Characteristics of nascent entrepreneurs in Germany

Liñán Alcalde, Francisco; Martín Martín, Domingo; González Rodríguez, Rosario Conference Paper Characteristics of nascent entrepreneurs in Germany 42nd Congress of the European Regional Science Association: "From Industry to Advanced Services - Perspectives of European Metropolitan Regions", August 27th - 31st, 2002, Dortmund, Germany Provided in Cooperation with: European Regional Science Association (ERSA) Suggested Citation: Liñán Alcalde, Francisco; Martín Martín, Domingo; González Rodríguez, Rosario (2002) : Characteristics of nascent entrepreneurs in Germany, 42nd Congress of the European Regional Science Association: "From Industry to Advanced Services - Perspectives of European Metropolitan Regions", August 27th - 31st, 2002, Dortmund, Germany, European Regional Science Association (ERSA), Louvain-la-Neuve This Version is available at: http://hdl.handle.net/10419/115638 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. 42nd ERSA Conference. Dortmund (Germany), 27-31 august 2002 Characteristics of nascent entrepreneurs in Germany Francisco LIÑÁN ALCALDE; Domingo MARTÍN MARTÍN; Mª Rosario GONZÁLEZ RODRÍGUEZ Departamento de Economía Aplicada I. Universidad de Sevilla. Av. Ramón y Cajal, 1; 41018-Sevilla. Tel.: 00.34.95455.4487/7184/7550. Fax: 00.34.95455.7547 [email protected] , dom[email protected] , [email protected] Abstract Currently, the rate of creation of new enterprises -or start-ups- is considered as one of the maximum exponents of competitiveness within the different economies. The interest regarding the appearance of new enterprises has made numerous governments in the continent implement a series of support measures in that direction. It is expected that the European productive fabric can be revitalized in this way, and therefore to be able to compete with the present superiority of the United States, especially in the field of the so-called “new economy”. However, very little is still known with respect to the specific characteristics of the individuals who create those enterprises. The knowledge about the figure of the entrepreneur could allow a greater effectiveness of the measures already in operation, as well as of others that may be implemented. In this work we seek to contribute to a better knowledge of the European entrepreneur through an analysis of the characteristics of nascent entrepreneurs in Germany, a major Central European country. Given that Germany is the most important economy in the continent, and taking into account the narrow cultural interrelations with the neighbouring countries, we understand that this study can be representative of that area in Europe. For this study we will be using data from the Panel Comparability (PACO) Project, which includes a longitudinal panel of German households. The Panel data provide information on economic variables and both personal and family characteristics that may influence the decision of creating a new enterprise. It allows us to analyze the features presented by the prospective entrepreneurs before the creation of their enterprises. Thus, we can approach to the ideal situation of studying which are the factors that cause the decision of settling down as an entrepreneur. 1 Characteristics of nascent entrepreneurs in Germany F. Liñán Alcalde; D. Martín Martín & M.R. González Rodríguez 1. Introduction The role of the entrepreneur in the process of economic development is nowadays broadly acknowledged. From the crisis of the seventies, a resurgence of the importance of small and medium enterprises in the economy took place, with these enterprises growing more and generating more employment. Besides these enterprises play an essential role in the R&D processes (Storey, 1994; Schwallbach, 1994; Cohen and Klepper, 1996). At the same time, as a logical consequence, research on those small enterprises and on the entrepreneur’s figure has acquired a greater relevance through time. Thus, the study of the process that leads from an inclination towards the entrepreneurial activity to the creation and development of one or several small enterprises has been called “entrepreneurship” (Gartner, 1990; Bygrave and Hofer, 1991; Woo et to the., 1991). Although it is still a relatively new field of study, where the agreements are still insufficient (Fiet, 2000), there is an important level of coincidence about the division of what we could denominate the “Entrepreneurial Process” (Krueger, 1993; Harvey and Evans, 1995) into a series of phases. So, those people who have not still taken any step towards the creation of an enterprise, but who are somehow prone to that activity, are considered Potential Entrepreneurs (Krueger and Brazeal, 1993). The individuals that are in the phase of effective creation of an enterprise are called Nascent Entrepreneurs (Carter et al., 1996; Reynolds, 2000), and their study is frequently known as Entrepreneurial Emergency (Cáceres, 1999). The study