The Journal of Technology Transfer

, Volume 41, Issue 3, pp 506–529

Technology use and availability in entrepreneurship: informal economy as moderator of institutions in emerging economies

  • Saurav Pathak
  • Emanuel Xavier-Oliveira
  • André O. Laplume
Article

DOI: 10.1007/s10961-015-9423-x

Cite this article as:
Pathak, S., Xavier-Oliveira, E. & Laplume, A.O. J Technol Transf (2016) 41: 506. doi:10.1007/s10961-015-9423-x

Abstract

This paper investigates the contextual influences of institutions on the use of latest available technologies by early stage entrepreneurs in emerging economies. Hypotheses are developed and then tested using multi-level modeling techniques on a dataset covering entrepreneurs in 20 emerging economies. We utilized 10,431 individual-level responses from the Global Entrepreneurship Monitor survey from 2002 to 2008 and complemented it with data on country-level institutions such as the size of a country’s informal economy, intellectual property rights (IPR) regimes obtained from the Index of Economic Freedom and inward foreign direct investment (FDI) from the World Bank Group. Results on the direct effects suggest that levels of FDI negatively influences the use of latest technology by entrepreneurs in emerging economies, while the moderation effects of informal economy suggest that as its size increases (1) the negative effects IPR on the use of latest technology by entrepreneurs strengthens, and (2) the negative effects of FDI on the use of latest technology strengthens. These findings support the overall proposition that the size of a country’s informal economy is an important moderator of institutional influences on technology use by entrepreneurs in emerging economies. More generally, the study proposes that institutions may not have the same effects on entrepreneurs in emerging economies that might be expected in developed countries, suggesting that future research should take the level of socio-economic development of a country into account when theorizing the role of institutions.

Keywords

Technology use in entrepreneurship Institutions Informal economy Emerging economies Multi-level modeling 

JEL Classification

L26 O34 P33 

1 Introduction

Most of the entrepreneurship literature considers ‘technology entrepreneurship’ to be predominantly associated with the development of new technologies by entrepreneurial ventures (Shane and Venkataraman 2003). The underlying assumption is that resources deemed necessary to develop new technologies are available for use by entrepreneurs, independent of the context. Here, we challenge this assumption and posit that contextual factors, manifested by a country’s institutions, exercise regulatory influences (Lund and McGuire 2005; Munir and Phillips 2005; Swan and Newell 1995) on the availability of resources for use in entrepreneurship. The outcome from the pursuit of technology entrepreneurship may be contingent upon the extent to which country’s institutions ease the use of latest technologies by entrepreneurs. This paper investigates contextual influences of country-level institutions on the use of latest technologiesby early-stage entrepreneurs, particularly in the context of emerging economies.

Whether or not entrepreneurs perceive the use of latest technologies as feasible and desirable may depend on a country’s institutions because institutions not only influence what individuals search and see, but also, how they react to what they see (e.g., Kim and Li 2013; Hwang and Powell 2005). We reason that the size of a country’s informal economy, the strength of intellectual property rights (IPR) regimes, and inbound foreign direct investment (FDI) are dimensions of the institutional environment that influence entrepreneurs’ perceived feasibility and desirability of using the latest technologies in their entrepreneurial efforts. In addition, the strength of association between IPR, FDI, and the use of latest technology may be contingent upon the size of a country’s informal economy, such that the size of the informal economy is expected to moderate the influences of IPR and FDI.

