STRATEGY Self-Employment Training Programs for the Poor Steven Balkin Roosevelt University ABSTRACT Asastrategytohelplmvincomepeoplecreatetheirownjobs,numerous trainingprogramshavebeen developed to assist the poor to start a small business. This effort is becoming uiidespread in both the U.S. and Europe. While there seems to be a consensus of what a self-employment training program should contain, the appropriateness of the typical program, for low income people ivith severe educational deficiencies, is questioned. A typology for differentiating self employment training programsis generated and their distribution is described from a survey of self-employment training programs for loiv income people in the United States. INTRODUCTION Assisting low income people into self-employment activity has its American roots in immigrant and ethnic mutual aid societies (3, 23, 9). 'hey arose among diverse populations such as Asians, Eastern Europeans, Afro-Americans, and West Indians. These societies provided a network of business contacts, knowledge of market niches, opportunities to apprentice, loan funds, and a supportive environment that inculcated a strong work ethic. Owning a business has been part of the American dream for a long time. The Roper Organization (30) conducted a telephone survey of adults to determine their views on ways to improve their economic situation through different types of employment. The highest rating was given to business ownership. Owning a business has been used strategically in two ways: as an advancement in one' career path, and as a survival mechanism by being the employer of last resort. Frequently the purpose behind business ownership is related to one's level of formal education. Highly educated people are more likely to use business ownership to further their career plans while those with low levels of education and skills frequently use it to provide a survival safety net. As a public policy strategy to create jobs and alleviate poverty, there has been great interest in programs that assist low income people to start businesses. The growth in client demand for these programs is from persons who have become unemployed, underemployed, or pushed out of the labor force due to structural changes in the economy, and from people experiencing changes in life circumstances such as becoming widowed, disabled, acqui ring a criminal record, or being forced to retire. The growth in the provision of these programs isa response to this client demand and a search for alternative ways to provide job training, reduce poverty, and accom- plish job creation. Part of this search for providing alternative training stems from a view that many poor people already engage in commercial activities in the informal economy and self- employment training may develop some of those underground skills into aboveground businesses (3, 12). 47 The spurt in selfwmployment training activity will be briefly described. This will be followed by the development of a typology for categorizing self-employment training programs and a description of the distribution of types of programs extant in the United States. SELF-EMPLOYMENT TRAINING ACI'IVITY Europe Fifteen European countries have entrepreneurial assistance programs directed to the unemployed, as do both Canada and Australia (5, 6). Among the most notable programs are the United Kingdom's Enterprise Allowance Scheme, France' Chomeurs Createurs (Unemployed Entrepreneurs), and Italy's Youth Enterprise Creation. In the United Kingdom, adults who have been unemployed 13 weeks or longer, have 1000 pounds to invest, and intend to work full time at their business, are entitled to select themselves into the program and receive a flat-rate allowance of 40 pounds a week for 52 weeks. Participants are eligible to obtain free business counseling but receiving technical assistance and preparing business plans are not mandatory for participation (6, 5). In France, a citizen who is entitled to unemployment compensation or welfare can apply to collect their cash benefits in a lump sum in order to start a business. The amount increases if additional employees are hired. Receiving technical assistance is not mandatory, although participants can use any of the many publicly supported entrepreneurial outreach support programs. Participants have to prepare a business proposal and pass a screening system (6, 5). In Italy, a person between 18and 29 years old who resides in the Southern half of Italy can apply for grants and loans to start a business (26). Grants can be for up to 60% of initial start-up capital; loans can be for up to 30% of start-up capital; grants are also available for part of the first three years operating expenses. Businesses in the retail and food service sectors are excluded and funding can not be used for purchase of an existing business. Potential participants have to prepare an elaborate business plan with the aid of a local business support group such as a chamber of commerce or trade association. A panel of business experts picksthe best plans to fund.'nited States The U.S. Department of Labor is currently engaged in an elaborate random assignment control group evaluation of projects providing financial capital and entrepreneurial training to recipients of unemployment insurance. The projects are in the states of Washington and Massachusetts and are modeled after the British and French transfer payment diversion programs for