of the development and growth of the enterprise is termed Entrepreneurial Dynamism (Santos, 1998). Some authors stress that process view when they speak of the Theory of the Entrepreneurial Career (Dyer, 1994). Meanwhile, decision-makers have being paying an increasing attention to entrepreneurship and new small enterprises for their capacity to generate employment and to stimulate the economy. For this reason, current strategies of development (as Endogenous Development or Local Economic Policy) clearly consider, as one of their main instruments, the promotion of the endogenous entrepreneurial initiative (Vázquez Barquero, 1993, 1999; Garofoli, 1992; Maillat, 1988). Nevertheless, many of those 2 initiatives have had very limited results (Liñán, 2001). This circumstance has established the need to know more about the characteristics of those individuals who decide to start a new enterprise, and the factors that influence that decision. Therefore, the analysis of nascent entrepreneurs has acquired great recognition, because they are individuals that have already overcome the phase of potential entrepreneurs (where their entrepreneurial inclination had not still been developed, remaining latent), but have not completely established their enterprise yet (they are not still active entrepreneurs). It is, then, a very delicate phase of the entrepreneurial process that corresponds to the realization of the first concrete activities that will lead to the creation of a new enterprise. The study of the aspects that make a greater number of people involved in the process of creation of new enterprises is, therefore, an essential element for the effective promotion of entrepreneurship. Accordingly, our objective in this work will be to analyse the factors that affect the decision to become an entrepreneur through the study of the characteristics of nascent entrepreneurs. We will focus specifically on the situation of the most important economy in the European Union, Germany. Reynolds (2000) offers data indicating that the situation and characteristics of nascent entrepreneurs in the countries of the centre and the north of Europe is quite similar, so we understand that the German case can be sufficiently representative. Besides, although the data we have used are available for other countries, it is in Germany where the time period embraced by the database is wider. 2. Conditioning factors of the decision to become an entrepreneur Numerous studies have been carried out to try to identify which factors influence more decisively the decision to become an entrepreneur. However, the results have not been sufficiently conclusive up to now for different reasons. Firstly, many of those studies suffer from a series of quite serious methodological problems. Secondly, the lack of a sufficiently developed agreement on the concept of the entrepreneur and on the entrepreneurial process has made the approaches differ, providing sometimes contradictory results. Methodological problems in the measurement of entrepreneurship In the first place, it seems sufficiently established today that a lengthy time period exists from the moment the individual decides to be a entrepreneur until the enterprise is in operation. It is a process that may take from some months to several years (Wolters, 2000; Carter et al., 1996). Therefore, it is necessary to analyze the 3 individual's circumstances before their enterprise is already established. However, if it is difficult to access reliable data on the newly established enterprises, it is much more complex to obtain information about people that have not yet created their enterprise. So, it is not strange that, until very recently, empirical studies have used ad hoc crosssectional samples. These studies obtain information, through surveys or other means, from active entrepreneurs. As a consequence, the entrepreneur's post-creation situation is being considered as the causal factor for establishing the enterprise. At the most, the respondent is asked retrospectively about their circumstances when the enterprise was created, with the corresponding risk of a significant bias (Reynolds, 1997; Santos, 1998; Cáceres, 1999). Thus, the consistency of results is doubtful. In this sense, studies based on panel data allow a much more correct approach from the methodological point of view. Another question to keep in mind regards the method to obtain the sample. It is relatively frequent the use of postal (or phone) questionnaires sent to a sample of individuals out of some database. Besides the problem of working with cross sectional data, already pointed out, there is an additional problem regarding the low answer rate usually obtained by this method. Normally, the correctly completed questionnaires are a very reduced percentage: between 18 and 36% of those sent (Birley and Westhead, 1994; Carter et al., 1996; Westhead and Wright, 1998). There is the risk, again, of a significant bias in the