Entrepreneurship scholars often limit their theorizing to the formal economy, leaving the segment of the economy operating outside the realm of formal institutions under-researched (Goel et al. 2015; Webb et al. 2013). The informal economy is frequently referred to using different terminologies such as shadow, hidden, black, underground, gray, clandestine, illegal, and parallel economy (Fleming et al. 2000). However, some efforts have been made to distinguish the informal economy. For instance, Webb et al. (2013) suggest that the scope of the informal economy is influenced by institutional incongruence, or differences between formal and informal institutional boundaries. The informal economy recognizes that some forms of production are illegal but legitimate—reflecting a gap between the law and social norms. The informal economy remains a controversial topic for many scholars and there is a lack of consensus about its strict definition. Nonetheless, several authors agree that informal economies are growing around the world, and consequently their relevance for effective policy making is rising (Feld and Schneider 2010; Giles 1999; Pedersen 2003; Schneider and Enste 2000). Here we adopt the definition and estimates (mainly due to data availability) of Schneider et al. (2010):

(…) the shadow economy includes all market-based legal production of goods and services that are deliberately concealed from public authorities to avoid payment of income, value added or other taxes; to avoid payment of social security contributions; having to meet certain legal labour market standards, such as minimum wages, maximum working hours, safety standards, etc.; and complying with certain administrative procedures, such as completing statistical questionnaires or administrative forms (Schneider et al. 2010: 444).

High quality formal institutions are credible, not corrupt and well-functioning in that they create positive fiscal outcomes (Torgler and Schneider 2009). Taken as a whole, a country’s institutional quality is expected to be better in developed than in emerging economies (Levchenko 2007; Meyer 2001). Institutions may also exert different constraints depending upon the context where they are implemented (i.e., developed or emerging economies). If so, then a one-size fits all policy toward institutions is likely to be ineffective at encouraging productive forms of entrepreneurship universally, yet conceptual development in entrepreneurship research typically examines mature or developed market conditions, leaving a sizeable gap for studies examining both informal and formal institutions in emerging economies. In this study we contribute to the literature by addressing this gap.

Our theoretical framework and empirical design has two distinct levels (individual-level entrepreneurial behaviors and country-level institutions), thus necessitating multi-level estimation techniques to test our hypotheses. Our dataset combines individual-leveltechnology use in entrepreneurship measures from the Global Entrepreneurship Monitor (GEM) and country-level data on (1) IPR from the Index of Economic Freedom, and (2) inward FDI from the Global Competitiveness Index. Due to data availability constrains across all of these sources, our sample comprises of 20 emerging economies from 2002 to 2008. The results for the main effects indicate that FDI has a negative effect on the use of latest technologies at the individual-level—the other two direct relationships, although in the proposed directions, were observed to be statistically non-significant. However, the size of a country’s informal economy was observed to act as a potent moderator—as its size increases (1) the negative effect of IPR strengthens; and (2) the negative effect of FDI strengthens.

The text has the following structure. Firstly, we explain how we arrived at our dependent variable (theuse of latest technologies by early-stage entrepreneurs) and elaborate hypotheses pertaining to the main effects of the informal economy, IPR and FDI. Subsequently, we hypothesize the potential moderating effects of informal economy on the two institutions. We then introduce our multi-level modeling methods and test our hypotheses using a sample of early-stage entrepreneurs in emerging economies. Lastly, we discuss the implications of our study, its limitations, and conclude.

2 Theory and hypotheses

Given that the literature covering technology entrepreneurship is somewhat vast, we are faced with the challenge of selecting a definition of technology entrepreneurship that is suited to the multi-level study we have pursued. In their introduction to the special issue in Research Policy entitled “Technology Entrepreneurship”, Shane and Venkataraman (2003) use the term synonymously with new-technology companies—the unit of analyses thus being the firm. Similarly, in Beckman et al. (2012) introduction to the special issue on the same topic in the Strategic Entrepreneurship Journal, they distinguish technology entrepreneurs as those that focus on how opportunities are fostered through innovations in science and engineering. The papers submitted to these special issues include samples of firms from the information technology, biotechnology, alternative energy and internet search sectors. In both cases technology entrepreneurship is considered in relation to companies rather than entrepreneurs and defined as the development of new technologies by ventures.