self-employment (33). The Self-Employment Investment Demonstration (SEID), funded through the U.S. Department of Health and Human Services and conducted by the Corporation for Enterprise Development and the Manpower Development Research Corpora- tion, is another large self-employment training evaluation project. It is directed towards women on welfare and is taking place in demonstration sites in Iowa, Maryland, Michigan, Minnesota, and Mississippi (15). In the first year of the program, 543 recipients of Aid to Families with Dependent Children (AFDC) have attended orientation sessions and 187 people have enrolled (16).This demonstration is based on a model developed by the Women' Economic Development Corporation in Minneapolis. After a rigorous process of screening and self-selection, enrollees participate in group workshops that teach business skills related to starting a business. Then enrollees must complete a business plan and are encouraged to seek financing for their enterprises. Most SEID programs have in-house loan funds attached; participants are also encouraged to seek loans from private banks. Those who open a business are exempt, for one year, from AFDC regulations that would handicap the operation of a business. Follow-up technical assistance is also provided (15, 17). 48 At the purely State level, numerous States are funding self-employment training programs on their own (27). The leader in the nation is Illinois, which has recently spent $800,000 to fund 23 self-employment training programs in poor and distressed areas (21,-8). Comprehensive Employment and Training Act. An early modern precursor to this spate of programs occurred during the late 1970sand early1980s under the Comprehensive Employment and Training Act (CETA) (4). Only a very small portion of programs that operated under the Comprehensive Employment and Training Act involved entrepreneurial training, and their existence largely went unnoticed.'nly 21 CETA programs provided entrepreneurial training that was oriented to business ownership after the program was completed. They did demon- strate that CETA eligible clients could be assisted to start their own businesses.'he programs, however, were small, and costs per participant were greater than other CETA training programs. The businesses that were started created jobs mainly for the owners. Substantial "creaming" occurred and participants were generally not the poorest of the poor (4). In 1982 CETA was discontinued and replaced by the Job Training Partnership Act (JTPA), under which self- employment training programs continue to be funded. Formal of Self Employment Training Programs. There seems to be a consensus of what a self- employment training program should do: screen applicants to pick the most likely candidates to initiate a business; provide classroom training on business skills; teach about starting businesses in general; focus on preparing a business plan; and assist a client to obtain a source of financing (19). All the U.S. programs mentioned above are of this type. Contemporary self-employment training programs are more varied than those in the past, as described above. For some projects, new program elements have been introduced and some of the old elements have been rearranged. Following the demise of CETA in the early 1980s, a great growth occurred in self-employment training programs targeted to the poor. A survey was conducted in 1988 to describe the distribution of the types of those self-employment training programs. A description of a program typology follows. TYPES OF SELF-EMPLOYMENT TRAINING PROGRAMS As a first attempt to develop a typology of self-employment training programs for low income people, projects can be classified according to the following three dimensions: target population, industry linkage, and business plan focus. From discussions with colleagues and operators of programs, these three characteristics seemed tobe the most salient for distinguishing programs. Further, there seems to be more variation on these characteristics than on others. Other attributes that could have been used to categorize programs are whether or not there is: a screening process, extensive academic classroom work, self-confidence improvement work, client access to a loan fund tied to the program, extensive post-training follow-up assistance, use of formal mentors, and reliance on government funding. Target Population Programs differ on whether they target a low income population in general versus targeting clients in a specific group, in addition to the low income criteria. Spedfic groups generally are categorized by demographic characteristics, type of transfer payment received, or by the cause of being low income. For example, some programs only serve women, youth, elderly, Hispanics, Blacks (demographic groups); welfare recipients, unemployment insurance recipients (type of transfer payment received); or dislocated workers, ex-offenders (cause of 49 being low income) (32).Some programs may combine elements of the possible targeting criteria; for example, women (demographic) receiving Aid to Families with Dependent Children (transfer payment). Industry Linkage Some programs specialize and train clients to start and operate a specific type of business versus training clients in entrepreneurship and business management in general. Examples of some of the businesses that programs specialize in are: child care, street vending of food, home sewing, and