Paper to First GEM Research Conference: Entrepreneurship, Government Policies and Economic Growth, Berlin, April 2004. Growth expectations by entrepreneurs in nascent firms, baby businesses and mature firms: Analysis of the Global Entrepreneurship Monitor surveys in Denmark 2000-2003 Torben Bager University of Southern Denmark, Department of Organization and Management Thomas Schøtt University of Pittsburgh, Department of Sociology Abstract Annual surveys undertaken by the Global Entrepreneurship Monitor project in Denmark during 2000-2003 document that entrepreneurs who are currently starting a firm expect their firm to reach a significantly larger size in 5 years than manager-owners of existing firm. Why is this so? Is it because growth-oriented nascent entrepreneurs more frequently than others fail to establish their firms? Or do entrepreneurs modify growth expectations as they gain experience and learn over the years? Addressing these questions comprehensively presumes a prospective research design, which our data at the moment cannot support. The data do, however, enable cross-sectional analysis of entrepreneur characteristics, identifying growth determinants, and relating these determinants to variation in the composition of nascent entrepreneurs and manager-owners of baby businesses and mature firms. These results do support the assumption that failure during the start-up process is more frequent among growth-oriented entrepreneurs than other starters. The authors acknowledge the financial support provided by GEM sponsors (Erhvervs- og Boligstyrelsen, Danfoss, Ernst & Young, Karl Petersen & Hustrus Industrifond, Industriens Realkreditfond, Vækstfonden og Tuborg Fondet).
2 Introduction An entrepreneur, who is starting a new firm or owning and managing a baby business (max. 3½ years old) or mature firm, forms an expectation about the future growth and size of the firm. They may expect expansion, stability, or (provided they have already expanded) contraction. The GEM project provides unique data for comparison of the growth expectations of starters and owner-managers through representative samples drawn each year. By adding these data for a number of years, statistical analysis becomes reliable. In the Danish case, each of the 3 categories of entrepreneurs adds up to about 200 during 2000-2003. As we demonstrate in table 1, the expected future size of the firm is substantially higher for nascent entrepreneurs than for the owner-managers of existing ones. About one third of the nascent entrepreneurs expect their firm to grow to a size larger than 10 persons in 5 years, while only about one fifth of the manager-owners expect to reach such size. The main question guiding this paper is: Why do we find such a substantial difference in the expected future size of firms? One explanation, which seems likely, is that growth-oriented nascent entrepreneurs more often than other starters fail to start their business. Likely reasons relate to financing, as they typically need to raise more money, and predictability, as their business ideas may be wilder and less predictable than those of other starters. Another explanation, which also seems likely, is that nascent entrepreneurs modify expectations as they gain experience. By enacting the firm, it gradually becomes clearer to the entrepreneurs what their firm is supposed to produce and what market opportunities are available. This experience-based learning process may either increase or reduce their ambitions and expectations, but perhaps the first option is less frequent than the latter, resulting in an overall reduction in growth expectations. The annual GEM survey data do not enable a prospective research design, studying a panel of entrepreneurs over some time and analyzing the sorting and learning processes as they occur. However, the GEM survey data we dispose of do open for cross-sectional analysis of the characteristics of the three groups of entrepreneurs, which may take us some way in understanding whether or not a sorting mechanism during the start-up process contributes to explain the observed difference in expectations. The paper therefore attempts to answer the following questions: - Are the personal characteristics of nascent entrepreneurs and manager-owners of existing firms identical? - Do personal characteristics - background characteristics as well as more proximate business related characteristics - affect growth expectations? 2
3 - Does the answering of the above questions suggest that growth-oriented nascent entrepreneurs more frequently than others are sorted out during the start-up process? - If affirmative, what can we do about it? The shaping of growth expectations Growth expectations stated by entrepreneurs are likely to be estimates influenced by their growth ambitions and the characteristics of the firm and its environment. Hence, when entrepreneurs are asked about the expected size of their firm five years from now relative to the present situation, they try to answer two questions simultaneously: Do I (and associates) want growth? Am I (and associates) able to achieve growth, given the resources I (we) command, and taking into consideration the characteristics of the firm related to environmental opportunities in its particular business field? In other words, they balance ambitions and dreams with reasoning on resources and opportunities when answering the question. In some firms, the estimate is relatively easy to make while in others it is very difficult. In mature firms growth