In this study, we take a different approach to technology entrepreneurship through the examination of contextual influences on the use of latest available technologiesby early stage entrepreneurs—the unit of analysis is thus at the individual-level. Whether entrepreneurs develop new technologies or use the latest ones in their ventures, adoption of new technologies renders itself as being important for competitive advantage and economic growth driven by technological progress (Fagerberg 1987; Porter 1985). Entrepreneurs exercise opportunism (Seth and Thomas 1994), limited foresight, and bounded rationality (Simon 1985) when deciding whether the net gain from engaging in entrepreneurial behaviors exceeds the cost of missing other opportunities (Douglas and Shepard 2002), or alternatively, when choosing the type of entrepreneurship in which to engage. Every society may be endowed with its share of entrepreneurial individuals but whether they use this propensity for productive endeavors depends in large part upon the calculus imposed by prevailing institutions (Baumol 1990; West III and Bamford 2005).

2.1 The informal economy

The size of the informal economy in an emerging economy is expected to affect the use of latest technologies by entrepreneurs via three main mechanisms: (1) attraction, (2) reduced access, and (3) voluntary avoidance. The first mechanism, attraction, pulls entrepreneurs from the formal economy into the informal economy. Informal economy entrepreneurs, who seek to avoid paying taxes, social security, and other regulation-related costs (Quintin 2008), may be the same type of individuals who would circumvent costly contractual access to new technologies and rather access them through questionably legal means (e.g., infringing, pirating and counterfeiting). Others may do so in order to achieve competitive parity through lower costs in an environment where operating in the official economy may create a competitive disadvantage. Thus, a large informal economy in an emerging economy suggests a potential increase in the use of latest technologies by entrepreneurs (Goel et al. 2015). However, this increase due to attraction mostly represents a shift from formal to informal use of technology; that is, those individuals who would have used the technology legally may be drawn to do so in the informal economy, thus, we would expect any net increase to be modest. Goel et al. (2015) suggest that innovations drive the attraction of the informal economy due to reduced competition at the early stages, delayed distribution to emerging markets, and lagging regulations.

The second mechanism, reduced access, requires the consideration of the technology provider’s perspective. Technology providers are firms that control large technology portfolios and that make their technologies available to other firms and entrepreneurs to be used through licensing arrangements. Firms in developed countries are expected to be the main source of the latest technologies and related knowledge spillovers because emerging economies tend to lag in terms of technological development (Narula and Sadowski 2002). However, contracting across a developed-developing country interface presents a number of challenges because of the varying levels of institutional development between the two contexts. Such contracting may be infused with higher levels of uncertainty and greater expectations of opportunism, especially when parties in emerging economies with larger informal economies are included. Weary technology providers may worry about the unreported use of their technologies, and may not trust the disclosures of emerging economy entrepreneurs. They may react by imposing excessive transparency requirements for the use and transfer of their technologies, or prefer to partner with established firms that they consider to be more legitimate. After all, legitimacy is important for gaining resources (such as technology) that are crucial for venture growth (Suchman 1995), especially for early-stage entrepreneurs who may have little past economic performance by which the holders of resources can rationally judge them (Zimmerman and Zeitz 2002). New ventures do not provide suppliers with confidence that they will survive and therefore should not expect special treatment or patronage (Starr and MacMillan 1990).

A third mechanism, voluntary avoidance, stems from the distrust that entrepreneurs may have for their own governments and their related formal institutions (Wallace and Latcheva 2006), especially tax authorities. However, a larger informal economy may be a normative reaction to government corruption or oppression that leads entrepreneurs to hide transactions that are vulnerable to intrusions by officials seeking bribes or who might use such tax revenues to support opposing groups (Choi and Thum 2005). Operating in the informal economy may eventually become socially embedded (Estrin et al. 2013), and thus come to be locally legitimate at the group-level (Webb et al. 2013), especially for those groups that are being oppressed or underrepresented. Under such circumstances, the use of latest technologies, especially those involving electronic transactions, may create vulnerabilities for entrepreneurs operating in the informal economy. For instance, online payment systems and software-based accounting systems record transaction details that could be made available to authorities, thus exposing hidden transactions. This may lead to injunctions, seizures, arrests or tax bills that are unwanted by entrepreneurs.

