Introduction

The purpose of this paper is to analyze the progress made by academia in the female entrepreneurshipFootnote 1 domain in the last 14 years, adopting a thematic approach that combines a managerial and a sociological perspective. To this end, a systematic literature review (SLR) of 248 papers has been conducted.

The rationale behind this research interest lies on the awareness that research on female entrepreneurship has expanded exponentially in the last years and thus the time has come to evaluate its progress and to reflect on its future directions in order to gain deeper insights into the topic. For the sake of clarity, other reviews on the topic, both systematic and narrative, already exist (for a comprehensive list see Jennings and Brush 2013). However, this work is different from previous analyses for at least three reasons. Firstly, it adopts a thematic approachFootnote 2 for the analysis of the selected 248 papers, thus contributing to fill a gap in the extant literature. The other existing SLRs use indeed different lenses to read the phenomenon (see Appendix) and those (narrative) reviews that are based on the same approach mainly consider papers published before 2002 (see Carter and Marlow 2006). Secondly, this study adopts both a managerial and sociological perspective in the analysis of the advancements in the academic literature in the last 14 years, thus responding to the numerous and recent calls for a more interdisciplinary approach (e.g. Brush and Cooper 2012; Jennings and Brush 2013). Thirdly, compared to previous SLRs, this work adopts a more inclusive search criterion, as the research is not limited to the selection of any specific journals, thus contributing to depict a more comprehensive picture of the female entrepreneurship phenomenon.

Stemming from these considerations, we examined the selected 248 papers on female entrepreneurship throughout a coding frame and we inductively synthesized and categorized them into five major themes, namely Entrepreneurial characteristics, Financing, Management & Strategy, Performance and Connections developing-developed countries. A detailed analysis of each of these allows us to state that research on the topic has definitely progressed in the last 14 years as new insights not, or only accidently, analyzed in the ’80s and ’90s (e.g. the role of venture capital, the non-economic measures of performance, the connections between developed and developing countries) have increasingly appeared in the researchers’ agenda. Moreover, advancements in those topics (or even the identification of new topics) that cross the different themes, such as the link between gender and entrepreneurship and the role of family factors, have been identified and commented, thus contributing to demonstrate that this field of research has definitely moved ahead.

Our paper is primarily intended for those scholars and practitioners who want to improve their knowledge about the evolution of research on female entrepreneurship. In the next sections, we briefly summarize the prior research findings; we then describe our research methods and discuss the research results. We finally draw on the main conclusions from this work for proposing potential avenues for the research agenda within this field.

Consolidated findings

In order to answer the stated research question, a brief overview of the main findings from ’80s and ’90s studies is needed.

Prior research was primarily committed in identifying the main features of both women-owned firms and women entrepreneurs by generally contrasting female and male entrepreneurs. Aggregating the themes identified by Carter and Marlow (2006), the main conclusions can be outlined as follows.

Entrepreneurial characteristics: educational levels are often the same for men and women business owners, but women are less likely to have a formal education in business or financial issues (e.g. Hisrich and Brush 1983; Chaganti 1986; Brush 1992). Women business owners often lack prior business experience, especially in managerial positions (Hisrich and Brush 1984). Female entrepreneurs’ motivation comes mainly from “push” factors linked to survival pressures, discouraging situations in previous job, economic downturn and unemployment (Moore and Buttner 1997). In particular, as Cromie (1987) suggests, women interpret entrepreneurship as a mean to accommodate their work and child-care responsibilities simultaneously.

Financing: women tend to be more risk-averse and less self-confident than men, especially in the areas of financial decision-making and investments (Powell and Ansic 1997), and encounter credibility problems when dealing with bankers (Carter and Rosa 1998). Difficult relationships with the credit institutions can be explained by considering that women-owned firms are smaller than male-owned firms in terms of sales, assets and number of employees (Kalleberg and Leicht 1991; Coleman 1999) and that most women-owned firms are in the service and retail industries (Loscocco et al. 1991).

Management & Strategy: women are relatively conservative in terms of growth expectations and have modest plans for growth and expansion (Chaganti 1986; Cliff 1998), largely due to time constraints imposed by family responsibilities (Lee-Gosselin and Grisé 1990). Networks of female-owned businesses are weaker and more informal than men’s (Cromie and Birley 1992; Greene et al. 1999).

Performance: in evaluating their firms’ performance, women pay attention to factors such as personal fulfillment, the search for flexibility, and the desire to serve the community rather than only to economic indicators (Anna et al. 1999).

During the Nineties, the desire to look for a solid theoretical framework to interpret women’s entrepreneurship and why female-owned firms may be different compared to male counterparts has encouraged the use of feminist theories, such as liberal and social feminism (e.g. Fischer et al. 1993).

Liberal feminist approach holds that men and women are essentially equal in their rational capacity. Thus, the theoretical explanation for the observed differences between men and women are grounded in discrimination or in the existence of structural barriers such as education, employment opportunities, social networks, and mentors. Differently, social feminism holds that men and women are not essentially the same and it focuses on the unique needs, experiences, competencies, and values of women. Specifically, social feminism suggests that differences between women and men are in early and ongoing socialization processes that shape an individual’s identity, influencing his/her behavioral characteristics, and not in their biological characteristics. Thus, the perception of low self-efficacy may restrict the possibility for women to recognize business opportunities or can even lead to self-imposed barriers when women perceive that they may not have the right opportunities. Moreover, women’s negative perceptions of their own abilities to access external finance can increase the difficulties that women entrepreneurs face regarding their access to credit.

Stemming from the highlighted ’80s and ’90s results and by considering the fundamentals of the two depicted theories, the following question leads our research: What is the progress made by academia in the female entrepreneurship domain in the last 14 years? In order to answer this question as precisely as possible, we adopt the SLR method, whose criteria and main results are hereafter highlighted. In particular, the selected papers have been analyzed by specifically focusing on the advancements in terms of reframed and new research questions, and new ways to answer them, as well as on the advancements in terms of methods adopted for investigating such questions.