expectations are likely to be more precise than in emerging firms because uncertainty is less: The firm knows what it is doing, whom it is employing, its market situation, etc. Moreover, manager-owners of innovative firms in turbulent environments are likely to have more difficulties in estimating the future size of their firm than those of stable firms in stable environments due to a higher level of uncertainty and complexity. However, growth expectations are also influenced by the motivation, reasoning and sensemaking of the entrepreneur, which depend on personal characteristics. Entrepreneurs do not respond the same way when put in similar situations due to variation in personal characteristics. In this paper we focus at the issue of personal characteristics. It therefore relates to the literature on the role of personal characteristics for firm formation and growth. Previous studies in this field have focused at background factors like gender, age and education, and more proximate variables like motivation, goals, competence and business network. This stream of research has, according to Delmar & Davidsson (1999), not been particularly successful in picking winners by relating to start-up conditions. Nevertheless, a number of studies have documented a relationship between the personal characteristics of entrepreneurs and the performance of firms: Stuart & Abetti (1990) demonstrated an impact of entrepreneurial and managerial experience; Brown & Kirschhoff (1997) demonstrated an impact on growth of entrepreneurs perception of resource availability; Chandler & Jansen (1992) and Shane (2000) demonstrated an impact of entrepreneurs competences and knowledge; Cooper, Gimeno-Gascon & Woo (1994) demonstrated an impact of gender, number of partners and industry-specific know-how, and results were confirmed in a later study by Dahlquist, Davidsson & Wiklund (2000); Bager & Schøtt (2002) demonstrated little impact of formal education on young firm growth, but an impact of the age of managerowners and the age structure of employees; Wiklund, Davidsson & Delmar (2003) demonstrated that expected negative consequences of growth influenced manager-owners 3
4 motivation for growth; and Hansen (1995) as well as Davidsson & Honig (2003) demonstrated an impact of personal business networks. Moreover, the aim of this article is not prediction of future growth performance by relating to start-up conditions, but simply to explore whether growth-oriented nascent entrepreneurs are sorted out more frequently than other entrepreneurs during the start-up process. Data from Global Entrepreneurship Monitor The GEM project includes a number of personal variables on the entrepreneurs. They can be divided into two groups of variables: personal background characteristics, such as age and gender, and more proximate personal business-related characteristics, such as business competence and business network. Personal background factors may influence growth expectations either directly or indirectly through more proximate variables, while personal business-related characteristics are likely to influence directly. The Global Entrepreneurship Monitor surveys have been conducted annually since 2000 in Denmark. The GEM population survey annually interviews slightly more than two thousand adults, drawn randomly from the adult population in the country, with approximately equal probability of inclusion. Being a random population survey, it includes a random sample of starters of new firms and owner-managers of baby businesses and mature firms. The random sampling enables confident generalization to entrepreneurs in the country. Starters are identified as those who answer yes to at least one of the following two questions: Are you, alone or with others, currently trying to start a new business, including any selfemployment, or selling any products or services to others? Are you, alone or with others, currently trying to start a new business or a new venture for your employer an effort that is part of your normal work? Owner-managers are identified as those who answer yes to the following question: Are you, alone or with others, currently the owner of a company you help manage, self-employed, or selling any goods or services to others? The so-identified owner-managers are then also asked: What was the first year the owners received wages, profits, or payments in kind? If that year is the year of the survey or one of the preceding three years, the firm is identified as a baby business. If that year is earlier, the firm is identified as a mature business. Owner-managers are asked about the present size of their firm by the question: Right now how many people, not counting the owners but including exclusive subcontractors, are working for this business? Owner-managers are asked about the expected future size of their firm by the question: Five years from now how many people, not counting owners but including all exclusive subcontractors, will be working for this business? Starters are asked about the expected future size of their firm by the question: How many people will be working for this business, not counting the owners but including all exclusive subcontractors, when it is five years old? 4