horticulture. General entrepreneurship training is provided on the assumption that participants need an idea for a business to initiate and can only become a business owner if they work on their own business concept. In contrast, training for specific businesses is provided by programs which feel the trainers should specialize in only the types ofbusinesses they know best, or those that are the most feasible to start; and it is not necessary for prospective participants to have their own business ideas. Business Plan Focus Some programs focus on preparing a formal business plan as the centerpiece of the project (e.g. two year cash flow projections, projected income and balance sheet statements, formal marketing strategy) versus not requiring clients to prepare a business plan. Similar to the latter, some projects interpret the notion of a business plan very loosely and make its preparation a small part of the program. The degree of focus on the business plan is likely a function of whether clients will be applying for business loans, and the educational background of the trainers. Many trainers in self-employment programs were educated in university business schools or read manuals prepared by business academics. According to a survey of small business educators, 80% viewed business plan preparation as an important element in a small business course (28). Program Type Matrix By considering all the combinations of these three program attributes, measured di- chotomously, there are eight possible mutually exclusive program types. These combinations are presented in the following matrix and numbered one through eight. See Figure l. Figure 1. Matrix of Eight Types of Self-Employment Training Programs for the Poor Specific Target Population General Target Population Business Plan Business Plan Business Plan Business Plan Important Unimportant Important Unimportant Industry Linked (I) (2 (3) (4) Training General Entrepreneurial (5) (6) (7) (8) Training 50 A SUMMARY DESCRIPTION OF UNITED STATES SELF-EMPLOYMENT TRAINING PROGRAMS SERVING LOW INCOME PEOPLE The population of programs here are those whose intentions are to assist a client in starting a business after the initial training part of the program is completed.'n 1988,as part of a research project, fifty such programs were found across the U S.that were currently operating or recently operated. The enumeration was accomplished by drawing on an extensive literature search, attending conferences on self-employment training, and using data from the Corporation for Enterprise Development and the National Conference of State Legislatures. A list of all 50 programs is available from the author. Descriptions of most of the programs can be found in (1, 3,,10, 11,13, 14, 19, 20, 24, 25, 27). Programs that do not predominantly serve low income people were not included.'or example, there were many programs serving minorities and women but that either do not serve very many low income people or intentionally screen them out. Results In Table 1, the matrix in Figure 1 is reproduced with the percentage inside each cell being the relative frequency for each of the eight types of programs. The cell percentages are proportions of the total 50 programs. See Table 1. Table 1. Relative Frequencies for Eight Types of Self-Employment Training Programs for Low Income People (n=50) Specific Target Population General Target Population Business Plan Business Plan Business Plan Business Plan Important Unimportant Important Unimportant Industry Linked (1) (2 (3) (4) Training 2% 12% 2% 2% General Entrepreneurial (5) (6) (7) (8) Training 42% 4% 32% 4% Programs that target a specific low income group are slightly more numerous than ones that do not. This is observed by comparing the left half of the table (60%) versus the right half of the table (40%).Programs that provide general entrepreneurial training are much more common than industry linked programs. This is observed by comparing the second row of the table (78%) to the first row (18%).Programs that focus on a business plan are much more common than ones that do not. This is observed by comparing the first plus the third columns (78%) versus the second plus the fourth columns (22%). 51 The most commonly occurring program is type ¹5: general entrepreneurial training, focusing on a business plan, and serving a specific target group. Forty-two percent of all the programs are of this type. Pmgrams of type ¹5 and type ¹7 (serving general low income clients but also a business plan oriented general entrepreneurial program) together comprise almost three quarters of all the projects. Thus, programs that provide general entrepreneurial training and focus on clients preparing a business plan are by far the most common. One reason for the predominance of the business plan focus is the influence of bankers and business academic textbook writers. Business plans are important if bank financing is being sought, the business is complex, and/or substantial growth is expected to occur.'owever, if the business is very simple, the size will be limited to no or few employees, no external financing is sought, and validated easy rules-of-thumb operation are available, perhaps the requirement of c)ients preparing a business plan is not necessary. One important function that business plans serve is to act as a screening mechanism to restrict the program to those with good literacy and math skills. The purpose of this screening