Methodology

As it is widely used in the management literature (Newbert 2007; Crossan and Apaydin 2010; Abatecola et al. 2013; Mari and Poggesi 2013), we adopted the SLR method as the research design most suitable to the aim of this paper. It differs from the traditional narrative reviews (Cooper 1998) by being more explicit in the selection process, by employing rigorous and reproducible methods of evaluation, and by “assisting in linking future research to the questions and concerns that have been posed by past research” (Thorpe et al. 2005, p. 258).

Based on the traditional SLR process, we established the following set of research criteria:

  1. 1.

    We chose the Business Source Premier (EBSCO), ABI/Informs and Ingenta (including Science Direct) as the research databases.

  2. 2.

    After having established that articles in the selected databases had to be written in English and published in peer reviewed journals in the time period January 2000-January 2014, we ensured their substantive relevance by requiring that the articles contained “female” or “wom*” or “gender” as keywords in their abstractFootnote 3.

  3. 3.

    Moreover, we required that articles selected in the previous phase also contained at least one of the following seven additional search keywords in their abstract: “firm*” or “enter*” or “own*” or “business*” or “corporation*” or “compan*” or “entrep*” or “venture*”. These keywords were chosen as they were particularly suitable for identifying those articles which strictly looked at the role of women as entrepreneurs.

  4. 4.

    All the authors further scanned the articles selected in phase 3 by reading all their abstracts to ensure substantive context. Thus, we verified the articles’ connection with our research topic. Rigorous inclusion and exclusion criteria were defined. Regarding the inclusion criteria, we decided to include in our research theoretical, qualitative and quantitative papers published in management and sociology journals. Regarding the exclusion criteria, we decided not to consider books, chapters in books, books reviews, practitioner papers, conference proceedings, working papers, reports, and other unpublished works; moreover, papers dealing with future entrepreneurs, students’ attitudes towards entrepreneurship and women employees were also excluded.

  5. 5.

    All the authors finally ensured the articles’ substantive relevance by reading the full text of all the papers selected through the previous screening of the abstracts. In particular, the articles have been reviewed according to two quality assessment criteria, theoretical robustness and methodology/data robustness, to which we assigned the following scores. 1: when articles show a poor awareness of the existing literature and/or poor research design. 2: when articles show a basic awareness of the existing literature and/or whose research design could be improved. 3: when articles show a deep and broad knowledge of the relevant literature and/or a sophisticated research design. By summing up the score for both the theoretical robustness and methodology/data robustness, those articles scoring less than 3 or equal to 3 have been excluded from the dataset.

  6. 6.

    We adopted the snowballing technique (i.e. searching in the reference lists of the selected papers to add other coherent papers) for consolidating the research outputs.

Table 1 shows the results from our search. In particular, this Table outlines the outputs from phases 2 to 6.

Table 1 Summary of the results
Full size table

In order to analyze the 248 papers, a two steps analysis has been conducted.

Firstly, by manually sorting three papers’ dimensions, namely: a) the individual paper’s keywords b) the paper’s research question(s) and, only for the empirical papers, c) the dependent variable(s), we have identified codes for each paper. Secondly, we have regrouped coherent codes into five themes (Table 2) and consequently elaborated a codebook, where for each theme its name, the list of associated codes and some examples are listed. To ensure the inter-rater reliability (Boyatzis 1998), the articles have been allocated independently by each author into the identified themes, according to a detailed examination of the article’s full-text and to the codebook; any disagreement in the categorization was resolved by jointly reviewing the doubtful articles until a consensus was reached. Consequently, some minor changes were made to integrate these disagreements.

Table 2 The research themes
Full size table

Following previous works on female entrepreneurship, themes 1 to 4 refer to “traditional” issues in female entrepreneurship research (Carter and Marlow 2006) “that are always ‘in fashion’ and continue to remain in the research agenda” (de Bruin et al. 2007, p. 325). For these reasons, we have labelled them “streams of research”. Conversely, the theme “Connections developing-developed countries” has been labelled “new research pattern” as papers clustered here deal with new and emerging topics on which knowledge has not yet been consolidated.

Results

The selected papers, published either in management (67 %) or sociology journals (33 %), have been carefully analyzed in order to investigate the progress made by academia into the female entrepreneurship research domain.

Stemming from the awareness that the context (be it institutional, legal, political, cultural or religious) impacts on entrepreneurship (e.g. Welter 2011) and that previous reviews on the topic have stressed a general lack of attention to it, in discussing the results we have decided to take a step further by explicitly considering the country in which the analysis is conducted (De Vita et al. 2014). We have distinguished between developed and developing countries by using the categorization by region provided by the World Bank which defines as developing countries the low- and middle-income economies (World Bank 2014). Consequently, the selected papers, already labelled as belonging to a specific stream of research or to the new research pattern, have been further allocated and discussed according to the investigated country(ies). In particular, 147 papers deal with developed countries, 82 papers deal with developing countries, while 19 papers both compare developing and developed countries and investigate female entrepreneurs from developing countries that are now established in developed economies. Interestingly, we have found evidences for each of the four identified streams of research in either studies on developed or developing countries, although with different levels of depth and sophistication. The new research pattern “Connections developing-developed countries” emerged, instead, as a joining link between developed and developing countries (Table 3).

Table 3 Streams of research and new research patterns according to the contexta
Full size table

Table 3 can be read either vertically or horizontally. This section explores its vertical reading, allowing us to deepen the most relevant advancements in each stream of research and in the new research pattern. The complementary horizontal reading of the Table is specifically analyzed in the “Discussion and conclusions” section, allowing us to identify and discuss those (new or renewed) topics that cross the different streams of research.

Table 4 presents a sample of evidences collected from the SLR.

Table 4 A sample of the SLR evidences
Full size table

Entrepreneurial characteristics

In this stream of research, we have grouped those papers that deal with women entrepreneurs’ motivations and personality.