sample. The individuals who do not answer may present certain characteristics that make them different, for what the sample would not be representative (Gartner, 1985). Finally, when using questionnaires, it is necessary to be very careful with their elaboration and with the manipulation of the data obtained, especially when we try to measure features not directly observable like the interviewee's opinion, their personality traits, their perceptions, etc. (Davidsson, 1991). In particular, when the essential motivation that takes the individual to the creation of the enterprise is analysed, it is necessary to be extremely careful. In this sense, Birley and Westhead (1994) found that the reasons to create the enterprise are very varied. These authors included in their questionnaire twenty-three statements in relation to motivation, they were later orthonormalized to identify seven main reasons for creation. However, it is interesting to note that a high level of agreement with some of those twenty-three statements could correspond with different core reasons (depending on the answers to other statements). Thus, the statement regarding the desire to obtain high incomes, appears with a high value both in the motivation “Need for Approval” and in “Instrumentality of wealth”. So, in the first case, individuals might desire wealth only as long as it can grant them 4 approval from the members of their community, while in the second case, it would correspond with the security it offers. Therefore, it is necessary to be extremely careful with the identification of the individuals’ motivations, trying to go beyond the most trivial questions. Conceptual problems In this section, it is necessary to highlight firstly, the problem with respect to personality traits. During a long time, researches have searched for a series of features of the personality that would allow to identify the future entrepreneur (McClelland, 1961). This conception -in their narrowest sensehas been severely questioned (Gartner, 1989). A more appropriate approach consists probably on pointing out that the individual's characteristics are one of the conditioning elements, although always in combination with the environment and the specific circumstances surrounding the individual (Gartner, 1985). Presently, due to the poor results offered by this “traits approach”, psychological studies focus on variables as the individual's perceptions and attitudes towards the entrepreneurial activity. These psychological variables would be the basic conditions of what we have called entrepreneurial potential (Krueger and Casrud, 1993; Krueger, 1993). However, a perception more or less favourable towards the entrepreneurial activity will be the result of individual's experiences and circumstances, and how he/she has interpreted and assimilated them. This explains why some authors have found, for example that being the first-born child, having parents entrepreneurs, or being a member of ethnic minorities influence positively the decision of becoming entrepreneur (Cáceres, 1999). It would not really be the circumstances themselves (being first-born) which promotes entrepreneurship, but rather the way in which the individual assimilates that circumstance. In certain subgroups of the population may arise a situation that helps that relationship (between, for example, primogeniture and entrepreneurial propensity). Again, sample selection can have a decisive influence on analysis results. Besides those subjective variables, we could consider what Krueger and Casrud (1993) call “precipitating events”, i.e., those environmental variables (economic opportunities, institutional framework…) that -in addition to their indirect effect through perceptions and attitudesexercise a direct effect on the decision of starting an enterprise in a certain moment of time and not in another. In the same way, the concept of “strategic windows” (Harvey and Evans, 1995) tries to show the need of a coincidence between personal characteristics and environmental circumstances for individuals to effectively start their entrepreneurial project. In this way, Birley and 5 Westhead (1994) found an important relationship among reasons for creation and age in which entrepreneurs start their venture. This is consistent with professional career theory in the sense that the reasons to start-up change along the different phases of the individual’s life. Along the same line, it is important to point out the great diversity existing among active entrepreneurs. As Gartner stresses (1985), the existing differences among entrepreneurs are probably higher that those between entrepreneurs on one side, and non entrepreneurs on the other. Thus, studies that try to classify active entrepreneurs in only two or three categories may be clearly inadequate. Woo et al. (1991) showed that the traditional division between “craftsmen” (traditional and imitators, preferring manual work and dedicating less time