is to increase success rates by admitting clients who, many times, would have started their businesses anyway, without the assistance of the program. This helps to generate a high success rate when success is defined by measures such as the percent of enrollees who started a business. However, from a social policy perspective this may not be a correct strategy because it may diminish the expected degree of impact.'mpact is defined as the outcomes of the program (e.g. increased business starts, improved earning streams, greater employment stability) that would not have occurred in the absence of the program. If clients are very likely to start their businesses even without the program, there is less likelihood of finding a big difference between the program outcome and the outcomes from a similarly situated control group who did not participate.'wenty-two percent of all of the programs did not focus their assistance on clients preparing business plans. Of the eleven programs that did not focus on client preparation of business plans, seven were industry linked. Program types ¹2 and ¹4 are those that are industry linked and do not focus on client preparation of a business plan. Essentially, these seven programs (15%of the programs) specialized in learning the essentials for operating and choosing a business in a certain industry and, therefore, could provide validated rules of operation for the business. Basically they operate like a franchise, and franchises purportedly have significantly lower failure rates than other forms of business." This can mitigate against the fear that targeting clients with severe skilldeficiencies will lead to high failure rates. Because of its modest demands on participants and its potential to achieve start-ups quickly, greater use of this type of program may permit more targeting on the poorest of the poor: those with very little formal education. These are people who would have great difficulty preparing a business plan and who would more likely not enter or would drop out of a self-employment training program that was classroom and analytically oriented. Sometimes self-employment training programs refer to orientation sessions, where the requirement of preparing business plans is presented, as "scare- off" sessions. The franchise approach should be explored as a means to develop self-employment programs that have the potential to reduce failure rates for low income clients. This could take the form of assisting clients into low cost franchises, stimulating the number and variety of low cost franchises, and having programs and assistance organizations act like franchisors by specializing in certain industries. If there are enough of these programs, and a client wants to start a business that a particular program doesn't specialize in, he/she can be sent to one that does. With the high rate of growth for these programs, there may be enough of them in a region so that clients can be served by a network of programs, rather than by only one. 52 It is, however, still too early to tell which type of program works or works best for which type of client. A variety of programs need to be tried with extensive evaluation implemented for the most promising ones. The evaluations should include use of control groups and long term follow-up. At this point we need to experiment and evaluate. CONCLUSION Variety in self-employment training programs ought to be encouraged because there are a variety of clients who have different needs and capacities. Some partidpants may be low in- come people but with a high level of formal education and skills who are looking for high profits and have high financial capital needs. Programs that focus on a business plan and provide general entrepreneurial training may be correct for this group because they are likely to need bank financing and their businesses will more likely be complex and have growth potential. The most common type of self-employment training program for the poor was of this kind. Other self-employment training participants may just want to enhance their survival capability or attain needed lifestyle flexibility. They may have a low level of academic skills, have a low tolerance for classroom education, and require little capital. Birch (7) would categorize this latter group as "income substitutors" rather than "entrepreneurs". They may be best helped by programs that specialize and train for starting a business in a spedfic industry without requiring the preparation of a business plan. We need to know what works, for whom, and at what cost. While there is reason to think that self-employment training belongs in the mix of programs geared to alleviate poverty, develop human capital, and create jobs, just how much of it should be there will likely depend on the development of new types of business advisory services that can help people with educational deficiencies to start a business. If programs can effectively target the poorest of the poor, show impact, and pass a cost-benefit test, there is reason to be optimistic about their growth. ENDNOTES 1. I want to gratefully acknowledge support from The Coleman/Fannie May Candies Founda- tion, The Joyce Foundation, the U.S. German Marshall Fund and Roosevelt University. I also want to thank the reviewers for their helpful comments. The views expressed are my own. 2. Part of the information on the Italian Youth Enterprise Creation program comes from my field notes from a research trip to Italy in 1988. 