Developed countries

As far as women’s motivations to undertake an entrepreneurial path are concerned, in line with ’80s and ’90s results, the traditional push/pull dichotomy (Mallon and Cohen 2001) still represents the main theoretical framework for investigating such a topic. However, the most current results show that some important steps have now been taken. Several scholars have, indeed, strongly stressed that the traditional push/pull categories “ignore the complexities of women’s working and domestic lives” (Patterson and Mavin 2009, p. 173) and that they rarely exclude each other (e.g. Kirkwood 2009). Moreover, as still no clear consensus has emerged on which factor exerts the greatest influence, scholars are advocating the need for new methodologies to investigate the reasons why women become entrepreneurs and for new response categories as the traditional surveys and narrow categories’ definitions can create interpretation bias (Hughes 2003). In this vein, the most intriguing results come from those recent papers that use new approaches, such as the life history approach, to investigate the subject, mostly implicitly, adhering to the social feminism or even to the post-structural feminism (e.g. Patterson and Mavin 2009). Interestingly Kirkwood (2009), comparing women and men entrepreneurs’ motivations, finds that a combination of push and pull factors motivates both genders, but differences in the impact of these motivations exist. In particular, the role of children influences women’s but not men’s motivations (p. 357) “as women consider (in advance of starting the business) the potential difficulties of integrating their family and the business” (see also DeMartino and Barbato 2003). Accordingly, in recent years more emphasis has been placed on the impact of family responsibilities (e.g. McGowan et al. 2012) as push factors, clearly showing that entrepreneurship cannot be considered a panacea for balancing work and family role responsibilities (Parasuraman and Simmers 2001; Kirkwood and Tootell 2008; Patterson and Mavin 2009; Poggesi et al. 2015) and this seems true also for mature women (Moult and Anderson 2005).

Regarding women entrepreneurs’ personality, over the years, several variables (i.e. personality traits), such as need for achievement, locus of control, self-efficacy, risk aversion, etc., have been investigated to understand whether female and male entrepreneurs can really be considered to be different but still results are not convergent. It is interesting to point out a recent perspective according to which behavioral differences between women and men entrepreneurs are sometimes minor, if compared to differences among women themselves who perform different kinds of entrepreneurial activities (Malach-Pines and Schwartz 2008), thus advocating for a more heterogeneous approach in investigating women entrepreneurs’ behavior.

Developing countries

Studies in developing countries adopt the traditional push/pull framework, trying to contextualize it according to the country-based specificities they investigate. Also in these countries a consensus on which push and pull factors exert the greatest influence does not emerge. Many studies, indeed, show that economic necessity is the primary motivation that pushes women to start up a business (e.g. Eversole 2004; Holmén et al. 2011) and other push factors are identified, for example, in the need to assist their families and in the strong gender inequity (e.g. Chu 2000; Terjesen and Amorós 2010). However, Gray and Finley-Hervey (2005) show that desire for achievement and for independence are the main motivations for Moroccan women to become entrepreneurs; similarly, Naser et al. (2009) show that women entrepreneurs in the United Arab Emirates are motivated by self-achievement, self-fulfilment and the need for being financially independent. In line with these results, the study by Welter and Smallbone (2008) on women entrepreneurs in Uzbekistan is particularly relevant as it stresses that “reported motives relating to factors such as’independence’, ‘autonomy’, and ‘self-fulfilment’ must be interpreted in the context of the environment and experience where such terms are used” (pp. 518–519). Conflicting situations between the role of entrepreneur and other roles traditionally ascribed to women – such as wife, mother and housewife – are also investigated here. Results are particularly worth noting due to the patriarchal model widespread in these countries (e.g. Ufuk and Özgen 2001). Accordingly, Rehman and Roomi (2012) claim that the concept of work-life balance is no longer a western phenomenon.

Regarding studies on women entrepreneurs’ personality, they are very scant and not as deep as those conducted in developed economies are. In South Africa, for example, Mahadea (2001) finds that only marginal differences exist between men and women entrepreneurs in autonomy propensity, need for achievement, risk-taking propensity and locus of control. Interestingly, Javadian and Singh (2012) claim that the fear of failure and the lack of self-confidence are “non-factors” for Iranian women due to the challenges they have to face daily in their lives.

Financing

Developed countries

As in the ’80s and ’90s studies, most of the papers clustered here contrast women against men entrepreneurs by statistically analyzing their main similarities and differences in the relationship with credit institutions, either from the demand side or from the supply side perspective, thus, implicitly, adhering to liberal feminism. However, with respect to previous research, interesting new insights emerge.

Regarding the demand side perspective of the relationship, scholars are still largely committed to verifying and testing the existence of real discriminatory behavior engaged in by lenders against women entrepreneurs (e.g. Brana 2013). Results converge that this discriminatory behavior does not exist because of either legal restrictions or a greater awareness of the importance of female firms for banks. Thus, to date, the gender-based differences in debt financing are explained in two interrelated ways. On the one hand, scholars agree on the existence of a kind of “self-discriminatory” behavior carried out by women themselves (e.g. Kon and Storey 2003; Orser et al. 2006; Wilson et al. 2007). On the other hand, scholars agree on the need to take into consideration the business characteristics (e.g. industry sector, firm size, and age of the firm) in order to control for systematic differences between male and female firms. By adopting this approach, mixed results emerge: some scholars indeed find no evidence of credit institutions’ discrimination against women entrepreneurs (e.g. Coleman 2002; Arenius and Autio 2006), while others assert the contrary (e.g. Muravyev et al. 2009). In this debate, worth noting is the study by Wu and Chua (2012) that takes into account as a relevant variable not only the gender of the entrepreneur, but also the “gendered structure” of the firm, which they call the “second order” gender effect. According to them, a significant effect on the relationship between female firms and credit institutions would not be generated by the gender of the borrower but is more likely to be associated with specific organizational characteristics. The authors, confirming Marlow and Patton’s (2005) results, find out indeed that, when the “gender” of the organization is unambiguously female, women entrepreneurs have to face higher borrowing costs.

When looking at the demand side perspective of the relationship between credit institutions and entrepreneurs, scholars generally confirm the ’80s and ’90s results. Differently from men entrepreneurs, women still show a negative perception of banks as sources of finance (e.g. Hill et al. 2006; Roper and Scott 2009) mainly because of their past negative experiences in seeking funding, which discourage them from seeking further external finance.