to strategic management tasks) and “opportunists” (innovative and dynamic, restless, attentive to the continuous exploitation of new opportunities) lacks enough rigor. According to these authors, several studies have derived that typology using different variables and approaches, so they would be inconsistent, and would refer to different realities -although using the same label to name them-. There is also an additional question to be addressed from the regional development policy. It regards the probabilities of success that different nascent entrepreneurs have. Its importance resides in the scarcity of public funds, so decisionmakers would like to identify and support only those entrepreneurial projects with higher possibilities of survival, and with a greater impact on the regional economy. In the first place, it would be necessary to define what to understand as entrepreneurial success (Santos and Liñán, 2001). In this sense, the variables used in the literature range from the mere enterprise survival to its growth and employment creation (Liñán, 2001). However, attempts to predict entrepreneurial success have not obtained satisfactory results up to now (Cooper, 1993). Anyhow, most promising results use models with a higher level of abstraction, combining economic variables with psychological ones (Davidsson, 1991) or with those relative to the entrepreneur's previous experience (Chandler, 1996). One can say therefore, that the entrepreneurial activity is not one-dimensional at all, but rather it is essentially complex and dynamic. So, as we have pointed out, the reasons that take the individual to the decision of creating an enterprise are expected to be different in each case. As a consequence, the variables that influence that decision can also be different for each nascent entrepreneur. Nevertheless, it does not imply that any attempt to look for a causal relationship is useless, although the results will be 6 necessarily modest until sufficiently wide theoretical models are developed, as well as databases specifically designed to contrast this type of models. Revision of some empirical analyses There are several studies that try to examine the factors explaining the decision to become a entrepreneur. However, many of those studies suffer from some of the methodological and conceptual problems identified above. Those problems would reflect on no significant coefficients, low predictive capacity, or spurious correlations. In spite of this, certain relevant regularities still exist. However, all this literature studies the transition to self-employment in a wide sense, and not the establishment as an entrepreneur specifically. Thus, in that literature, independent professionals, self-employed persons or farmers are considered equal to individuals that create a new venture. In our study, we will focus exclusively on those people that decide to become entrepreneurs by creating a new enterprise, because they may have a greater effect on the productive fabric and, through it, on the regional or local economic development (Santos, 1998). In the first place, the positive influence of the income level on the decision to be self-employed is the most broadly documented. Van Praag and Van Ophem (1995), for example, found that variable to be the most important determinant of what they call the “opportunities” to be entrepreneur. It is obvious that the creation of an entrepreneurial initiative requires the investment of resources; therefore the availability of them facilitates the decision. This relationship can be reinforced because a higher income level is usually related to higher qualifications and social status (contact networks). Thus, the individual’s perceptions of capacity and self-confidence may be greater. Another influence frequently found refers to age. That relationship usually has a bell shape (Reynolds, 1997). The probability of becoming an entrepreneur initially rises with age (up to thirty or thirty five years), to descend gradually and continually later. Specifically, Reynolds’ study found that people in the 25-34 age group had a probability of being a nascent entrepreneur three times that of the other groups. Considering now the educational level, it seems to have a positive effect on the decision to create an enterprise. Wolters (2000) found that the entrepreneurial propensity is higher among those individuals with university education. Other studies (Evans and Leighton, 1989; Blanchflower and Meyer, 1994) also find this relationship 7 statistically significant, with the effect increasing as the level of studies rises, reaching their maximum for university graduates. Other positive relationships have been repeatedly found in the empirical literature about enterprise creation (Storey, 1994): unemployment situation, availability of wealth, and previous experience. Regarding unemployment, it is usually considered that the loss of a previous employment (or not finding a satisfying