3. Prime sponsors were the local administrative bodies of CETA. In August 1982, an initial screen questionnaire was sent to all 524 CETA prime sponsors in an attempt to obtain a census of all CETA entrepreneurial programs. 286 responses were received, for a response rate of 53%. Of these 286 responses, 17% or 49 replied that they did have some sort of an entrepreneurial oriented program. It appeared that of these 49 entrepreneurial oriented programs, only 21 were genuinely entrepreneurial in the sense of having an ownership orientation, 12 seemed to be involved in providing training in just business skills, and 16 were very marginally entrepreneur- ial (e.g. training employees for local businesses) (4). 4. Different programs had different eligibility criteria depending on the title of the CETA law the program was under. The following are some CETA eligibility criteria: Eligible CETA clients could be unemployed for 30 days or more (in areas of 7% unemployment or more, this could be reduced to 15 days); underemployed which means part-time or full-time workers whose earnings 53 were below the poverty level or persons from families receiving AFDC or SSIor persons working part-time but seeking full time work; long term unemployed which means persons who exhausted unemployment insurance benefits or were ineligible for such benefits or those who had been unemployed 15 or more weeks; and low income long term unemployed which means persons unemployed for at least 15 weeks and whose family income was not above the poverty level and persons in families receiving AFDC without regard to their employment status. In addition, several groups were targeted to get special consideration. These included Vietnam-era veterans, ex-offenders, persons of limited English language proficiency, handicapped, women, single parents, displaced homemakers, minorities, youth, older workers, individuals lackingeducational credentials (e.g. high school dropouts), migrant workers, and public assistance recipients. 5. Many small businessprogramsarerunbycommunitycolleges,universities,and government agencies and are geared to provide either general business education or how-to workshops for particular types of business. But it is usually not expected that the student or client will initiate the business immediately after the course and post-program assistance is usually not provided. For a survey of community college and university programs in entrepreneurial education see Solomon (29). Many local private non-profit agencies and community organizations provide technical assistance, networking, and referral services for low and moderate income people who are interested in starting a business. Fora directory of such organizations in Chicago, see Balkin (1).There are also myriad forms of for-profit private sector assistance to assist someone into self- employment. These mostly consist of professionals, (e.g., attorneys, accountants, brokers) who will provide advice for a fee, and franchises. The most common form of assistance into self- employment is through informal networks of family, friends, and associates. 6. In this analysis, low income people includes the newly unemployed and the long-term structurally unemployed, as well as discouraged workers. Discouraged workers are not technically in the labor force but would be willing to work if suitable employment were found. These groups have different capabilities and aspirations. The newly unemployed have recent work experience, are likely to be more skilled, and have more savings, assets, and effective social networks. It would seem to be easiest for the newly unemployed to start a business,but it would be more difficult for their businesses to match their previous or potential earnings (including fringe benefits from a wage job). The long term structurally unemployed and discouraged workers maybe harder to work with to accomplish a business start,but it would be easier for their businesses to match or exceed their previous or potential earnings from the labor market. 7. The 1982 Survey of Business Owners (31) reports that only about a third of U.S. business owners used some borrowed capital for financing their start-up. 8. The Supported Work experiment was a random assignment control group evaluation of a job training program targeted to four groups: AFDC women, ex-addicts, youth, and ex-offenders. It found that the program had the greatest impact with the participants who were the worst off. The greatest impact occurred for groups with the lowest rate of employment and earnings, and higher crime and drug use rates (18, p. 45). 9. The poverty population is not static; people do leave poverty on their own. Only a relatively small proportion are in poverty or on welfare continuously for long periods of time. (22) It is contended that many training programs "cream" by selecting clients from the "temporarily" poor segment and then claim their exit from poverty as a program success. If this is the only process operating, targeting the poorest of the poor will more easily reveal the ineffective programs. 54 10. Literature, citing the International Franchising Association as the source, reports that franchises have the lowest rates of failure compared to non-franchises. Using Dun and Bradstreet data, along with data from the U.S. Department of Commerce, I could not replicate that finding (2). 