Besides these “classical” topics, new insights have started to emerge. Indeed, researchers have taken a step further in the analysis of the interaction between banks and firms, by studying the personality, experiences, beliefs and perceptions of loan officers. Specifically, these studies consider the sex of the loan officers as an aspect that could influence either the criteria or the processes used to assess the loan application of male and female entrepreneurs, thus influencing the entire process of negotiation. Three recent papers based on UK and Italian data examine this topic; gender-based differences in the relative importance given to the assessment criteria for approving loan applications emerges in the UK (Carter et al. 2007), although female bank loan officers are as likely as their male counterparts to draw gender distinctions between business owners (Wilson et al. 2007). In Italy, female loan officers appear to be more risk-averse or less self-confident than male loan officers are since they tend to restrict credit availability to new, unestablished borrowers more often than their male counterparts (Bellucci et al. 2010).

A growing interest is also emerging in the relationship between venture capitalists/business angels and female entrepreneurs (e.g. Carter et al. 2003; Harrison and Mason 2007). Researchers are now debating a new approach to investigate this relationship, that of homophily which means that entrepreneurs prefer to seek funding from investors of the same sex. In the venture capital industry, the concept of homophily has been introduced by Brush et al. (2002), who state that the low level of venture capital funding in female firms is in part ascribable to the relatively small number of women employed in the venture capital industry. Accordingly, in the angel capital market, Becker-Blease and Sohl (2007, 2011) by using a liberal feminist perspective show that the difference in seeking angel funding rates appears, at least in part, to be driven by the small number of women business angels.

Developing countries

Financing issues have also been deepened in those studies dealing with female entrepreneurs in developing countries. As far as the relationship between women entrepreneurs and credit institutions from the supply-side perspective is concerned, results mainly show how strong is the relevance of the socio-economic context in limiting women’s possibilities to access to financial resources (e.g. Kevane and Wydick 2001; Aidis et al. 2007; Belwal et al. 2012).

When looking at the demand side perspective of such relationship, the results are in line with those of developed countries as a general reluctance of women in asking for capital to credit institutions is tested (e.g. Fletschner and Carter 2008; Fatoki and Garwe 2010). However, the main reasons for such behavior are identified here in the social norms, as well as formal and informal prescriptions that, for example, prohibit or discourage women from interacting with men other than their own relatives (e.g. Guérin 2006; Fletschner and Carter 2008), in countries where loan officers are predominantly male.

Therefore, in these countries, the entrepreneurial path is often an unsustainable challenge for women who are obliged to establish low-intensity, low-returns, and frequently close-to-home firms whose contribution to their income, however, is really quite modest. What scholars highlight is on the one hand the need to foster microcredit programmes, so that it would be possible to raise the women’s income-earning potential and empower them socially and economically (Botha et al. 2006; Afrin et al. 2010; Nyamwanza et al. 2012). On the other hand, they encourage women’s engagement in higher-intensity and higher-returns activities (Guérin 2006; Cetindamar et al. 2012).

Management & strategy

In this stream of research, we have grouped those papers that deal with the way in which women manage their firm, the strategy they undertake and their networking practices.

Managerial practices

Developed countries

In their theoretical work, Bird and Brush (2002) invite researchers “to look for and measure the feminine/personal and traditional/masculine dimensions of individuals and process in order to identify ways that men and women behave similarly and differently” (p. 58); however, still few empirical papers have analyzed how women manage their firms. An explanation for this gap can be identified in the sophisticated and complex methodologies needed to unveil this topic, methodologies that require analyzing women entrepreneurs within their context, by specifically focusing on their interactions, tacit processes, and often hidden beliefs and values. Accordingly, the papers clustered here are mainly based on interviews, with the most current paper taking a step further as it adopts an ethnographic research approach and explicitly considers gender as being socially constructed rather than being a fixed and universal category (Eriksson et al. 2008).

Turning to the results, evidences are mixed. Some scholars, for example, state that women as entrepreneurs tend to use a less formal and more relational approach, especially with employees (e.g. Farr-Wharton and Brunetto 2009). On the other hand, scholars such as Cliff et al. (2005) assert that “the effect of owner sex on organizational characteristics and managerial practices is more of a myth than a reality” (p. 87) and that entrepreneurs themselves perpetuate the belief that men and women organize and manage their firms in different ways. By adopting a different perspective, Eriksson et al. (2008) test that the way in which women build cooperative relationships is not gendered in a unified manner and that the contexts count.

Developing countries

Topics related to managerial practices are not particularly deep in those theoretical and empirical studies that deal with female entrepreneurs in developing countries to the point that no papers fit within this category.

Strategies

Developed countries

Despite the relevance of the topic, research in this area is still in its infancy, thus confirming the gap already identified by Brush in 1992. Two main reasons can be identified for such a low number of publications: 1) in order to analyze strategies, longitudinal analyses are needed and access to this type of data is particularly difficult, 2) strategy is less developed in smaller (as female firms generally are) than in larger businesses due to the entrepreneurs’ lack of time and money (e.g. Verheul et al. 2002). However, due to the relevance of female firms, in terms of employment and GDP, the topic can no longer be neglected and, indeed, in the last 14 years scholars seem particularly interested in investigating the female firms’ internationalization process and their entrepreneurial growth.

As far as the internationalization process is concerned, two papers are worth noting, those by Welch et al. (2008) and Orser et al. (2010), as they take a step further in linking this process with feminist arguments. Welch et al. (2008) adhere in their research to “gender as social identity” and do not consider “gender as variable”, as most scholars instead still do, not only in this stream of research. Their results show that exporting is not just a strategy for the firm but also a life-changing experience for the women entrepreneurs and that there is the need to develop “constructs” (such as the characteristics of the ideal exporter) that are less “masculine” in their definition. Orser et al. (2010) show that both liberal and social feminism fail to explain the differences in export propensity between female and male entrepreneurs leading scholars to ask for the adoption of a feminist entrepreneurship theory in the investigation of female entrepreneurship.