one) acts as a “precipitating event” of the decision to become an entrepreneur. Availability of wealth (either the family’s or an inheritance) would act similarly as the income level, allowing the exploitation of a business opportunity that has been discovered. Previous experience, finally, allows the individual knowledge of the activity to be developed, making his/her self-efficacy perception rise (feeling of being able to successfully carry out that task). Besides, experience will probable help develop contact networks, very useful for business success. Among the analyzed studies about the characteristics of nascent entrepreneurs, the one by Reynolds (1997) is especially relevant, although his definition of nascent entrepreneur -as the other studies pointed outrefers only to becoming self-employed. That analysis is based on panel data and uses a model of logistic regression, getting very interesting results that seem to confirm the existence of different “types” of nascent entrepreneurs. Therefore, Reynolds divides the sample into subgroups. Thus, between 25 and 34 years old the presence of nascent entrepreneurs more than triples that of the remaining age groups. Besides, within this subpopulation, self-employed, unemployed and students are those showing the highest nascent rate. In this group, employees only present a high entrepreneurial inclination if they have an educational level above secondary school. In the other age groups the rate of nascent entrepreneurs is more reduced. Anyhow, self-employment stands out within the 35-54 year group; and belonging to a household of three or more adults stands out within the 18-24 year group. With respect to its predictive capacity, the percentage of correctly classified cases in that study is about 75%. Nevertheless, it is necessary to keep in mind that it uses samples where half the cases correspond to nascent entrepreneurs, so that the null model (that predicts that nobody is nascent entrepreneur) would offer 50% of correct cases. On the other hand, the number of nascents is very small, so the results should be taken with caution. In the following section we describe the empirical part of our study. Working with a panel of households we are able to minimize some of the problems found in other 14 Lastly, the analysis shows that being the reference person's son (in a wide sense, including natural, adopted and cohabitant’s children) diminishes the probability of becoming an entrepreneur. In this line, Reynolds’ study did not find the number of children or adults living in the household to be significant (except in the age group from 18 to 24 years). However, our study shows that the position occupied by the individual in the household is significant for the transition to entrepreneurship. It may be that individuals, before their transition to entrepreneurship, prefer to settle down in their own household (so they would become the reference person) and to acquire certain work experience. Regarding the continuous variables, in order to interpret more appropriately the coefficients of the model, the marginal effects have been calculated. In the logit model, and in the case of continuous variables, the expression to be used is: [ ] ( ) 1 ( ) E y F F β β β ∂    ′ ′ = − ∂ xx x x, which will vary with the values of x. Two alternatives exist for this analysis. Firstly, we could evaluate that expression taking the sample means as x. Secondly, the marginal effects may be evaluated on each observation and the mean of all marginal singular effects calculated afterwards. If we use the first of those alternatives for age, the marginal effect of an additional year diminishes the probability of becoming entrepreneur in 0.00472. Using the second alternative, Figure 1 shows the mean marginal effect for each observed value of age. In this way, the variation in the marginal effect when age changes may be observed. Although the marginal effect is always negative, it increases in absolute value until approximately 30 years of age, and diminishes lineally in absolute value from that age. This means that for the youngest individuals, the probability of transition to entrepreneurship (if all other circumstances remain constant) decreases faster every year. However, above thirty years of age, the probability decreases more and more slowly (again, if all the other circumstances remain constant). However, when we represent the observed probability (the relative frequency) of becoming entrepreneur for each age, the graph is bell-shaped, consistently with other studies. This does not imply any contradiction, because of the ceteris paribus hypothesis assumed. In this sense, during the first years of working life, “qualifications” and “opportunities” of the individual (education, higher level of income, experience, 15 and so on) are improving significantly, making the total probability of becoming entrepreneur rise. On the contrary, above thirty years of age, these other circumstances