On balance, we expect that despite the attraction of the informal economy as a means of circumventing technology access costs, the mechanisms of reduced access and voluntary avoidance are more likely to reduce the use of latest technologies by early-stage entrepreneurs operating in emerging economies. Accordingly, we hypothesize that:

H1

As the size of the informal economy increases, the use of latest technology by early-stage entrepreneurs in emerging economies decreases.

2.2 Intellectual property rights

The IPR system is a formal institution that regulates the acquisition and use of latest technology—older inventions are typically in the public domain (e.g., most patents are valid for 20 years or less). Formal institutions affect what potential entrepreneurs search, see, and react to by adjusting the incentives for selecting entrepreneurship as a career path, (Hwang and Powell 2005). However, formal institutions also affect the opportunity and threat recognition of technology providers. Thus, the strength of IPR is expected to affect the use of latest technologies by early-stage entrepreneurs in emerging economies through two competing mechanisms: (1) increased access, and (2) reduced access.

At first glance, we might expect that stronger IPR in an emerging economy would give confidence to technology providers, thus increasing access to new technology for domestic entrepreneurs. The availability of legal processes should encourage technology providers and multinationals to implement fewer isolation mechanisms (Oxley 1999), and rely instead on the enforcement processes of IPR regimes to protect their technology portfolios, thus facilitating technology licensing. On the flip-side, strong IPR systems make it easier for technology providers to protect their designs and inventions, which can also create impediments to the use of latest technologies by indigenous entrepreneurs, thus reducing access. Technology providers often view their intellectual property as key to their competitive advantage (Rivette and Kline 2000) and may use, for instance, the patent system to take out competitors, build a reputation for toughness, to protect the ‘crown jewels’ of the firm, or simply to extract unaffordable royalties (Somaya 2003). Early-stage entrepreneurs in emerging economies may be made to pay higher fees to compensate technology providers for the increased perceived risk that they face, and a strong IPR system makes it easier for technology providers to collect license fees or impose injunctions to stop the unwanted use of their technologies.

Whether the net effect of these two mechanisms results in more or less access to new technology for emerging economy entrepreneurs may depend upon the beliefs and behaviors of technology providers. Signaling processes (Connelly et al. 2011) may lead technology providers to believe that emerging economies generally have loose IPR protection (Lall 2003; Forero-Pineda 2006), regardless of the actual strength of the institutions in specific countries. These signals may manifest themselves as ‘country of origin effects’ (Hong and Wyer 1989; Verlegh and Steenkamp 1999), and ‘reverse ethnocentrism’ (Agbonifoh and Elimimian 1999), which work upon the bounded rationality of technology providers, leading to unwarranted stereotypes that work against the interests of early-stage entrepreneurs in emerging economies. Indigenous entrepreneurs already face higher technology integration costs than their competitors in developed countries because of their lower absorptive capacity (Borensztein et al. 1998). Stereotypes leading to higher licensing royalties relative to average income levels may keep early-stage indigenous entrepreneur out of the game entirely as they begin to perceive a negative net benefit to pursuing new technology use. They may instead choose the less desirable, but more affordable options, of using older technologies or substituting more labor. In short, these arguments suggest that stronger IPR protection leads to reduced access, thus we hypothesize:

H2

As the strength of IPR increases, the use of latest technology by early-stage entrepreneurs in emerging economies decreases.

2.3 Foreign direct investment

FDI may be expected to affect the use of latest available technology by early-stage entrepreneurs in emerging economies through two competing mechanisms: (1) knowledge spillovers and (2) crowding effects. Knowledge spillovers from developed country multinationals are expected to be an important route through which the latest technologies are made available to indigenous entrepreneurs in emerging economies (Acs et al. 2009; Audretsch and Keilbach 2007). According to the knowledge spillover theory of the firm, FDI has the potential to promote economic growth via the transfer of technology and associated management practices (Acs and Szerb 2007), but only if the local conditions are right (Propris and Driffield 2006; West III and Bamford 2005). The problem is that most emerging economies have the wrong conditions: they have too little accumulated stock of human capital and absorptive capacity (Borensztein et al. 1998), and too little complementary capital (De Mello 1999). Given the wrong conditions, FDI may not provide the entrepreneurial opportunities expected for defecting local managers and engineers, and for independent local entrepreneurs observing multinational enterprises (Parker 2010; Stieglitz and Heine 2007).

