Introduction

Small and medium enterprises (SMEs) are smaller, employ fewer people, and invest less money than larger corporations; nevertheless, the definition varies by industry and country (Ayyagari et al., 2007). SMEs are diverse, often resource-constrained ventures, ranging from necessity-driven, survival-oriented micro-businesses to opportunity-driven high-growth start-ups that create jobs, support livelihoods and develop human capital (Angulo-Guerrero et al., 2017; Doh & Kim, 2014; Morris et al., 2015). SMEs vary in focus, and many exist to generate a family livelihood and may not for growth (Morris et al., 2015; Shane, 2009). However, there is growing evidence that even these survival-oriented SMEs may benefit from shifting their mindset from a traditional scarcity perspective to a more entrepreneurial mindset (Morris et al., 2020). This shift often translates into improved performance (Feroz et al., 2008).

Government support policies (GSPs) aim to help SMEs overcome financial and nonfinancial constraints to improve performance (Nakku et al., 2020). GSPs help SMEs start, grow, and develop by (i) directly providing capacity-building management assistance and (ii) indirectly by sha** the economic, regulatory, and institutional environment to be more supportive of SME needs. There is increasing evidence that GSPs improve SME performance directly (** entrepreneurial orientation (EO) (Shu et al., 2019; Nakku et al., 2020).

To explain what makes a firm entrepreneurial, Lumpkin and Dess (1996, pp. 136–137) defined EO as “processes, practices, and decision-making activities that lead to new entry, as characterized by one or more of the following dimensions: innovativeness, proactiveness, risk-taking, autonomous and competitive aggressiveness”. EO is a dynamic concept (Wiklund et al., 2019) that could be applied at the individual, organizational and environmental levels (Medina Molina et al., 2022). Recent work by Wales et al. (2020, p. 2) expanded the conceptualization of EO that included “entrepreneurial top management style, organizational configuration, and new entry initiatives”.

Recognizing the importance of entrepreneurship to economic growth and development, governments have employed GSP to stimulate entrepreneurial activity (Minniti, 2008). Despite the evidence, there remains no clear understanding of what types of GSPs are effective in develo** an EO in SMEs and if their benefits are accessible across different types of SMEs. Although it is not straightforward to untangle how firm size and the country context shape the impacts of GSPs (Grimm & Paffhausen, 2015), this review provides a pattern of impact in different contexts. Moreover, there is a lack of consensus on the efficacy of GSPs in building entrepreneurial capacity in SMEs (Grimm & Paffhausen, 2015; Nakku et al., 2020). It is also apparent that ‘one size does not fit all’ regarding entrepreneurship policy, and that governments can only provide an environment conducive to the growth of productive entrepreneurship in the long term (Minniti, 2008).

While some scholars believe that there is less interest in assisting necessity-driven SMEs (Morris et al., 2015), others argue that due to climate change, poverty, and economic issues, there is a growing need for governments to focus their support on necessity-driven SMEs, given their ability to enhance livelihoods (Dung et al., 2021). GSPs can be generic or specifically target necessity- or opportunity-driven SMEs (Buffart et al., 2020; Colombelli et al., 2020).

Governments and non-governmental organizations (NGOs) increasingly leverage entrepreneurial competencies for broader economic development (Bonney et al., 2013; Dung et al., 2021). Social entrepreneurship, for instance, tackles issues that governments may be unable or unwilling to address directly (Muldoon et al., 2022). Moreover, firms’ decisions to adopt EO are influenced by their pursuit of legitimacy, which stems from the normative, regulatory, and cognitive institutional environments (Dickson & Weaver, 2008). A conducive institutional environment is thus crucial for fostering economic growth. Recognizing this, policymakers extend support to both necessity- and opportunity-driven SMEs, with the expectation that their initiatives will cultivate EO and enhance firm performance.

Several studies have investigated GSP–performance and EO–performance relationships separately. Consequently, empirical studies were summarized as literature reviews separately considering the GSP–performance relationship (Dvouletý et al., 2021; Grimm & Paffhausen, 2015; Hogendoorn et al., 2019; Kersten et al., 2017); and the EO–performance relationship (Rauch et al., 2009; Slevin & Terjesen, 2011; Wales, 2016). Our research is motivated by a central question: What is the current understanding in the entrepreneurship literature regarding how the two forms of GSPs impact EO and the performance of SMEs in different contexts? Therefore, this paper aims to clarify the relationships among GSPs, an SME’s EO, and its performance.

We first conducted a systematic literature review of published studies investigating GSPs, SMEs’ EO, and performance nexus. By doing this, we aim to contribute to the ongoing discourse as to whether governments should follow Shane (2009), who strongly advocated that more focus should be given to opportunity-driven high-growth ventures rather than subsidizing all forms of start-ups, or Morris et al. (2015), who suggested inclusive support for all forms of SMEs.

Methods

We used the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) (Liberati et al., 2009). A sco** exercise using two databases, the Web of Science and Scopus (Snyder, 2019; Zhu & Liu, 2020), was also used to identify the correct keywords and check the richness of the study’s keywords. We searched for two sets of keywords: (i) ‘entrepreneurial orientation’ OR ‘strategic orientation’; and (ii) ‘government’ OR ‘state’ OR ‘public’ OR ‘institution*’ in the title, abstract and keywords in the databases, including ABI/INFORM, ScienceDirect, Scopus, and Web of Science, in October 2023. The steps and sampling criteria used to identify relevant articles for this systematic review are provided in Fig. 1. We have used the following inclusion and exclusion criteria to select the papers.

