after business start-up, 84 percent of business were surviving, but this had dropped to 53 percent after 3 years.

Job creation in SEA-sponsored businesses is similar to that of businesses started by comparison group members. SEA businesses created, on average, 1.5 full-time jobs and 1.8 part-time jobs, representing an average of 5.6 months of employment per business. During 1993/94, the 1992/93 group of 12,349 SEA participants created the equivalent of 7,264 full-time jobs.

Local economies appear to benefit from SEA-sponsored businesses. Considering salaries to employees, goods and services purchased by the businesses and the reduced draw on UI in the post-program period, and adjusted to account for program “dead weight” and the increased draw on UI during the program period, participants who began SEA in 1992/93 generated a net economic benefit of $188 million during fiscal 1993/94. As a result, SEA begins to produce a positive return to society 6 to 10 months after program completion.

The proportion of visible minority, disabled and aboriginal SEA participants is low compared to their proportion of the comparison group (about 50 percent lower for all groups).

These groups fared less well in their businesses after SEA than did non-equity participants: they were less likely to receive a loan from a BDC or a bank; their business was significantly less likely to survive (78 percent vs. 84 percent); they reported significantly lower personal earnings from their business; and they were somewhat more likely to collect social assistance in the post-program period. However, when earned income was considered, their earnings were no different from non-equity groups and they were no more likely to collect UI after SEA completion.

SAR participants experienced considerable barriers to starting a business, having a lower initial personal equity in their business, being more likely to require a loan, and having more difficulty in obtaining a loan from a bank (they relied mainly on loans from BDCs and family and friends).

In spite of this, SAR participants did as well as non-SAR participants on all measures and earned significantly more from their businesses than non-SAR participants. Their businesses were more likely to be home-based and less likely to hire employees than non-SAR businesses. As well, 49 percent ceased collecting social assistance benefits.

The amount of training received was not a significant predictor of business success, as measured by business survival, employment created and sales. The number of training hours varied significantly across the country from a low of 45 hours in Newfoundland to a high of 250 hours in Saskatchewan. A puzzling finding is that one-third of participants reported receiving no training at all, in spite of the inclusion of a mandatory training component.

Since the international experience on self-employment programming indicates that review of business plans, training and advising can reduce failure rates by up to 50 percent, these results were unexpected. In future evaluations it would be worthwhile to examine in greater detail the type of training received by participants.

Business success was associated with higher levels of initial investment, receiving financing from a bank, managerial experience in a previous job, and greater number of weeks employed in the 24 months prior to SEA. Higher levels of business sales were obtained by business operated by men, businesses in the Sales industry, businesses in the Quebec and Atlantic regions, and businesses operated by individuals who’s spouse was employed.

Although the program appears to be successful in helping its clients make the transition from unemployment to self-employment, it also appears to have provided services to individuals who would have started their business even in the absence of the program. Non-incremental program effort, or “dead weight”, is estimated to fall between 40 and 60 percent.

Conclusion

SEA is a viable method for increasing labour market self-sufficiency, at least in the short-term. Compared to unemployed persons who had an interest in self-employment but did not enter the program, participants used less UI and less social assistance, had higher earnings, and earned more of their income from self-employment.

However, the program appears to suffer from conflicting objectives. Given the high level of non-incremental program effort (40 to 60 percent), and the high level of labour market self-sufficiency of many participants before they applied to the program, it must be considered whether the program should be selecting those most likely to establish a successful business or those who have potential to build such a business but require significant help, such as that provided by the SEA program, to achieve that goal.