5 The respondents are also asked about their background such as age. Entrepreneurial competence is tapped by asking: Do you have the knowledge, skill and experience required to start a new business? Risk-willingness is elicited by asking: Would fear of failure prevent you from starting a business? Optimism is tapped by the question: In the next six months will there be good opportunities for starting a business in the area where you live? Networking is queried by asking: Do you know someone personally who started a business in the past two yours? Obviously, these operationalizations are neither perfectly valid nor perfectly reliable. The personal business related characteristics should be regarded as proxies for the variables as they only rely on the answering of one question. Moreover, some factors which previous research has pointed out as important for the survival and growth of firms are not included in the analysis due to lack of data, e.g. the sector-specific experience an entrepreneur has gained before he or she starts a business in that sector (Storey 1996). During 2000-2003 the GEM population surveys interviewed 8044 adults. Among these, the survey identified 552 starters of new firms (job-related starters were a little more frequent than autonomous starters) and 747 owner-managers, including 286 owner-managers of baby businesses and 347 owner-managers of mature firms. Some owner-managers did not report age of the firm, so they drop out of analyses. Indeed, data are missing here and there, and the number of cases available for examination differs from one analysis to another. Present and expected size of the firms Size here denotes people working in the firm, including the entrepreneur. The work force is estimated as one plus the number of people reported to be working for the business. One is added because the reported number excluded any owner. The sizes of the firms, as reported by the entrepreneurs, are listed in Table 1. Starters of new firms differ greatly in their expectations about the future size of their firms. Only 19 % of the starters expect their firms to be one-man shows, most starters expect their nascent firms to grow by adding more workers over the next five years. The typical size can be indicated as the median. The median of the expected future size of the nascent firms is 5 persons working in the firm. Baby businesses are mostly small, indeed as many as 40 % are one-man shows. The median current size is 2, undoubtedly often a family business employing wife and husband. Ownermanagers of baby businesses often expect their business to grow over the next five years. The median of the expected future size of baby businesses is 2-3. Mature businesses also tend to be small, as 31 % of them are one-man shows. The median of their current size is 2, and also the median of their expected future size is 2. Starters expectations thus exceed, typically, both current size and expected size of ownermanagers firms, whether they are young or mature firms. 5
6 Table 1. Size of firms, presently and expected in 5 years (entrepreneurs included). Nascent firms Baby businesses Mature firms Expected Present Expected Present Expected 1 person 19 % 40 % 34 % 31 % 34 % 2 persons 9 % 15 % 11 % 19 % 16 % 3 persons 9 % 14 % 10 % 10 % 11 % 4 persons 8 % 5 % 8 % 4 % 3 % 5-9 persons 21 % 16 % 17 % 18 % 17 % 10-99 persons 30 % 9 % 17 % 16 % 16 % 100-999 persons 3 % 1 % 2 % 1 % 4 % 1000 + persons 1 % 0 % 0 % 0 % 0 % Total 100 % 100 % 100 % 100 % 100 % No. of firms 187 256 235 283 270 Median 5 2 2-3 2 2 Expected change in the size of firms The above Table 1 of size of firms indicates their size at various times, but does not show the change in size of the firms individually. How do entrepreneurs expect the size of their firm to change? Only 19 % of the starters of new firms expected not to have any workers in their firm after five years, not counting the owner(s), so these are expected to be stable. The other 81 % expected to have some workers after five years of business, so these are expected to expand, as listed in Table 2. About half of the owner-managers of baby businesses reported their expected number of workers in another five years to be the same as their number of workers at present, so these are not expected to expand, cf. Table 2. Quite many expected an expansion whereas quite few expected a contraction. By far most of the owner-managers of mature businesses expected stability, as seen in Table 2; rather few expected expansion and few expected a contraction. Table 2. Expected change of firms. Nascent firms Baby businesses Mature firms Expansion 81 % 41 % 23 % Stability 19 % 53 % 67 % Contraction 6 % 10 % Total 100 % 100 % 100 % Number of firms 187 234 267 6
7 By comparison we see that starters of new businesses more frequently expect expansion than owner-managers of baby businesses, whereas owner-managers of mature firms least frequently expect expansion. That new businesses are expected to expand more frequently than older businesses is not surprising. Contraction is not possible in 100 % of the nascent firms, 40 % of the baby businesses and 31 % of the mature firms. Personal characteristics related to growth expectation Some entrepreneurs expect their business to expand, and others do not expect positive growth. The question here is, which kind of person expects the business to expand? We can formulate hypotheses, where each hypothesis specifies how growth expectation relates to a personal characteristic, Region: Urbanites expect growth more frequently than rural dwellers. Gender: Men expect growth more frequently than women. Age: Young ones expect growth more often than older entrepreneurs. Competence: Entrepreneurially competent expect growth more than others. Risk-willingness: The risk-willing expect growth more often than risk-averse. Optimism: Optimistic