Fig. 1
figure 1

PRISMA flowchart

Inclusion criteria

Articles were selected based on the following criteria: (i) conceptual and empirical studies published in peer-reviewed journals; (ii) that used the firm as the level of analysis or that included firm-level analysis if a multi-level approach was adopted; (iii) studies focussing on micro, small, and medium firms or SMEs (allowing for variance in SME classification); and (iv) studies that cover specific or all sectors of the economy where industry classification of focal firms was not used as a discriminator.

Exclusion criteria

Articles that did not study GSPs and EO (or dimensions thereof) were excluded. Likewise, articles that were not published in English articles or did not provide access to the full paper were also excluded.

The initial search produced 3594 articles. This was filtered for articles published in English peer-reviewed journals, resulting in 2773 articles. Duplicates were then eliminated (1066), yielding 1707. In line with this study’s main aim, an initial manual screening of article titles, abstracts, and keywords filtered out a further 1331 articles that were not relevant. The remaining 409 full-text articles were subsequently evaluated for the relevance of this review. Of these, 344 articles were excluded because of the following: (i) they did not include either government policy/support/intervention (152) or the EO concept (51) or both concepts (110); (ii) they were not at the firm level of analysis (19); and (iii) they were non-empirical studies (12). Some studies were retained by considering innovativeness as an EO proxy (Ciabuschi et al., 2020). The full-text analysis resulted in a final sample of 65 articles (details of the studies are provided in the "Appendix" section), which were then subjected to NVivo 12 and Leximancer 4.5 analyses.

The details of the 65 articles, including authors and journal, country of study, methodology, industry studied, the theory used, analysis employed, and key variables used in the study related to GSP, SMEs’ EO and performance, were recorded in NVivo 12. Descriptive analysis of the articles and thematic analyses were carried out to identify the interrelationships between the major constructs considered (GSP, EO, and SME performance).

Next, text analysis was conducted using Leximancer 4.5 software to triangulate the findings and reduce researcher bias. Leximancer is appropriate for performing a content analysis of text data as it assists the researcher without any bias in discovering first-order concepts, second-order themes, and relationships (Verreynne et al., 2013). The map was developed and re-clustered until the software generated a stable map (Fig. 8), identifying emergent themes, concepts and relationships. Leximancer extraction was further used to manually review text blocks in the sample papers identified to have a co-occurrence of emergent themes and concepts related to the relationships among GSPs and an SME’s EO and performance.

Results

Since the first study in 2008, the number of studies investigating the relationship between GSP and EO has steadily increased, reaching its peak in 2020 with 11 out of a total of 65 (Fig. 2). This indicates a growing interest in the role of GSP in SME growth and entrepreneurial development. In the first study of our sample, Dickson and Weaver (2008) investigated how the institutional environment, shaped by GSP, impacted firms’ EO. This study found that drives for the legitimacy of the institutional environment, including regulatory, normative, and cognitive dimensions, motivated EO. Therefore, it was found that the strategic choice of SMEs’ EO was impacted mainly by the country’s regulatory structures, rules, and regulations.

Fig. 2
figure 2

Number of publications by year

Geographic representation of studies favors develo** economies, with Asia the most investigated region (Fig. 3 and Fig. 4a) (41 studies), followed by Africa (9) and Europe (6). When considering countries where the studies have taken place, China and Malaysia were the most studied countries, followed by Indonesia (Fig. 3). In Africa, most studies were undertaken in Nigeria (3).

Fig. 3
figure 3

Reviewed studies per country

Fig. 4
figure 4

Percentage of studies by various analytical procedures (PLS-SEM – Partial Least Squares Structural Equation Modelling)

Thirteen studies primarily focused on manufacturers (secondary sector). Nine studies targeted service (tertiary sector) firms such as tourism healthcare, and nine concentrated on agriculture and food (primary sector) industries (Fig. 4b). The four remaining studies combined two or more sectors. A significant proportion of the studies (30) did not identify their sample firms’ sectors or included all firms regardless of sector (Fig. 4b).

Institutional theory (11) and the resource-based view (08) (Fig. 4c) were the dominant theoretical lenses used in research. Seventeen studies used various combinations of institutional theory, resource-based view, dynamic capability view, strategic choice theory, social cognitive theory, theory of planned behavior, upper echeolons theory and network theory. Other publications purported to use resource dependency, grounded, contingency, or social capital theories, whilst 21 studies did not mention any theory.

Fifty-seven studies used quantitative methodologies, four used qualitative, and four employed mixed-methods (Fig. 4d). In the EO literature, there were several calls for more qualitative research (Miller, 2011; Wales, 2016), to advance a deeper understanding of the impact of various dimensions in diverse contexts. Hence, this review also suggested conducting more qualitative, in-depth studies to understand the context, with deeper insights into how EO’s dimensions and various government policies are manifested and may be empirically captured more specifically.

Figure 4e depicts the data collection methods employed. Four studies used secondary data for their analysis, and these studies were primarily in developed countries such as Australia and Germany and a cross-cultural context. Only one study from a develo** economy, China, used secondary data. The rest of the studies that were conducted in develo** economies used surveys (52), followed by interviews (5), or a combination of data collection methods (4). Because the effect of GSPs in develo** EO and performance is lagged (Rauch et al., 2009; Shu et al., 2015; Nakku et al., 2020), more longitudinal studies are needed to precisely explain the relationships between GSPs and SMEs’ EO (Shu et al., 2015; Nakku et al., 2020). Rauch et al. (2009) suggest that more longitudinal studies are needed to precisely explain the relationships between GSPs and SMEs’ EO.