The growth process is a more investigated topic; scholars are interested – by mainly using statistical analyses – in understanding which factors influence male and female entrepreneurs in their decisions to pursue business growth and why female entrepreneurs are less likely to exhibit strategic growth orientation than their male counterparts. Results are not convergent: Orser and Hogarth-Scott (2002), for example, test that male and female owners exhibit strong similarities in how they come to the growth decisions, but women are discouraged by the growth-related stress associated with personal demands made on their time and family. Alsos et al. (2006) indicate that gender makes an important difference in raising the amount of loan and equity capital needed to develop the business with a consequent lower growth of female firms than male firms. Differently, Morris et al. (2006) show that growth orientation is associated with whether a woman is “pulled” or “pushed” into entrepreneurship, the former being more growth-oriented than the latter.

In this stream of research, a first attempt to move the research towards a more dynamic understanding is proposed by Davis and Shaver (2012), using the life course theory. These scholars, investigating how the relationships among gender, career stage, and family status jointly influence growth aspirations over the life course, test that women and men differ little in their overall intent to pursue high-growth entrepreneurship. However, a different set of factors pull men and women into pursuing high-growth entrepreneurship and, surprisingly, a positive relationship between parenthood and women’s growth intentions, certainly deserving further attention, has been verified.

Developing countries

In line with what has been studied in developed countries, growth intention is the main topic analyzed also in the works clustered here. In particular, these papers are mainly focused on the reasons why women show lower growth intentions than men and the main reason is often identified in the high formal and informal barriers women face (disadvantageous tax laws, difficulties in access to finance, social legitimation, etc.). For example, in contexts such as those of Nigeria or Zimbabwe, women often perceive the environment as “hostile” (Mboko and Smith-Hunter 2009), thus negatively impacting on their decision to pursue growth.

Another reason is suggested by Yordanova (2011), who states that, in the case of Bulgaria, not only the perceived environmental hostility, but also the tendencies of women entrepreneurs to employ mainly family members and relatives have a detrimental impact on their growth intentions. These results are in line with another study conducted in Bulgaria, that by Manolova et al. (2007), testing that the support that networking generates results to be positively and significantly associated with entrepreneurs’ growth expectancies for men but not for women.

Networking

Developed countries

Despite the evidence that networks are important to business success and the awareness that networks are based on social interactions and, thus, gender “lenses” are relevant (Hanson and Blake 2009), few empirical studies exist on women entrepreneurs’ networks. However, these papers are worth noting as they take a step further in the female entrepreneurship research by adopting a more dynamic way of understanding network development.

Several scholars indeed investigate the dynamics of female entrepreneurial networks at the different stages of the business life cycle. Roomi (2009), for example, shows that women entrepreneurs can more successfully use their social capital in the survival, success and take-off stages of their businesses if they start by building it into the start-up or even in the pre-start-up phase. Greve and Salaff (2003) explore how each phase of establishing a business requires a different emphasis on networking. Hampton et al. (2009) finally suggest that enhanced quality and network benefits evolve in line with business development.

Developing countries

In less developed countries, networking is mainly perceived as a vehicle enabling women to countervail resource paucity and environmental adversity by means of personal connections. As it emerges in studies on developed countries, female networks are mainly based on relatives and friends. Differently from men, only when the ties become weaker does the women’s proportion of non-family ties increase (Rutashobya et al. 2009). In this vein, also Kuada (2009), analyzing women entrepreneurs in Ghana, finds out that, even if the family definitely is a key resource for women entrepreneurs, women work better without such strong social relations as these might obstruct their business growth.

Performance

Developed countries

Most papers clustered here statistically compare female and male firms’ performance and, although results are still mixed, female firms are more likely to be labelled as “under-performing” (Du Rietz and Henrekson 2000).

The “under-performing” hypothesis – or “myth” (Marlow and McAdam 2013) – of female firms has animated the last 14 years’ debate with important new insights to note. On the one hand, scholars have tried to test this hypothesis by considering the systematic differences between male and female entrepreneurs. Thus, by generally adopting (implicitly or explicitly) a liberal feminist approach, scholars have begun to control for the potential effect of the economic structure of the market as well as of the firm characteristics (e.g. business age, business size) but no conclusive results emerge (e.g. Collins-Dodd et al. 2004; Fairlie and Robb 2009; Lee and Marvel 2014).

On the other hand, several scholars underline, by generally explicitly adopting a social feminist approach, that the tested performance differences between men and women owned firms, and the mixed results obtained after controlling for specific business characteristics, are either the results of inappropriate performance measures or the scant (or absent) consideration of the non-economic results. Regarding the former, for example Watson (2002), Watson and Robinson (2003), Johnsen and McMahon (2005) and Robb and Watson (2012) point out the need to relate output measures to input measures (i.e. the need to use ROA, ROE) and to consider risk when making comparisons of business performance. In doing so, the previously cited scholars test no significant difference in the performance of female and male owned firms. Thus – they conclude – women are not less effective in business than men because they are discriminated against (e.g. by lenders) or because of factors that limit them (e.g. education background), but women adopt different approaches which may – or may not – be equally effective. Regarding inappropriateness of the performance measures, the current critique asserts that the “primary performance measures” (Rosa et al. 1996) are not entirely able to grasp the non-economic results of female firms, fundamental to evaluate female entrepreneurs’ success and that “concepts such as ‘success’ should be restructured into a collective vision that includes both genders” (Morris et al. 2006, p. 226). So, to date, the debate is focused on the identification and operationalization of the non-economic results that seem to be more important for women than for men, such as employee satisfaction, social contributions, goal achievement, effectiveness, and desire to balance family and work responsibilities (e.g. Brush 1992). Despite the overwhelming acceptance of this general principle, only one paper in our dataset uses non-traditional measures of success, i.e. satisfaction with status and with employee relationships (Powell and Eddleston 2013).

Beyond the underperformance hypothesis and its latest insights, other important developments have to be highlighted. Worth noting are those papers that push further into the less-explored terrain on the work-family front (Jennings and McDougald 2007). One paper in particular seems to contribute in advancing knowledge on the topic, that by Powell and Eddleston (2013). The scholars analyze the positive side of the work-family relationship in order to understand how family might also help female and male entrepreneurs in their endeavors. They test that US women entrepreneurs experience positive linkages of family-to-business enrichment and support to entrepreneurial success, whereas male entrepreneurs do not. This is explained by women’s socialization process which allows them to create a more synergistic link between family and work.