tend to stabilize. It is then when the negative effect of age begins to be apparent, thus making the overall observed probability decrease. Figure 1: Marginal Efect of Age -0,0045 -0,0040 -0,0035 -0,0030 -0,0025 -0,0020 -0,0015 -0,0010 -0,0005 0,0000 15 25 35 45 55 65 Change Prob. The marginal effect of income, on its part, is relatively small. A thousand additional marks on the average individual hardly increase his/her probability of becoming entrepreneur by 0.01630. The graph in the Appendix shows the marginal effect by income intervals. A slightly growing trend may be appreciated (as income increases, the probability of transition to entrepreneurship rises), but this relationship is very weak. It may also be interesting to show the marginal effects for the dichotomic variables (that we have denoted by d). In this case, the marginal effect would be: ( 1 , 1) ( 1 , 0) P Y d P Y d = = − = = x x , whith x usually obtained using sample means of all other explanatory variables. Table 2 shows the marginal effects of the dichotomic variables that were significant. The first thing to highlight is the huge effect that having previous experience as an entrepreneur has on transition. However, this information should be considered cautiously, as the corresponding coefficient was not significant (possibly due to the reduced number of observations). The situation of being self-employed but not entrepreneur also has a great marginal effect (0.55269). A third group of variables with a high effect would include: to be unemployed, to be a farmer, to be a contributing 16 family worker, or working in a small company (up to 20 employees). In all these cases, the probability of suffering a transition to entrepreneurship rises considerably (between 0.25 and 0.40). The remaining variables have marginal effects below 0.15. Among the situations unfavourable to entrepreneurship (negative marginal probability), there are that of being apprentice, being son of the reference person, and working in social services. Table 2. Marginal effect in categorical variables β ββ β Mean Change Prob. ES4 1,75268 0,06301 0,35727 ES567 0,46604 0,24536 0,07032 WORK1 1,76240 0,00487 0,37142 WORK2 2,66152 0,10807 0,55269 WORK3 1,64049 0,01126 0,33992 WORK6 -1,53335 0,02770 -0,13036 NUM12 9,25707 0,05449 0,89070 SIZE1 1,60274 0,24323 0,28394 SIZE2 0,48965 0,14155 0,07652 PERIODO2 -0,23929 0,49924 -0,03342 EXTRANJ -0,31018 0,20244 -0,04072 SECTOR6 0,72400 0,10928 0,12054 SECTOR9 -0,88844 0,15038 -0,10070 EDUC3 0,41543 0,62618 0,05607 EDUC4 0,59293 0,12116 0,09545 RELAC3 -0,69114 0,12359 -0,08093 It can be said, then, that the highest marginal effects are concentrated on only a few variables. Thus, professional situation and activity sector are especially relevant, with all effects above 10%. Therefore, the kind of work experiences (and not only the duration of those experiences) would have a substantial effect on the probability of becoming an entrepreneur. 4. Conclusions In this paper we have tried to analyze the characteristics of nascent entrepreneurs in Germany, which may be representative of Central Europe. To do so, we draw on previous work that point to a certain sociocultural homogeneity in the different countries of that area. Thus, Reynolds (2000) found that the proportion of nascent entrepreneurs in the centre and north of the continent is quite similar. On the other hand, Wolters’ (2000) results for the Dutch case confirm ours in many aspects. In the first place, a series of important methodological and also conceptual questions have been dealt with. Regarding the former, some practices that are common 17 in the empirical studies analyzed may cause an important bias in the results. In particular, on the one hand, it is essential that the explanatory variables are measured before the individual becomes an entrepreneur, and not afterwards, not even with retrospective questions. Therefore, it is usually necessary the use of panel data. On the other hand, in the elaboration and manipulation of the questionnaire a great care is needed, specially when measuring non tangible variables such us interviewee’s motivations, perceptions or attitudes. Finally, the non response rate should be taken into account, because it may also cause biases in the sample composition. As for the conceptual aspects, we have adopted a view of entrepreneurship as a process, in line with the most recent trends in this field. This approach considers the entrepreneurial activity as a result of the combination of personal (individual's perceptions and attitudes, which are in turn affected by the person’s background) and environmental factors (not only economic, but also family and social environments). Therefore, it is probable that a great diversity of types of entrepreneurs exists, with different motivations and objectives. Factors that lead to enterprise creation may vary for each social group. Nevertheless, to the extent that environmental circumstances