Of the studies examined, structural equation modeling (SEM), a quantitative method for analyzing several concepts simultaneously, was the most popular choice used in 30 studies. Another 22 studies used regressions. The remaining 13 studies employed qualitative data analysis methods such as thematic analysis, content analysis, or fuzzy-set qualitative comparative analysis (Fig. 4f).

Publication outlets included the International Entrepreneurship and Management Journal (06), Sustainability (05), Journal of Business Research (03), and Journal of Small Business Management (03). These high-ranking journals as publication outlets indicate the research quality of the papers. The remainder of the papers were published in a diverse range of journals. These publication outlets indicate the need for further excellence in research to be published in high-quality journals.

The conceptualization of government support policies

The studies under review conceptualized GSP differently. Prompted by the lack of consistency, we followed best practice to categorize GSPs as direct (targeted) and indirect (non-targeted) support (OECD, 2020), which are further classified as either financial or nonfinancial (Nakku et al., 2020) as depicted in Fig. 5 and discussed next.

Fig. 5
figure 5

Types of government support policy intervention to assist SMEs

  1. (i)

    Direct government support policies

    Direct GSPs are intended to offer users quick and targeted assistance, and the government has greater control over how the assistance is used. Direct financial GSPs (mostly investigated) involved grants, loans, guarantees, and risk-sharing schemes, whereas nonfinancial support included education, training, information, advice, consultation, and promotion of technology and innovation (Table 1).

    Table 1 Explanation of various direct support policies and number of studies investigated

    Before the work of Storey (2003), as operationalized by Nakku et al. (2020), most studies combined financial and nonfinancial programs into one variable when analyzing the impact of a GSP and did not consider the impact of a particular support measure on the SMEs’ level of EO. Moreover, within direct GSPs, financial support was the main form of support investigated (Ciabuschi et al., 2020; Firman et al., 2018; Hassan et al., 2019; Khattak & Shah, 2020; Nakku et al., 2020; Núñez-Pomar et al., 2020; Yusoff et al., 2021a, b).

  2. (ii)

    Indirect government support policies

    Indirect GSPs aim to indirectly promote industries or firms by establishing a conducive institutional environment and ecosystem for their development. Examples of indirect GSP mechanisms include using regulations, tax, or development policies to create a more favorable environment for SMEs (Dennis, 2011). Even though the immediate benefits of these GSPs to SMEs are sometimes difficult to identify, the findings suggest that they play a vital role in SMEs’ creation, development, and growth.

    At a broader level, the institutional environment comprises the political, economic, social, and legal properties that legalize individual and organizational activities (Busenitz et al., 2000). The past, present and future institutional environments are paramount for existing and new firms. Entrepreneurs’ attitudes and motivations toward entrepreneurship are elevated when the institutional conditions are promising (Dai & Si, 2018). SMEs strengthen their chances of survival and success by responding appropriately to the institutional environment. Governments generally implement non-targeted GSPs to improve sector-wise, state-wise or nationwide institutional conditions for SMEs.

    The entrepreneurial ecosystem comprises the “key entrepreneurs and firms that govern, integrate and perform all of the functions required for entrepreneurship to flourish in a territory” (Stam & van de Ven, 2021, p. 821). Entrepreneurial ecosystems are complex, multi-actor, multi-scale phenomena requiring tailored governmental responses (Brown & Mason, 2017). Indirect GSPs impact the entrepreneurial ecosystem, but there is no clear understanding of how and which policies are promoted effectively (Mason & Brown, 2014). The entrepreneurial ecosystem provides a comprehensive and integrated supportive environment for start-ups and scale-ups. It is essential to identify the entrepreneurial ecosystem’s elements and how GSPs influence these elements. Table 1 illustrates the different GSP interventions (financial and nonfinancial) implemented to improve the institutional environment and entrepreneurial ecosystem, as found in this review.

    Seventeen studies investigated how the institutional environment impacted EO. A number of these papers focussed on indirect GSP, investigating the impact of indirect GSPs on EO (Ajayi, 2016; Alarjani et al., 2020; Basco et al., 2020; Chege et al., 2020; Dickson & Weaver, 2008; Gomez-Haro et al., 2011; Hanle et al., 2021; Kungwansupaphan & Leihaothabam, 2019; Musa et al., 2017; Parga-Montoya & Cuevas-Vargas, 2019, 2023; Rasiah et al., 2023; Roxas & Chadee, 2013; Siahaan & Tan, 2022; Urban, 2018; Wales et al., 2021; Zhou et al., 2022). Roxas and Chadee (2013) conceptualized the institutional environment using different constructs, including business support, regulatory quality, government policies and rules of the law. Further, some scholars investigated direct and indirect GSPs as a single construct (Hutahayan, 2019; ** SMEs’ EO. Yang and Yu (2022) recently demonstrated that both formal and informal institutional support can foster greater investment in innovation. This assistance aids new ventures in gaining a better understanding of market opportunities, mitigating the risks associated with innovation activities, and enhancing their overall organizational innovation performance.