ones expect growth more often than less optimistic ones. Networking: People with business network expect growth more often than those without. To test the first hypothesis we find that there are 44 starters of new firms in the urban region and 89 % of them expected expansion, whereas there are 64 starters in the rural region and 82 % of them expected expansion (Table 3). The difference between the two percentages is too small (for the small samples) to be significant, so the evidence does not corroborate the first hypothesis for the nascent firms. For baby businesses the difference between 55 % and 30 % is significant and supports the hypothesis. The hypothesis is not supported by the data on the mature firms. In short, there is some evidence in favor of the first hypothesis about region. The hypothesis about gender is supported by the data in Table 3, as men expect expansion more frequently than women do, regardless of the phase. The hypothesis about age is also fairly well-supported by the data, as young ones expect expansion more frequently than older entrepreneurs except for nascent entrepreneurs. The hypothesis about competence is corroborated by the data, as entrepreneurially competent starters and owner-managers expect expansion more frequently than less-competent ones. 7
8 Table 3. Percentage of entrepreneurs expecting expansion; by phase of the firm and by region, gender, age, entrepreneurial competence, risk-willingness, optimism, and networking. Nascent firms Baby businesses Mature firms Percent N Percent N Percent N Region Urban 89 % 44 55 % 58 19 % 54 Rural 82 % 64 30 % 44 30 % 45 Gender Male 85 % 137 42 % 154 27 % 200 Female 70 % 50 39 % 80 13 % 67 Age Young 79 % 109 45 % 130 42 % 55 Older 83 % 78 36 % 104 18 % 212 Entrepr. Present 87 % 115 41 % 156 27 % 202 competence Absent 61 % 23 39 % 44 11 % 55 Risk- Present 80 % 165 43 % 185 25 % 216 willingness Absent 89 % 18 28 % 36 13 % 45 Optimism Present 86 % 108 38 % 106 27 % 119 Absent 80 % 44 40 % 80 24 % 102 Network Present 82 % 148 42 % 163 29 % 155 Absent 84 % 39 36 % 67 15 % 105 The hypothesis about risk-willingness is partly supported by the data, as risk-willing ownermanagers expect expansion more frequently than risk-averse persons. The hypothesis about optimism only supported limitedly, as optimistic ones expect expansion only about as frequently as less-optimistic entrepreneurs except for nascent entrepreneurs. The hypothesis about networking is supported, as networking entrepreneurs expect expansion more frequently than entrepreneurs who network less. Each of the above analyses relates expectation to one condition, while ignoring the other conditions. An association may be causal, or it may be caused by a common underlying condition, and, unfortunately, the above analysis cannot reveal that. However, the growth expectation determinants identified in this study (gender, entrepreneurial competence, riskwillingness and networking) have also been identified in other growth studies, making it likely that they do have an impact. 8
9 Characteristics of nascent entrepreneurs and owner-managers of baby businesses and mature firms So far the analysis has shown that some personal characteristics affect growth expectation. Therefore variation in the composition of personal characteristics in the three groups may suggest, particularly if that variation is found between nascent entrepreneurs and managerowners of young firms who frequently are likely to be identical with the original starters, that there is a sorting mechanism during the start-up process which affects growth oriented entrepreneurs more than others. Table 4. Distribution of nascent entrepreneurs and manager-owners of baby businesses and mature firms across the selected variables. Nascent Baby Mature Firms Businesses Firms Region Urban 40 % 57 % 55 % Rural 60 % 43 % 45 % Gender Male 73 % 66 % 75 % Female 27 % 34 % 25 % Age Young 58 % 56 % 19 % Older 42 % 44 % 81 % Entrepr. Present 83 % 78 % 79 % Competence Absent 17 % 22 % 21% Risk- Present 90 % 84 % 83 % willingness Absent 10 % 16 % 17 % Optimism Present 71 % 57 % 54 % Absent 29 % 43 % 46 % Network Present 79 % 71 % 60 % Absent 21 % 29 % 40 % Table 4 demonstrates that the group of nascent entrepreneurs is composed somewhat differently from baby business owners on personal characteristics (11 % more males, 4 % more young, 6 % more entrepreneurially competent, 7 % more risk-willing, 25 % more being optimistic, and 11 % more having an entrepreneurial network). As all of these variables correlated positively with growth expectation, varying from limitedly to strongly, the group of nascent entrepreneurs counts relatively many people with personal characteristics which correlate with growth expectation. Provided ownership shifts in baby businesses since they 9