are similar, and that certain shared sociocultural features exist, it may be expected that certain variables influence the whole group of nascent entrepreneurs. In this paper, we have worked with a very wide data panel including the whole German population, which has allowed to avoid many of the methodological problems indicated. Nevertheless, our database did not include questions about perceptions, attitudes and motivations of the entrepreneur, which has undoubtedly been an important limitation. The results obtained are, considered as a whole, quite satisfactory. In the first place, they have confirmed a series of causal relationships that had been already found in previous studies, and have also identified other relationships. Secondly, we have obtained a reasonable predictive capacity, since our model correctly classifies 89.4% of the sample, and in particular, it classifies 56.5% of nascent entrepreneurs correctly. Based on these results, we can say that the following personal factors positively affect the decision to create an enterprise: a higher income level, higher educational level, lower age, not being the household reference person's son (those still living with their parents are less likely to become entrepreneurs), and not being a foreigner. Regarding professional features, the positive effect is derived from either being unemployed, or -if workingdoing so in small companies, as self-employed or with a high degree of autonomy, and also doing it in the hostelry and trade sector. 18 In future, however, it is necessary to advance in the design of new research to allow a more precise identification of the factors causing the decision to create an enterprise within the different social groups. In the first place, it is our intention to divide this sample in more homogeneous subgroups, since its size allows it. Repeating the analysis we have carried out here to each one of those subgroups could allow identifying the specific factors that affect enterprise creation in different social groups. Secondly, another line of research we intend to continue would consist on the comparison of these results with those corresponding to other countries, both European and otherwise. To do so, we will use as far as possible the information available in the Panel Comparability Project. Lastly, we stress again the need to have much more specific information relative to the personal and psychological features of individuals, to be able to validate the different theories about enterprise creation and entrepreneurship. It is more and more urgent to build a sufficiently wide database, both in sample size and in time period covered, and that it is specifically intended for the measurement of entrepreneurship. References AYDALOT, P. (1988): “The role of small and medium-sized enterprises in regional development: conclusions drawn from recent surveys”; in Giaoutzi, Nijkamp and Storey (eds.): Small and medium-sized enterprises and regional development. Routledge. Londres. BIRLEY, S. and WESTHEAD, P. (1994): “A taxonomy of business start-up reasons and their impact on firm growth and size”. Journal of Business Venturing, vol. 9, num. 1, pp. 7- 31. BLANCHFLOWER, D.G. and MEYER, D.B. (1994): “A longitudinal analysis of the young self-employed in Australia and United States”. Small Business Economics, vol. 6, pp. 1- 19. BYGRAVE, W.D. and HOFER, C.W. (1991): “Theorizing about entrepreneurship”. Entrepreneurship Theory and Practice, vol. 16, num. 2, pp. 13-22. CÁCERES, R. (1999): Función empresarial y emergencia empresarial. Un análisis empírico sobre los factores económicos y socio-institucionales que explican la aparición de empresarios en Andalucía Occidental. Doctoral Dissertation. Departamento de Economía Aplicada I. Universidad de Sevilla. Sevilla. CARTER, N.M.; GARTNER, W.B. and REYNOLDS, P.D. (1996): “Exploring start-up event sequeces”. Journal of Business Venturing, vol. 11, num. 3, pp. 151-166. CHANDLER, G.N. (1996): “Business similarity as a moderator of the relationship between preownership experience and venture performance”. Entrepreneurship Theory and Practice, vol. 20, num. 3, pp- 51-65. COHEN, W.M. and KLEPPER, S. (1996): “A reprise of size and R & D”. The Economic Journal, vol. 106 (July), pp. 925-951. COOPER, A.C. (1993): “Challenges in predicting new firm performance”. Journal of Business Venturing, vol. 8, num. 3, pp. 241-253. 