    The conceptualization of GSP is dynamic and subjective, making comparison of studies difficult. We argue that this is caused by the lack of a clear basis for classifying and typifying GSPs. Therefore, it is recommended that a consistent construct employing the many aspects of GSPs be conceptualized and designed to make studies more comparable. Figure 5 depicts the major categories of GSPs.

    In essence, some GSPs (direct) are targeted explicitly to develop entrepreneurs’ specific skills and provide resources. In contrast, others (indirect) generally impact the business environment where the businesses operate. As discussed in the next section, both direct and indirect GSPs impact the dimensions of the firm’s EO.

Relationships between GSPs, EO and SME performance

From the preceding, it is clear that GSPs “use taxpayers’ funds to, directly and indirectly, target primarily or exclusively SMEs” (Storey, 2003, p. 474). Studies have shown four different pathway relationships through which governments influence SMEs. These pathway relationships are (i) GSPs’ direct influence on SME performance; (ii) GSPs can alter SMEs’ EO; (iii) GSPs mediate or moderate the EO-performance relationship; and (iv) the GSPs-performance relationship is mediated via EO. The four-pathway relationship framework was used to explore the nature of the relationships of GSPs on the SME performance, EO and their dimensions among the reviewed articles. Our review revealed evidence for all four relationship pathways. However, their nature, direction, and impact dimensions varied according to the countries, stage of development, industries’ affiliation, and the level of government policies in direct or indirect support.

  1. (i)

    GSPs’ direct contribution to SME performance

    Fifteen of the studies under review provided significant support for GSPs direct influence on SME performance. Most of these studies (11) were conducted in develo** countries, with the remaining four in developed countries. An Australian study found that government regulations and financial assistance positively impacted export performance, with only a small number of exporting SMEs who had received government funding being significantly hampered by government regulations or compliance (Chang and Webster, 2019).

    Most types of direct GSPs were found to positively impact SME performance. A Nigerian study suggested that the government’s direct GSP (various skills, knowledge and capabilities development) improved firm performance (Buba et al., 2021). Likewise, Hassan et al. (2019) found that government financial support (direct GSP) also had a significant influence on SMEs’ nonfinancial performance (such as management commitment and worker involvement, safety rules, procedures, and safety promotion policies) in Pakistan’s food industry. Interestingly, when investigating direct GSPs in a specific industry, most studies investigated how financial GSPs influenced the SMEs’ performance and found that financial support directly and positively influenced the SMEs’ performance.

    Most studies conceptualized GSPs broadly and did not specifically consider the idiosyncratic impacts of various GSPs. Zainol (2013) revealed that GSPs, including all direct and indirect support, had a significant positive relationship with firm performance in Malaysian indigenous family firms. Other studies also highlighted the positive impact of GSPs on production, adoption of new technology (Wang et al., 2023), financial performance (Shu et al., 2019), nonfinancial performance (Shu et al., 2019), innovation performance (Lek et al., 2022; Sinatoko Djibo et al., 2022), sustainable performance (Ishaq et al., 2023), internationalization, and growth (Yazici et al., 2016). Among these studies, most specifically investigated financial support, while the rest investigated more than one GSP, including financial support, as a single construct.

    At the same time, most indirect GSPs were found to have a similar positive impact on firm performance. For example, some studies showed that regulatory policies positively impacted business performance (Chang & Webster, 2019; Roxas & Chadee, 2013). Similarly, the institutional environment where these SMEs functioned also significantly impacted their performance (Alarjani et al., 2020; Doblinger et al., 2016), but not in all cases. Some studies showed that the institutional environment could negatively affect export (Ajayi, 2016) and innovation performance (**e et al., 2013). When businesses used government support, they often faced additional compliance issues (Chang & Webster, 2019). The fact that some regulatory policies and institutional environments can impact SME performance positively and negatively suggests that they could be enabling or impeding. Distinguishing the regulatory and institutional environment as enabling or impeding may help to explain the causal direction of these indirect nonfinancial policies on SME performance.

    While the studies on the impact analysis of GSPs on SMEs performance did not differentiate between necessity- and opportunity-driven SMEs, it is important to note that accessing and recognizing resources have varied depending on the characteristics of SMEs (lifestyle, survival, managed-growth, and high-growth). Cassia and Minola (2012) found that extra-ordinary resource access and opportunity recognition were linked directly as causal antecedents of hyper-growth SMEs; however, Munoz et al. (2015) established that higher-performing businesses sought less government support. This could be attributed to various reasons, including that proactive SMEs may leverage necessary support from sources within their business network and opt to go with efficient sources rather than wait for a lengthy and complex bureaucratic government process (Sinatoko Djibo et al., 2022), or their enhanced capability may enable them to discover innovative solutions for their needs.

  2. (ii)

    GSPs’ impact on SMEs’ EO

    Nineteen of the reviewed studies examined how GSPs affect SMEs’ EO. Most studies conducted in develo** countries identified that GSPs, including direct and indirect GSPs, positively impacted the EO or its underlying dimensions (innovativeness, risk-taking, proactiveness, autonomy and competitive aggressiveness). For example, **ao et al. (2021) suggested that GSP participation positively correlated with proactiveness, innovativeness, and risk-taking. Further, a simulation study conducted using ‘pharmaceuticals industry data’ found that government incentives could raise the firm’s innovativeness, which varied depending on the incentive type, size of the firm, and stage of the research and development of the firm (Ciabuschi et al., 2020). Government institutions such as laws, regulations, policies, and support are statistically significant in positively impacting SMEs’ EO (Roxas & Chadee, 2013; Rwehumbiza & Marinov, 2020; Shu et al., 2019). Furthermore, Tarraço et al. (2023) differentiated that institutional regulatory pressures exert a more significant strategic influence on eco-innovation in firms in emerging countries than in developed countries.