10 were started have not implied a major change in the composition of the group of baby business owner-managers on personal characteristics, these results do suggest that a sorting mechanism during the start-up process contributes to explain the observed difference in growth expectations. However, the result should be looked at as only a first approximation. Through a prospective study of the already interviewed entrepreneurs we will during 2004 be able to study the sorting and learning processes more detailed, including clarifying to which extent the original entrepreneurs remain owners of the baby businesses. Moreover we need during the next phase of our research efforts to clarify whether the huge difference in growth expectations found in Denmark also applies in the other GEM countries. Policy implications Further studies will not only confirm the existence of a sorting mechanism, and whether experience based learning also contributes to explain the observed difference, we will also know more about the reasons why growth oriented entrepreneurs are sorted out more frequently than others. This clarification is needed before policies can be elaborated, but then it is possible based on other information in the GEM study to suggest national policies which can remedy the situation. If for instance a major reason for the sorting process is lack of finances, or a biased financial structure, the financial level and structure would be targets for new political measures. In Denmark, the GEM study for 2003 has revealed two biases in her financial structure which are likely to affect the starting process of growth oriented entrepreneurs more than other starters, namely lack of pre-seed capital and a bias in the Danish incubators towards technology driven start-ups and neglect of market oriented innovative start-ups which often grow faster than technology driven ones (Storey 1998). Conclusions Entrepreneurs differ in their expectations for growth. Starters of new businesses frequently have high expectations for growth, much higher than the expectations that owner-managers have for their baby businesses and mature businesses, and nascent entrepreneurs expect their firms to become larger than owner-managers in existing firms, whether they are young or mature. Growth expectations depend not only on the phase of the firm, but also on the entrepreneurs personal backgrounds such as region, gender and age and their business related characteristics such as entrepreneurial competence, risk-willingness, optimism and networking. 10
11 Expected growth correlates positively by a number of personal characteristics, in particular being male, having entrepreneurial competence and having a network encompassing other entrepreneurs. Moreover, the group of nascent entrepreneurs is composed somewhat differently from baby business owners on personal characteristics (11 % more being males, 4 % more being young, 6 % more being entrepreneurially competent, 7 % more being risk-willing, 25 % more being optimistic, and 11 % more having an entrepreneurial network). As all of these variables correlated positively with growth expectation, varying from limitedly to strongly, the group of nascent entrepreneurs counts relatively many people with personal characteristics which correlate with growth expectation. Provided ownership shifts in baby businesses since they were started have not implied a major change in the composition of the group of baby business owner-managers on personal characteristics, these results do suggest that a sorting mechanism during the start-up process contributes to explain the observed difference in growth expectations. However, this study, applying a cross-sectional design, can only indirectly provide us with insight into the dynamics of the shaping of growth expectations among entrepreneurs. Further studies applying a dynamic design, studying how growth processes and growth expectations evolve as entrepreneurs develop their firms, are therefore needed. References Bager, T. & T. Schøtt (2002): Growth of young firms determinants revealed by analysis of registries in Denmark. LOK proceedings 2002. LOK Research Center. Bager, T. & M. Hancock (2003): Global Entrepreneurship Monitor Denmark 2003, Part 1: The Danish Entrepreneurial situation; Part 2: The Growth of Danish Firms. Børsen/Syddansk Universitet. Brown, T. E. & B.A. Kirschhoff (1997): The effects of resource availability and entrepreneurial orientation on firm growth. Frontiers of Entrepreneurship Research, 1997: 32-46. Chandler, G.N. & E. Jansen (1992): The founder s self-assessed competence and venture performance. Journal of Business Venturing. 7 (3): 223-236. Cooper, A.C., F. J. Gimino-Gascon & C.Y. Woo (1994): Initial human and financial capital as predictors of new venture performance. Journal of Business Venturing, 9(5): 371-395. Dahlquist, J, P. Davidsson & J. Wiklund (2000): Initial conditions as predictors of new venture performance: A replication and extension of the Cooper et al. study. Enterprise and Innovation Management Studies, 1 (1): 1-17. Davidsson, P. & B. Honig (2003): The role of social and human capital among nascent entrepreneurs. Journal of Business Venturing, 18(3): 301-331. Delmar, F. & P. Davidsson (1999): Firm size expectations of nascent entrepreneurs. Frontiers of Entrepreneurship Research. Babson College. 11
12 Hansen, E. (1995): Entrepreneurial network and new organization growth. Entrepreneurship Theory & Practice, 19 (4): 7-20. Shane, S. (2000): Prior knowledge and the discovery of entrepreneurial opportunities. Organization Science, 11 (4): 448-469. Storey, D. (1996): Understanding the small business sector. Routledge, London. Storey, D. (1998): The Ten Percenters. Fast growing SMEs in Great Britain. (3 rd and 4th reports). Deloitte & Touche, London. Stuart, R.W. & P.A. Abetti (1990): Impact of entrepreneurial and management experience on early performance. Journal of Business Venturing, 5 (3): 151-162. Wiklund, J., P. Davidsson, F. Delmar (2003): What do they think and feel about growth? An expectancy-value approach to small business managers attitudes towards growth. Entrepreneurship Theory & Practice, Spring 2003. Correspondence to authors: Thomas Schøtt: Tschott+@pitt.edu Torben Bager: tob@sam.sdu.dk 12