19 DAVIDSSON, P. (1991): “Continued entrepreneurship: ability, need, and opportunity as determinants of small firm growth”. Journal of Business Venturing, vol. 6, num. 6, pp. 405-429. DYER, W.G. (1994): “Toward a theory of entrepreneurial careers”. Entrepreneurship Theory and Practice, vol. 19, num. 2, pp. 7-21. EVANS, D.S. and LEIGHTON, L.S. (1989): “Some empirical aspects of entrepreneurship”. American Economic Review, vol. 79, num. 3, pp. 519-535. FIET, J.O. (2000): “The theoretical side of teaching entrepreneurship”. Journal of Business Venturing, vol. 16, pp. 1–24. FRESE, M.; KRING, W.; SOOSE, A. and ZEMPEL, J (1996): “Personal initiative at work: differences between East and West Germany”. Academy of Management Journal, vol. 39, num. 1, pp. 37-63. GAROFOLI, G. (1992): Endogenous development and Southern Europe. Avebury Ashgate Publishing Limited. Aldershot. GARTNER, W.B. (1985): “A conceptual framework for describing the phenomenon of new venture creation”. Academy of Management Review, vol. 10, num. 4, pp. 696-706. GARTNER, W.B. (1989): “ “Who is an entrepreneur?” is the wrong question”. Entrepreneurship Theory and Practice, vol. 13, num. 4, pp. 47-68. GARTNER, W.B. (1990): “What are we talking about when we talk about entrepreneurship?”. Journal of Business Venturing, vol. 5, num. 1, pp. 15-28. HARVEY, M. and EVANS, R. (1995): “Strategic windows in the entrepreneurial process”. Journal of Business Venturing, vol. 10, num. 5, pp. 331-347. KRUEGER, N.F. (1993): “The impact of prior entrepreneurial exposure on perceptions of new venture feasibility and desirability”. Entrepreneurship Theory and Practice, vol. 5, pp. 5-21 KRUEGER, N.F. and BRAZEAL, D.V. (1994): “Entrepreneurial potential and potential entrepreneurs”. Entrepreneurship Theory and Practice. Spring. KRUEGER, N.F. and CASRUD, A.L. (1993): “Entrepreneurial intentions: applying the theory of planned behaviour”. Entrepreneurship and Regional Development, vol. 5, pp. 315- 330. LIÑÁN, F. (2001): Hacia una nueva política de fomento empresarial: al apoyo a la esfera impulsora del empresario. Dpto. Economía Aplicada I, Universidad de Sevilla. Sevilla. MAILLAT, D. (1988): “The role of innovative small and medium-sized enterprises and the revival of traditionally industrial regions”; in Giaoutzi, M.; Nijkamp, P. and Storey, D.J. (Eds): Small and medium-sized enterprises and regional development. Routledge. Londres. McCLELLAND, D. (1961): The achieving society. The Free Press. London. REYNOLDS, P.D. (1997): “Who start new firms? – Preliminary explorations of firms-in- gestation”. Small Business Economics, vol. 9, num. 5, pp. 449-462. REYNOLDS, P.D. (2000): “Netherland nascents: entrepreneurship in northern Europe”; in EIM: Entrepreneurship in the Netherlands. Opportunities and threats to nascent entrepreneurship. Ministry for Economic Affairs / EIM. Zotermeer. SANTOS, F.J. (1998): La Teoría de la Función Empresarial: una aproximación cualitativa al empresario sevillano. Tesis Doctoral. Departamento de Economía Aplicada I. Universidad de Sevilla. Sevilla. SANTOS, F.J. and LIÑÁN, F. (2001): “Hacia una metodología empírica para la medición del empresario de calidad. Aplicación al empresario sevillano”. XXVII Reunión de Estudios Regionales. Asociación Española de Ciencia Regional. Madrid, 28-30 november. SCHWALBACH, J. (1994): “Small business dynamics in Europe”. Small Business Economics, vol.6, págs. 21-25. STOREY, D.J. (1994): Understanding the small business sector. Routledge. London. VAN PRAAG, C.M. and VAN OPHEM, H. (1995): “Determinants of willingness and opportunity to Start as an entrepreneur”. Kyklos, vol. 48, num. 4, pp. 513-540. VÁZQUEZ BARQUERO, A. (1993): Política económica local. Pirámide. Madrid. 20 VÁZQUEZ BARQUERO, A. (1999): Desarrollo, redes e innovación. Lecciones sobre desarrollo endógeno. Pirámide. Madrid. WESTHEAD, P. and WRIGHT, M. (1998): “Novice, portfolio, and serial founders: are they different?”. Journal of Business Venturing, vol. 13, num. 3, pp. 173-204. WOLTERS, T. (2000): “Nascent entrepreneurship in the Netherlands: a glimpse behind the scenes of business start-ups”; in EIM: Entrepreneurship in the Netherlands. Opportunities and threats to nascent entrepreneurship. Ministry for Economic Affairs / EIM. Zotermeer. WOO, C.Y.; COOPER, A.C. and DUNKELBERG, W.C. (1991): “The development and interpretation of entrepreneurial typologies”. Journal of Business Venturing, vol. 13, num. 3, pp. 173-204. Appendix Marginal Efect by levels of Income 0,0000 0,0020 0,0040 0,0060 0,0080 0,0100 0,0120 0,000 2,000 4,000 6,000 8,000 10,000 Change Prob. Sample distribution by year Year No. Individuals % of sample 1984 267 8,13 1985 226 6,88 1986 238 7,25 1987 204 6,21 1988 180 5,48 1989 217 6,61 1990 313 9,53 1991 272 8,28 1992 271 8,25 1993 340 10,35 1994 347 10,56 1995 410 12,48 Total 3285 100,00 21 Simple means of the variables used in the empirical analysis Variable Mean Variable Mean EMPRESAR 0,1970 SXV 0,5209 INGMIL 2,2782 SXF 0,4791 EDAD 41,3598 NACIONAL 0,7976 ES2 0,0889 EXTRANJ 0,2024 ES3 0,6027 SECTOR1 0,0146 ES4 0,0630 SECTOR3 0,1963 ES567 0,2454 SECTOR5 0,0548 WORK1 0,0049 SECTOR6 0,1093 WORK2 0,1081 SECTOR7 0,0362 WORK3 0,0113 SECTOR8 0,0542 WORK4 0,2350 SECTOR9 0,1504 WORK5 0,2304 SECTO240 0,0338 WORK6 0,0277 NOMINUSV 0,9336 WORK7 0,0387 MINUSGRA 0,0664 NUM0 0,9455 EDUC1 0,0505 NUM12 0,0545 EDUC2 0,2021 SIZE1 0,2432 EDUC3 0,6262 SIZE2 0,1416 EDUC4 0,1212 SIZE3 0,1306 RELAC1 0,5111 SIZE4 0,1406 RELAC2 0,3546 PERIODO1 0,5008 RELAC3 0,1236 PERIODO2 0,4992 RELAC4 0,0107 C2 0,3817 C1 0,6183