  3. (iii)

    GSPs mediate or moderate the EO-performance relationship

    The well-established EO-performance relationship is said to be moderated or mediated by several factors, including GSPs. Twelve studies reviewed investigated the moderation effect of GSPs. Most of the studies (ten) found that GSPs significantly and positively moderated the EO-performance relationship (Bahari et al., 2017; Falahat et al., 2021; Khattak and Shah, 2020; Nakku et al., 2020; Seow et al., 2021; Shehu and Mahmood, 2015; Zeebaree and Siron, 2017; Zeebaree et al., 2018). At the same time, GSP negatively moderated the EO-performance relationship of highly proactive start-up businesses in Malaysia (Musa et al., 2017).

    Further, Basco et al. (2020) identified that the SMEs’ institutional (entrepreneurial business environment) and cultural differences moderated the EO-performance relationship in a multi-country context. In contrast, studies investigating the moderation effects of GSPs and external environments (including government interference) were insignificant and unsupported (Shehu and Mahmood, 2015; Hutahayan, 2019). Nevertheless, Tyler et al. (2023). Demonstrated that regulatory pressure enhances the impact of proactive orientation on environmental practices. Interestingly, only one study confirmed that GSPs positively and significantly mediated EO and SME performance, while EO had no direct effect on business performance (Udiyana et al., 2020).

  4. (iv)

    The GSPs-performance relationship is mediated by EO

    Eight studies investigated the mediating effects of EO on the GSPs-performance relationship. Most of these studies (07) were conducted in develo** countries. Roxas and Chadee (2013) found that EO mediated the GSPs–performance relationship, i.e., formal institutions (law, policies, regulations, and support) significantly and positively influenced performance via EO. This study identified that the indirect effects of GSPs via EO had larger impacts than direct and total effects and indicated that EO partially and strongly mediated the relationship between GSP dimensions and SME performance.

    Firman et al. (2018) confirmed that GSPs impacted firm performance and strategic renewal through EO. EO also positively affected business sustainability via green radical innovation (Han & Niu, 2023) and entrepreneurial satisfaction in businesses that started with government support (Lee & Kim, 2019). Recently, Yang and Yu (2022) demonstrated that EO is an intermediatory factor in the relationship between institutional support (formal and informal) and innovation performance. Further, they established that innovative resource acquisition reinforces the mediating role of EO between institutional support and innovation performance. On the other hand, a study on Malay indigenous family firms revealed that government-aided programs and firm performance were not mediated by EO (Zainol, 2013).

Conceptual framework

Based on the above review, we developed an integrated conceptual framework to address our research question of how EO, GSPs and performance are interrelated. Figure 6 summarizes the various relationships identified in the empirical studies reviewed, with direct relationships depicted as continuous black lines, indirect mediatory relationships depicted as dotted purple lines, and indirect moderator relationships depicted as continuous red lines. Both the most observable and lesser divergent views on the nature of the interrelationships between the constructs were reflected in the linkages and different sizes of the positive or negative signs provided in Fig. 6. This visualizes the empirical landscape to inform further research required in this field.

Fig. 6
figure 6

Conceptual framework. Note Signs in the brackets show the signs of the relationship

Content analysis using Leximancer

Leximancer analyzes keywords and phrases from source texts and shows the extracted information visually in an interactive conceptual map (Leximancer, 2018). Smith and Humphreys (2006, pp. 262–263) explain Leximancer extraction by classifying it as: ‘semantic’ and ‘relational’ extraction. Semantic extraction comprises the automatic discovery and extraction of thesaurus-based words or ‘concept seeds’ in the text data. With minor manual adjustments, the default settings were used to form concepts that naturally avoided potential researcher bias. These minor adjustments included the following: After the generation of concepts, similar concept words (for example, businesses, business, firm, firms and enterprises) were merged, and some of the ‘name-like’ (proper noun) concepts and research terminology concepts were removed after being investigated (for example, SME, China, entrepreneurship, management). In addition, general words that did not contribute to the meaning and interpretation or were over-represented due to the research language used in the academic publications were also removed. These terms frequently lacked semantic meaning relevant to this investigation and hence tended to obstruct interpretation by overpopulating the idea maps, concealing more important concept clusters.

The second stage of Leximancer analysis, relational extraction, extracts the lexical co-occurrence of the given ideas to exhibit semantic and grammatical grou** around prominent themes (Smith & Humphreys, 2006). This study employed four sentences per block as sliding text instead of two sentences in the default settings because repeated analysis indicated that thesaurus abstraction using four sentences per block delivered more accurate relational extraction. Using Leximancer, we developed a two-dimensional concept map (Fig. 7) of the data. Grey dots in the map indicate thesaurus concepts, and the dots’ sizes vary depending on their respective incidence frequency. These concepts, linked to other concepts by grey lines, indicate that they commonly appear together in a phrase, a four-sentence sentence block, or a paragraph and share similar meanings or close relationships. Highly connected concepts appear close to one another on the concept map. The most dominant or interconnected concepts are identified as themes. These themes are presented as circles covering closely linked groups of concepts. The theme circles are presented in varying degrees of brightness or ‘heat’ (the hottest being red and so on, according to the color wheel), which shows the relative intensity of the theme. The closeness between the bubbles indicates the close association between the themes.