Also in this stream of research, an attempt to move research towards a more dynamic understanding of the female entrepreneurship can be identified. By using the Life Cycle Model and by considering traditional performance measures, Coleman and Kariv (2013) test for no significant performance differences between female- and male-owned firms during their early years, while significant difference in performance were tested starting from the sixth year. Explanations are found in the gender differences in terms of the impact of specific financial strategies on the amount of capital raised.

Developing countries

The interest in female firms’ performance is really strong in papers on developing countries and particularly by considering the peculiar socio-economic and cultural contexts, the need to identify a new and correct way to measure female firms’ performance seems even more important than in studies on developed countries. According to Kantor (2002), economic results can shed only partial light on the motivations of women operating in developing countries for entering into self-employment. The empowerment of State or male family members, the need to improve a family’s livelihood position, and self-fulfilment are indeed examples of important outcomes for those women. For researchers, it means analyzing the socio-cultural context in which the female entrepreneurs operate (the socialization process, the culturally imposed attitudes, the gender stereotypes, etc.) and operationalizing this information in order to understand how entrepreneurship contributes in achieving the depicted results.

Unfortunately, except for Kantor (2005) who, in her study on Indian women entrepreneurs, employs a two-dimensional definition of success, i.e. economic outcomes and empowerment outcome, scholars still fail in this effort. In particular, the majority of works in our dataset do not properly consider the social-cultural context in which female entrepreneurs operate (e.g. Singh et al. 2001; Chirwa 2008; Bardasi et al. 2011; Fuad and Bohari 2011), although some exceptions exist. Aterido and Hallward-Driemeier (2011), for example, test a gender gap in performance in six countries in Sub-Saharan Africa, pointing out that the entrepreneur’s marital status does matter. In particular, unmarried women appear less able to operate more productive firms as, according to the cultural practices in the region, daughters receive less access to assets than sons. Roomi (2013) also considers some socio-cultural variables in analyzing female firms’ growth in Pakistan and, by taking into account the Islamic traditions, “moral support of immediate family”, “independence to have meetings with opposite gender”, “independent mobility” are added to the statistical model.

Connections developing-developed countries

In this emerging new research pattern we have grouped papers that deal with: 1) female entrepreneurs that have moved from a developing to a developed country, 2) female entrepreneurship by comparing developed and developing countries.

Regarding immigrant female entrepreneurs, although ethnic or immigrant entrepreneurship is not a new topic in the literature (Light 1972; Waldinger et al. 1990), the role of gender has frequently been neglected. Often referred to as “ethnicity entrepreneurship” or “female ethnicity”, immigrant women entrepreneurship refers to the impact that either “ethnic diversity” or “gender” may have on entrepreneurship, thus running to the concept of intersectionality (Crenshaw 1995). The relevance of the topic is undeniable: in light of the ongoing globalization process, the demographics of many countries have indeed changed dramatically and are still evolving. The number of ethnic women that now live in a new country, usually following their husbands and fathers, is rising, even though the possibility of their entering the labour market is not. Entrepreneurship can be for them the only way to work and consequently to earn money and it can be a possible endeavor thanks to their access to resources by means of close ties and family relations. However, as immigrants, these women entrepreneurs may experience specific forms of discrimination that native women do not encounter, leading to a “double discrimination”, for the gender and for the ethnicity. The analysis of the way in which immigrant women “do” entrepreneurship exactly represents the newness and relevance of the papers clustered here as, in conducting this analysis, they contribute to broaden the debate on women entrepreneurship.

In looking at the papers’ content, both the analyzed countries – Japan (Billore et al. 2010), The Netherland (Essers and Benschop 2007, 2009; Essers et al. 2010), Israel (Heilbrunn and Abu-Asbah 2011), USA (McQuaid et al. 2010; Wang 2010), New Zealand (Pio 2006, 2007), Australia (Collins and Low 2010), UK (Strüder 2003) – and the ethnicity of surveyed immigrants – i.e. Turkish, Moroccan, Indian, and Arab women – are heterogeneous. However, the results are unanimous: when settled in the host country, what leads immigrant women to start their own activity is the difficulty of finding a job and, thus, the need to overcome the traditional barriers they encounter, for example religious, cultural, as well as financial and institutional ones (i.e. Heilbrunn and Abu-Asbah 2011; Pio 2006, 2007). Generally, the firms set up by the investigated women are micro and serve primarily ethnic customers; in some cases, they help other migrants to enter the labour market (Pio 2006, 2007). Interestingly, some pioneering scholars pay particular attention to the intersections of gender, ethnicity and religion within the context of entrepreneurship (Essers and Benschop 2007, 2009; Essers et al. 2010). Although these papers represent the first attempts to analyze the topic, they clearly show that investigating women entrepreneurs according exclusively to their being, or not, natives of the hosting country is not helpful in capturing the overall experience of immigrant women entrepreneurs (Essers et al. 2010).

As far as papers that compare female entrepreneurship in developing and developed countries are concerned, their relevance and newness have to be underlined. Interestingly, all of these few papers try to rule out countries’ differences by considering institutional/economic factors institutional environment and macroeconomic characteristics – such as regulations, GDP, maternity leave coverage. Minniti (2010), for example, finds that a significant portion of the gender gap in start-up activity is associated with differences in self-perception (i.e. having the necessary skills and knowledge to be firm founders) and with the per capita GDP of the country. Although conclusive results do not emerge, also because each paper differently accounts for economic, technological, demographic, institutional, personal and cultural variables, they converge on the fact that entrepreneurship is a social activity hampered by gender-based constraints.

Discussion and conclusions

The main aim of this paper was to review the literature on female entrepreneurship published in the last 14 years in order to identify the progress made by academia. To address this objective, we conducted a SLR of 248 papers published between 2000 and 2014 in both management and sociology journals. The selected papers have been categorized into four different streams of research and into a new research pattern, then further distinguished according to the socio-economic context to which they refer. Results clearly show that the female entrepreneurship research domain has progressed (Table 5).