11.Consideration should be given to supportive public policy and ancillary services that can enhance the success of self-employment training programs targeted to the poor. This involves such areas as expansionary macro-economic policies, welfare reform, national health insurance, day care subisidies, changes in tax reporting requirements, changes in zoning laws, incentives for outside investment in micro-enterprises, creation of micro-loan funds and rotating credit associations, incentives for business owners to act as mentors, and creation of peer self-help groups. It may be true that the poor can be effectively encouraged to enter setfemptoyment through initiatives such as the above without the necessity of training programs. REFERENCES 1. Balkin, S. Directory of the Chicago Self-Employment Program Network. (Chicago, IL: Roosevelt University, Department of Economics, 1987). 2. Balkin, S. Franchises as Potential Self-Employment Opportunities. Working Papers in Economics and Political Economy. (Chicago, IL: Roosevelt University, 1988). 3. Balkin, S. Self-Employmen%r Lore Income People. (New York, NY: Praeger Publishers, 1989). 4. Balkin, S. and Czechowski, D.E. Survey of CETA Entrepreneurial Programs: Supplement Report to National Commission on jobs and Small Business. (Sprtngfield, VA: National Technical Information Service, 1986). 5. Barker, P.S. Self-Employment Schemes for the Unemployed. ILE Notebooks No.10. (Paris, France: Organization for Economic Co-Operation and Development, 1989). 6. Bendick, M. Jr. and Egan, M. L. "Transfer Payment Diversion for Small Business Development: British and French Experience." Industrial and labor Relations Reoieto, (july, 1987, pp. 528-542). 7. Birch, D. L. job Creation in America. (New York, NY: The Free Press, 1987). 8. Chicago Tribune. "Business Training Gets Funding Boost." (February 27, 1990, Sec-2, p. 3). 9. Cummings, S. (ed.) Self-Help in Urban America: Patterns of Minority Business Enterprise. (New York, NY: Kennikat Press, 1980). 10. Corporation for Enterprise Development. The Entrepreneurial Economy, (june,1986,entire issue). 11. Feit, R. "Expanding Entrepreneurial Opportunity: An Alternative Route to Self-Sufficiency." Background Paper for the Corporation for Enterprise Development, (Washington, D.C.:1987). 12. Ferman, L A., Henry, S.and Hoyman M. (eds.) The Informal Economy, Special Edition of The Annals of the American Academy of Political and Social Science. (September, 1987). 55 13. Friedman, R. E. "Expanding the Vision of the Entrepreneurial Economy."The Entre- preneurial Economy (December/January, 1986, pp. 2-3). 14. Gould, S. and Lyman, J. "A Working Guide to Women's Self-Employment." The Entrepreneurial Economy (March, 1987, pp. 5-14). 15. Guy, C.A. "Overview of the Self-Employment Investment Demonstration." Evaluation Forum. (Forthcoming, 1990). 16. Guy, C.A., Fink, B.and Doolittle, F. Self-Employment Investment Demonstration -Early Implementation Issues. (New York, NY: Manpower Development Research Corpora- tion, 1989). 17. Hinds, M. "Door to a Business of One's Own Can Be Good Exit From Welfare." New York Times. (June 3, 1990.Sec. 1, pp. 1 and 20). 18. Hollister Jr., R.G., Kemper, P. and Maynard, R.A. (eds.) The National Supported Work Demonstration. (Madison, WI: University of Wisconsin Press, 1984). 19. Illinois Department of Commerce and Community Affairs. Organizing Self—Employment Training Programs: A Guide for Development Organizations. (Springfield, IL: 1987). 20. Illinois Department of Commerce and Community Affairs. Self-Employment Training Programs: Case Studies. (Springfield, IL: 1987). 21. Illinois Department of Commerce and Community Affairs. Self-Employment Training Program1989-1990Requestfor Proposals. ProgramAdministrationBureau. (Springfield, IL: 1989). 22. Levitan, S. A. and Johnson, C. M. Beyond the Safety Net. (Cambridge, MA: Ballinger, 1984). 23. Light, I. "Immigrant and Ethnic Enterprise in North America." Ethnic and Racial Studies. (April, pp. 195-216). 24. Mangum, S.,Tansky, J.and Keyton, J. (1988)."Small Business Training as a Strategy to Assist Displaced Workers: An Ohio Pilot Project. " Journal of Small Business Manage- ment. October, pp.14 -21. 25. McDermott, M. "Maryland Targets the Disadvantaged." Franchise. (May/June, 1986, pp. 42%5). 26. Minister for Spedal Intervention in Southern Italy. Giovani &1m presa. (June, 1987,entire issue). 27. Puls, B.From Unemployed to Self-Employed: A Program Analysis. (Denver, CO: National Conference of State Legislatures, 1988). 28. Ryan,C.C.,etal."Educator'sViewsonSmallBusinessTraining." journalofSmallBusiness Management. (Washington, D.Cz 1987, pp. 80-83). 29. Solomon, G. National Survey of Entrepreneurial Education 3rd ed.. (Washington, D.Cc U.S. Small Business Administration, 1986). 30. The Roper Organization. The American Dream: A National Survey by the Wall Street journal. (Princeton, NJ: Dow Jones and Company, Inc., 1987). 56 r 31. U.S. Department of Commerce, Bureau of the Census. Characteristics of Business Oroners. (Washington, D.Ca U.S. Government Printing Office, 1987). 32. U.S. Department of Labor (1989).firmstart- an Examination of Self—Employment. (Washington, D.Cc Employment and Training Administration, 1989). 33. Wandner, S.A. and Messenger J.C. The Self Employment Experience in the United States: Demonstration Projects in Washington State and Massachusetts. Unpublished Paper submitted to the Organization for Economic Co-Cooperation and Development. (U.S. Department of Labor, 1990). 57