Fig. 7
figure 7

Leximancer concept map (unstable) at the theme size 60% with the visible concepts of 100%

The first-round analysis resulted in an unstable concept map due to a huge mountain in the data (overrepresented concepts) that nullified small hills and valleys (Leximancer, 2018). Semantic analysis requires a stable map. Stability can be achieved when underlined themes and concepts are identified consistently and reliably. The over-connected concepts lead to a loss of differentiation, and stability should be recognized. The themes are identified and represented each time differently. As stated in the methodology section, all the articles investigated the firm-level analysis. As a result, the over-connected concepts directly related to firms and the predominant ‘firm’ theme in the unstable map were deleted. The new map was re-clustered until the generation of a stable map (Fig. 8). Finally, the above conceptualization of GSPs and their relationships with EO and SME performance are triangulated by the emergent themes and concepts inductively, resulting in Leximancer extraction by manually reviewing the actual text relating them.

Fig. 8
figure 8

Leximancer concept map (stable after five iterations) at the theme size 60% with the visible concepts of 100%

The dominant themes generated in the analysis are ‘performance’, ‘innovation’, ‘development’, ‘industry’ and ‘entrepreneurship’. These themes are closely intertwined with ‘industry’ and ‘performance’ overlap** with ‘innovation’, indicating the intimate association between the themes. The most frequently occurring themes, ‘performance’ and ’innovation’, are presented in warmer colors, and the less recurrent themes, ‘development’, ‘industry’ and ‘entrepreneurship’, are in cooler colors. The Leximancer analysis includes concepts’ frequency count and relevance within the data set. This map can be interpreted using the frequency, indicating a concept’s co-occurrence within a text block of four sentences and relevance relating to the relative importance of the concept with the top occurring concept’s count (see Fig. 8).

In the stabilized concept map (Fig. 8), the theme ‘performance’ has the highest count (8592) with 100% relevance, followed by ‘innovation’ (6219), with a relevancy of 58%. All these themes are interconnected, with the most prominent theme being ‘innovation’, which overlaps with more than a quarter portion of the other main themes, ‘performance’ and ‘industry’. Within ‘performance’ theme, concepts of entrepreneurial and strategic orientation were closely linked with performance, resembling the main relationship investigated in the selected studies. Concepts such as financial capital, social resources, capabilities, and networks were identified in the region and delineated by significant themes of ‘performance’ and ‘innovation’. This highlights the importance of various resources and capabilities for both innovation and performance.

The key concepts of government, support, environment, institutional and institutions are closely associated with strategic/entrepreneurial orientation and performance within the theme ‘performance’. The notion of the government is intricately intertwined with several other concepts, including financial, capital, resources, role and policies. Moreover, there exists an indirect connection between government and performance construct, established via entrepreneurial and strategic orientation, confirming our systematic literature review findings that GSPs affect firm performance indirectly through EO.

Further, various resources financial, capital, social, capabilities, and network intertwined between government and performance. This suggests that the government facilitates the provision of various resources that impact performance. The direct connection between government and performance via resources further confirms that GSPs affect firm performance directly. Additionally, this government concept is also related to the institutional, environment, EO, policies, financial, capital and resources. These closely clustered concepts support our literature review findings on the relationship between various conceptualizations of GSPs on SME performance. This proximity of concepts and the overlap** themes of various GSPs, EOs, and SME performances also support our conceptual framework.

The ‘industry’ theme possessed various interlinked concepts, such as market, opportunities, sustainable, environmental, practices, technology, process, products and services, implying that industrial elements of market opportunities, sustainable practices and environmental are essential for innovation, which is important for economic growth and development of the sector. Additionally, the concepts of strategy, management, competitive advantage, organizational, learning and ties are closely interlinked with innovation and reflect the underlying literature’s stance that firms’ success depends on opportunity identification, competitive advantage and innovation. Overall, the numerous paths studied by the manual analysis were confirmed by the themes, concepts, and how they spread and interlinked.

Discussion

This review explores the epistemological nature and the interconnections among the constructs of EO, GSP, and the performance of SMEs. While the relationship between EO and performance has been well-established for over three decades (Rauch et al., 2009; Wales, 2016; Wales et al., 2021), the investigation of bilateral EO - GSP relationships or trilateral EO-GSP-FP relationships started appearing vividly in scientific outlets after 2008. This may be because scholars became more interested in investigating increased government involvement in SMEs’ growth and entrepreneurial development after the global economic crisis in 2007/2008. The scientific publications investigating the impact of GSP peaked in 2020, immediately after the COVID-19 pandemic. This trend suggests that researchers are increasingly interested in exploring these relationships. Notably, the higher number of studies in Asia may be attributed to the government’s increased involvement in develo** the SME sector due to its significant contribution to employment creation and poverty alleviation. China and Malaysia, two of the world’s leading develo** economies (World Bank, 2021), further contribute to this growing body of research.