Table 5 Main advancements in the female entrepreneurship research domain
Full size table

Table 5 can be read either vertically or horizontally. Its vertical reading allows us to identify and deepen the most relevant advancements in each stream of research, distinguishing between developed and developing countries, and in the new research pattern. Its horizontal reading, instead, allows us to develop an integrative perspective in analyzing this literature review’s findings, highlighting a number of (new or renewed) topics that cross the different streams of research and new methodologies to answer new or renewed research questions.

Regarding the vertical reading of Table 5, as the main results have previously been deeply highlighted, our aim in this section is to identify a number of possible future research directions. By considering the studies on developed countries it could be worth: 1) adopting a more dynamic look to investigate if and how women entrepreneurs’ motivation factors change over the life cycle of the firm, and in particular if a “conversion” from push to pull factors may happen and with what consequences; 2) expanding the “second order” gender effects’ analysis over the financial transactions; 3) strongly considering the spatial segmentation patterns of women entrepreneurs’ networks; 4) really shifting the women entrepreneurs’ performance evaluation operationalizing socio-emotional factors, such as satisfying relationships with employees, the involvement of the family into business, etc. Moreover, although some attempts to expand the research beyond the “boundaries” of each stream of research have been realized (above all in the case of the “Performance” stream of research, which is studied in relation to strategy, networking and financing), more efforts are needed in order to really understand how women manage their firms. Our analysis also shows that the theoretical approaches used in the majority of the selected papers are grounded on reviews of previous works; thus, the adoption of emerging theories or theories from other research domains (e.g. life cycle approach) is urgent in order to catch female entrepreneurship peculiarities not yet sufficiently explored.

Considering those studies that deal with female entrepreneurs in developing countries, interestingly, no relevant advancements are highlighted in Table 5. Indeed, these studies represent an advancement on their own, as they have been able to unveil the female entrepreneurship phenomenon in countries that are mostly unexplored, providing a key contribution to fill the existing geographical bias of much female entrepreneurship research towards the western and more developed economies. However, the majority of these papers are descriptive in nature, based on very small samples and generally aimed at profiling the main characteristics of women entrepreneurs, thus mainly adopting traditional approaches in the analysis.Footnote 4 This can be, at least partially, justified by the overall difficulties that scholars have to face when doing research; indeed, according to Brush and Cooper (2012) and Henry et al. (2015), they have few government listings of women, minimal resources to carry out the research and data gathering difficulties. By considering these constraints, a possible future development can be identified in mirroring the research evolution occurring in developed countries, thus shifting towards an in-depth investigation of specific topics, supported by the adoption of more focused and sophisticated methodological techniques, still rarely employed to date.

Regarding the new research pattern, stemming from the awareness that the literature on immigrant or ethnic entrepreneurship often ignores gender and that immigrant female entrepreneurs are important for the hosting country’s economy, more theoretical and empirical effort on such topics is urgently needed. At least two future research avenues can be identified. First, in order to understand this “type” of entrepreneurship, future studies could better employ the concept of intersectionality, which allows an in depth understanding of how gender dynamically shapes ethnicity and vice versa. Second, stemming from the awareness that female entrepreneurs are heterogeneous, it could be interesting to analyze not only those women pushed into the entrepreneurship path by necessity factors, but also the experience of those highly educated immigrant women entrepreneurs who could be motivated by pull factors.

As far as the horizontal reading of Table 5 is concerned, the main advancements in those topics that cross the different streams of research and the bases for future research avenues are hereafter highlighted.

Regarding “The link between gender and entrepreneurship”, since 2000 we are witnessing an evolution in the debate on the topic, which is now more articulated and better theoretically grounded. Therefore, the awareness that gender is a socially constructed and a learned set of behaviors deriving from being male or female and that it does not refer to simply the sex of the entrepreneur seems widespread among scholars to date in each stream of research. However, theoretical or empirical analyses are not always able to properly mirror the actual knowledge of the topic. First, the feminist agenda is often hidden as only 16 % of the papers are explicit about which perspective, if any, they actually adopt. Second, independently of the specific stream of research, in a large number of studies on developed countries (59 %) the term “gender” is still considered a measurable and independent variable and treated as a dummy variable according to the sex of the entrepreneur. In doing so, scholars persist in comparing female and male entrepreneurs, mainly using male norms to judge women’s activities, and mostly adhering to liberal feminism (Henry et al. 2015). However, in line with social feminism, the assumption of gender (and not sex) as a result of the early and ongoing socialization process clearly emerges in several papers. In particular, this perspective helps to better understand how women “do” entrepreneurship and how women implement entrepreneurial behaviors. Accordingly, the research is mainly focused on women’s experiences by means of qualitative methods (discourse analyses, ethnographic studies, narrative approaches, life histories, in-depth case studies, phenomenological approaches) (e.g. Bourne 2010), thus abandoning the contrast between women’s and men’s experiences. Although such approach contributes to better consider the context in which women work, the female subordination, resulting from the socialization process, is still evident. Also in this case, the proposed business model is indeed designed with male features, thereby setting a male standard for female entrepreneurial activity (Gunnarsson-Östling 2011). A step further, and worthy of mention as it represents an advancement in the research and a very promising future research development, is also taken, to date, by those (very few) scholars who adhere to post-structural feminism (e.g. Farr-Wharton and Brunetto 2009; Ezzedeen and Zikic 2012). According to Ahl (2006, p. 597) in post-structural feminism “gender refers to what is regarded as masculine or feminine and is independent of a person’s biological sex”, meaning that both male and female entrepreneurs may adopt a behavior that would typically be expected to be adopted by the opposite sex. Thus the compliance with fixed (male) standards could finally be given up as, according to Ahl and Marlow (2012), the feminist perspective has to be applied not only to women’s business ownership but to the field of entrepreneurship more broadly. Interestingly, from our review it clearly emerges that management and sociology journals differ on the topic. Sociology journals, and not management journals, are indeed greatly devoted to analyzing the ways in which the processes of social construction may affect (and explain) women entrepreneurs’ behavior and attitude, above all in terms of human and social capital as well as in terms of women’s motivations and work-life balance. Conversely, management scholars, although theoretically acknowledging this progress, still mainly consider gender as a mere dummy variable in their analyses that mainly focus on female firms’ performance and management and strategy. Stemming from these results, the importance of a theoretical fertilization between the two disciplines clearly arises and it could be reached by, for example, creating authors’ teams with different backgrounds.