When considering the reviewed studies, as evidenced in the strategic management literature, the focus has primarily been on the manufacturing sector, indicating a need for more research in other sectors, especially agriculture. Despite governments’ strong support for agriculture in develo** countries, given its relative economic importance, agricultural SMEs are underrepresented in this body of literature (Dias et al., 2019; Fitz-Koch et al., 2018), suggesting a need for further studies. Furthermore, the review found that scholars are more inclined to statistically examine the relationships among these constructs: GSP, EO, and SME performance, which is necessary to establish relationships. While they may offer the benefits of generalization, this review highlights the need for more qualitative studies to capture the deeper nuances of GSP support to SMEs in diverse socio-economic, geographic and industry contexts.

In the identification of complex relationships, the review identified four pathway relationships: (see "Relationships between GSPs, EO and SME performance" section) (i) GSPs’ direct influence on SME performance; (ii) GSPs can alter SMEs’ EO; (iii) GSPs mediate or moderate the EO-performance relationship; and (iv) the GSPs-performance relationship is mediated via EO.

The first pathway relationship is a direct influence of GSPs on SME performance. While explaining this pathway relationship, the findings suggest that governments play a vital role for SMEs in facilitating, regulating, and catalyzing their performance (Hutahayan, 2019). As facilitating institutions, governments can assist SMEs in achieving their established goals by providing necessary financial support in grants, credits, guarantees and subsidies, and nonfinancial support such as training, guidance, informational support, and marketing assistance (Hutahayan, 2019; Petti et al., 2023; Wang et al., 2023). These direct supports enhance a firm’s performance in two ways. Firstly, these supports offer various resource portfolios, including financial and physical resources, which cause better performances (Buba et al., 2021). Often, businesses are in search of new and additional resources. In this situation, governments’ direct support is generally offered free of charge or at a lower rate, and this support is valuable for businesses. Secondly, policies that provide financial support directly lower the costs and efficient regulatory policies reduce associated transaction costs (business registration, licensing, and permit applications) (Roxas & Chadee, 2013), reducing production costs. Therefore, such assistance will contribute to some of the costs in the cost-benefit analysis and help improve the firm’s performance (Ciabuschi & Lindahl, 2018). The direct GSPs–SME performance relationship is also moderated by several factors, including networking, political ties, the external environment (Buba et al., 2021), and EO (Cassia & Minola, 2012).

Although the review suggests that both direct and indirect GSPs generally positively impact SMEs’ performance, we argue that governments’ direct policy support towards a small number of opportunity-based SMEs may negatively impact a large number of other SMEs who would not receive similar types of support. Nevertheless, the government might need to support necessity-based SMEs directly because their main objective is survival. To enhance the efficacy of the GSP, we suggest that both government and SMEs can benefit by assuring indirect non-targeted support (financial and nonfinancial) to opportunity-based SMEs and targeted criteria-based support to necessity-based SMEs. Additionally, it is essential to incorporate SMEs into policy formation because they possess industry-specific knowledge and understand their operational difficulties. Moreover, their involvement can shed light on the practical effects of policies and foster a sense of ownership, ensuring their opinions and interests are considered. Therefore, participatory policymaking fosters inclusivity and enhances efficacy.

The second pathway relationship of GSPs positively influence a firm’s EO in three ways. First, direct GSPs such as education and training (Wickramaratne et al., 2017) improve the EO by develo** skills and competencies to undertake entrepreneurial activities, which will aid strategic decision-making and differentiation in competitive and changing markets (González-Tejero & Molina, 2022). Meanwhile, financial support (Ciabuschi et al., 2020; Wang et al., 2023), information, advice, and coaching and mentoring (Falahat et al., 2021; Wickramaratne et al., 2017) increase the resource base of the firms and thereby improve SMEs’ EO by involving more innovative and risk-taking activities (Khattak & Shah, 2020). It is also suggested that the greater the perception of the effectiveness of the policies, the more likely the SMEs are to become involved in entrepreneurial activities (Dai & Si, 2018), as this creates confidence, proactiveness, and autonomy (Cassia & Minola, 2012; ** economy perspective. Journal of Small Business Management, 58(1), 2–31." href="/article/10.1007/s11365-024-00993-3#ref-CR60" id="ref-link-section-d96305076e2159">2020). This was due to GSPs that assisted SMEs in overcoming financial and nonfinancial constraints that impeded their business operations while also strengthening their entrepreneurial practices. For instance, proactive firms were less dependent on direct GSPs (Musa et al., 2017). In certain scenarios, regulatory pressure enhances the relationship between strategic orientation (proactive and innovative) and firm performance (Tyler et al., 2023). Our argument on whom to support is further strengthened by the fact that innovative, proactive, and risk-taking firms are often categorized as necessity- and opportunity-driven SMEs. Therefore, it is worth exploring the position of GSPs on EO performance, considering the dichotomy between opportunity-driven and necessity-driven SMEs.

The fourth pathway relationship is GSPs-performance relationship is mediated by EO. Government institutions can support firms with the necessary resources to implement innovation activities, mitigate the associated risks, foster a greater inclination to innovate, and ultimately enhance innovation performance (Shu et al., 2019; Yang & Yu, 2022). The influence of government institutions manifests in various ways in SMEs’ performance, including the rapid acquisition of key resources for innovation, effective exploration of external financing opportunities to enhance innovation willingness, and the identification of potential market opportunities aligned with government planning. Consequently, this heightened engagement with informal institutional support contributes to increased innovation input capacity and improved innovation performance. This variation in the role of EO on the GSP–SME performance relationship might also be explained by the differences in the SME’s intentions. Therefore, we suggest that further research exploring the relationships between these three constructs – GSP, EO, and SME performance – and the relationship efficacy of GSPs could use opportunity- and necessity-driven classification of SMEs as a contingent factor in the study of the impact of GSP on SMEs’ EO.

In summary, regarding who should receive preference from government support, entrepreneurial or non-entrepreneurial SMEs, our review shows that only one study differentiated between necessity- and opportunity-driven SMEs while examining EO and institutional environments (Parga-Montoya & Cuevas-Vargas, 2019). This review has identified the glaring gap in the literature that puts all SMEs in a single basket while studying the relationships between GSP and SMEs’ EO. It constitutes a significant missed opportunity for government and development partners who need to optimize their limited resources to support SMEs that need it most or can contribute most to the economy. While Parga-Montoya and Cuevas-Vargas (2019) established that institutional burden (including regulatory, normative and cognitive) significantly increased the EO of the opportunity-driven SMEs as opposed to the necessity-driven SMEs, more studies of a similar nature are required to confirm the finding. Although based on a single study, the finding reveals that the role of institutional mechanisms in promoting SMEs’ EO, albeit at varying degrees, is worth considering in future studies.

Conclusion

GSPs play a vital role for SMEs in several ways. Since resources are limited, appropriate allocation will be helpful in efficient and effective resource usage. This systematic review identified and categorized the impact of government policy on SMEs’ EO and performance by implementing various direct (targeted) support policies, including financial (grants, credits, subsidies, incentives) and nonfinancial (technical and training, advisory, and trade-related) support initiatives, and via indirect (non-targeted) support policies to develop a conducive institutional environment and entrepreneurial ecosystem (regulatory framework, taxation, competition policy and institutional environment). This type of categorization would make it easier for future studies to identify the impact of GSPs on SMEs’ EO and performance.

The findings suggest that the impact of GSPs on an SME’s EO and performance can be explained in four pathway relationships: (i) GSPs directly influence SME performance; (ii) GSPs change SMEs’ EO; (iii) GSPs mediate or moderate the EO-performance relationship; and (iv) GSPs-performance relationship is mediated by EO.

Except for one study, the studies did not differentiate whether the SME was necessity- or opportunity-driven. Although this is an essential issue, this study cannot identify a clear answer to this vital issue because 64 of the 65 studies that we examined did not differentiate SMEs based on their intention or characteristics.

Among the four possible pathway relationships, the effectiveness of the pathway could vary depending on the institutional context (develo** or developed country), the type of GSPs (direct or indirect, and financial or nonfinancial), the EO level of the SMEs (high or low level), and SME characteristics (objectives of the SMEs, whether growth-oriented: opportunity-driven, or survival-oriented: necessity-driven). We argue that direct GSPs are more effective and utilitarian for necessity-driven SMEs because this direct support significantly adds resources to these SMEs. On the other hand, indirect GSPs are more effective and sustainable for opportunity-driven SMEs. These policies create a conducive business environment for growth-oriented SMEs to perform more efficiently and effectively by optimizing their entrepreneurial behavior.

This review has fundamental limitations. The primary limitation was that although we did our best to include most of the previously published studies based on the rigorous methodology, there may also be studies not included because they did not appear in the keyword search. The other limitation is that the term GSP is not consistently used in the literature, and therefore, all government projects, programs, supports, and policies were considered GSP in this review.

This review has three main contributions. It contributes to the EO and GSP literature, which generates additional knowledge for policy debate and theoretical discussion. First, it discusses SME intentions and characteristics of necessity-driven (survival, lifestyle, and managed growth) and opportunity-driven (high growth) SMEs. If government support is extended to ventures based on their intention and growth target, that will be more effective and efficient for SMEs and governments. It is argued that policies for all kinds of SMEs are inefficient based on the types of GSP measures. Some direct GSP measures, such as training and grants, could target necessity-based ventures, and indirect GSP measures, such as establishing a supportive entrepreneurial ecosystem, could be targeted to opportunity-based ventures to ensure optimal use of limited government resources. Survival and lifestyle ventures, however, would be more interested in all types of direct support, as it would significantly change their resource base. Therefore, it can be suggested that government policy intervention should be tailored towards different types of ventures based on their characteristics and future aspirations by implementing arrays of support policies targeting ventures differently. Policymakers have to decide and implement particular support policies after analyzing the needs of the SMEs as they vary in industry, culture, and country (context).

Second, a framework was developed by identifying different pathway relationships of how GSPs influence SMEs’ EO and performance. This framework can be further explored in different economic and cultural contexts to holistically understand GSPs’ effect. Governments can also use this framework to plan and implement policies based on the firms’ total entrepreneurial activity and driving motivations.

Future research should analyze direct GSPs based on what firms receive and indirect GSPs that shape their institutional environment in study designs controlling for diverse contexts. Future researchers should, therefore, investigate the effectiveness of different policies and pathway relationships to clarify the interaction between the three main constructs. Future research may also look at the influence of GSPs on entrepreneurial experimentation in business model dynamics, such as business model innovation, validation, scaling, and pivots (Sanasi, 2023). Furthermore, when designing and implementing new GSPs, governments should be more conscious of the following: (i) what types of GSPs are more effective in supporting firms and (ii) to whom direct support should be provided to necessity- or opportunity-driven firms. To have successful policies, governments should take a holistic approach and actively engage with stakeholders collaboratively rather than adopt a top-down structure.