Turning to “The family factors”, several scholars in our dataset include in their analyses such factors in a more intriguing way than in the past, thus answering Aldrich and Cliff’s (2003) call to “incorporate family considerations in conceptual models and empirical investigations” (p. 574), previously highlighted by Brush (1992). In particular, the role of family emerges in all of the identified streams of research, leading us to claim that the most current research clearly supports the family embeddedness of women’s entrepreneurial activity. The search for a better balance between work and family is indeed considered as one of the strongest motivations leading women to start and run their own business. Family may also play a pivotal role in influencing women growth-oriented strategies and this influence seems to be strictly connected to women’s family life cycle stage. Family may help women entrepreneurs by lending the financial resources necessary to set up the endeavour. Family may also impact on the female firms’ performance, leading women to think and operate with different goals from men. Two specific advancements in respect of the previous research are worth noting. First, family is no longer analyzed as only a liability for women but is always considered more as an important asset (Powell and Eddleston 2013). Second, a more dynamic outlook is used to investigate the female entrepreneurs’ experience; thus, by means of the life cycle model, some pioneer scholars are starting to shed light on how, using a life course perspective, the centrality of career, the sense of self-efficacy, and personal work values change over the years. As far as future research directions are concerned, time has come to better define the meaning of “family”. Indeed, as Aldrich and Cliff (2003) claim, the nature of families is evolving and a broader definition of what is meant by family seems appropriate to date. Moreover, it could be useful to better understand, for example, how “family crises” may impact on female firms as, for example, in the case of divorce, at present investigated only by Galbraith (2003). What could be of interest to understand is not only the impact of such crises on female firms’ performance but, above all, how women manage the crisis and the consequent impacts on the firm’s strategy; in this vein, longitudinal analyses could provide very interesting “dynamic” results. More in general, longitudinal analyses are needed as they could help in showing if and eventually how women entrepreneurs deal differently with family issues at different life cycle stages and, consequently, with different family burdens.

Regarding “The business characteristics”, it is interesting to note that, differently from the past, in almost all the identified streams of research scholars have started, in their statistical analyses, to control for variables such as the sector of activity, the size and age of the firm. However, the business characteristics are introduced in the statistical models as dummy variables with a limited debate on how such a context dimension really influences the entrepreneurship phenomenon. Moreover, the other dimensions of the context (Welter 2011) are still often ignored in research centered on developed countries or are not adequately/only accidentally considered (e.g. Perry 2002; Welsh et al. 2014)Footnote 5. Very few papers indeed compare different countries (spatial context dimension) (one of the exceptions is Lee et al. 2009), and the institutional context – meaning to consider both formal and informal institutions – is also underinvestigated. What emerges is that the context is often taken for granted. Going back to the link between gender and entrepreneurship, the widespread adoption of social feminism – and even of post structural feminism – could foster the attention on context variables, as these theories emphasize how women’s experiences are embedded in the context in which gender is performed.

Differently – and thus we turn to the last cross-streams topic: “The role of context” – research in developing countries does not take the context for granted. Without forgetting these research limits, scholars describe the socio-economic characteristics of such economies, emphasizing institutional constraints women face, both at a formal and at an informal level, the role of the family and of the network in which the women work (e.g. Jamali 2009). These first attempts to shed light on unexplored socio-economic contexts show how female entrepreneurship differs from country to country and make those (still few) analyses that compare developing and developed countries particularly worth noting. If it is true that women business owners across different countries (both developed and developing ones) share some characteristics – it is well established that they do play an important role in the society in which they operate, by contributing to employment and wealth creation; they are more interested than men in the non-economic results of their firm. More than men, they face difficulties for example in access to credit, in developing strong networks, in balancing work and life – what clearly emerges from our review is that the influence of the socio-cultural context in which women entrepreneurs are grounded cannot be ignored as it can exacerbate or reduce difficulties and liabilities, and limit or create new opportunities. The local traditions and norms, the societal legitimation to act as an entrepreneur, the “power” of religion, the social segregation, are all examples of how the context can influence women entrepreneurs’ dynamics and behaviors. Thus, a better consideration of the effect that the socio-cultural context exerts on (female) entrepreneurship will allow researchers to better understand the phenomenon answering the when, how, and why entrepreneurship happens and who becomes involved questions (Welter 2011).

We believe that the results of our analysis may offer a clear synthesis of the recent and most remarkable advancements of academic research on female entrepreneurship that could also be of interest to practitioners. Stemming from our results, policy makers should on the one hand work hard to overcome the traditional stereotypes associated with the role of women in society that influence the existing social construction of gender roles. On the other hand, they should adopt a more heterogeneous approach in the development of policies, also taking into account the specific life cycle stages of women entrepreneurs. Moreover, time has also come to develop specific policies to exploit the potential of women immigrant entrepreneurs. Interestingly, although most of the analyzed papers discuss a number of specific policy indications – for example to balance work and family life, to supply networking facilitating services etc. – the analysis of the impact of these targeted policies is still limited. Indeed, a time lag between the initiative and the effect exists and although it is difficult to isolate such an effect, an effort should be made to evaluate the efficiency of these policies and their social impact.

Although this SLR was conducted in a rigorous manner, potential limitations must be acknowledged. We limited the search process to peer-reviewed published articles written in the English language. Given the universal interest in female entrepreneurship, as shown in our review, studies published in other languages could enrich our discussion by either adding to or contradicting some of its conclusions. Moreover, the selected inclusion/exclusion criteria may have led us to identify the final sample in ways that other keywords and/or other researchers may not have. Nevertheless, our results show that much has been done to advance research on female entrepreneurship during the last 14 years. In future scholars should start from this relevant basis in order to foster those new and emerging insights that currently are ready to be developed.