Instead of promoting new technology use through spillovers, high levels of FDI in emerging economies may reduce the proportion of domestic individuals’ engagement with new technology through crowding effects in labor and financial markets (Danakol et al. 2013; De Backer and Sleuwaegon 2003; García et al. 2013). Crowding effects refer to the repercussions of the increased density of players caused by the presence of foreign firms competing with local entrepreneurs in factor and labor markets. For example, the best candidates for domestic technology-based entrepreneurship may choose to work for the higher wages of multinational enterprises. Similarly, funding that could have been received from banks and investors for the purchase of new technologies by indigenous entrepreneurs may instead be pulled toward vertical opportunities (e.g., supply and distribution) created by multinational enterprises (García et al. 2013). For example, domestic bankers and investors may prefer to lend money to and buy equity in projects involving multinational enterprises, which may be perceived to involve less risk, rather than support domestic entrepreneurs directly. In summary, the potentially muted utility of knowledge spillovers and augmented crowding effects of FDI in emerging economy contexts combine to suggest that:

H3

As inward FDI increases, the use of latest technology by early-stage entrepreneurs in emerging economies decreases.

2.4 Intellectual property rights in the presence of the informal economy

As previously argued, strong IPR may reduce access, whereas the size of the informal economy may affect the use of latest technologies by early-stage entrepreneurs in emerging economies via the mechanisms of attraction, reduced access and voluntary avoidance. These same mechanisms can also help to explain how the size of the informal economy might moderate the negative effect of IPR. The mechanism of attraction applies both to entrepreneurs using the latest technologies and to those using older technologies. However, as property rights strengthen, the mix is likely to be altered as the mechanism of voluntary avoidance becomes stronger. Stronger intellectual property rights imply that legal recourse exists allowing rights-holders to go after infringers (Rivette and Kline 2000; Somaya 2003)—i.e., reduced access. A side effect of these activities is that infringing entrepreneurs operating in the informal economy are likely to be pulled into the light of courts and authorities, including tax collectors.

Infringing entrepreneurs would be weary of being discovered as they do not want their activities to be revealed and therefore are likely to voluntarily avoid the use of latest technologies in their ventures. Similarly, stronger IPR suggests that acquiring pirated or counterfeited technology would be more difficult in the first place (i.e., reduced access), encouraging entrepreneurs to use legal versions of those technologies. The use of legal versions of technologies implies the payment of sales taxes and also suggests that authorities would be more likely to be made aware of entrepreneurs’ operations, and thus be able to acquire transaction records embedded, for instance, in databases and other technologies. To the extent that technology providers are more willing to comply with authorities in emerging economies (e.g., because of legislation such as the Foreign Corrupt Practices Act in the U.S.; also see Christmann and Taylor (2001), for an example involving environmental compliance), it may be easier to expose informal economy entrepreneurs through such means. Thus, we hypothesize:

H4

The size of the informal economy moderates the relationship between the strength of IPR and use of latest technologies by early-stage entrepreneurs in emerging economies, such that as the size of the informal economy increases, the negative effect of IPR strengthens.

2.5 Foreign direct investment in the presence of the informal economy

Drawing upon prior arguments, FDI into emerging economies may induce crowding effects and allow for reduced knowledge spillovers, meanwhile, the informal economy elicits the mechanisms of attraction, reduced access and voluntary avoidance. Some of these mechanisms may combine to explain how the size of the informal economy might moderate the influence of FDI. Under the right conditions, multinational enterprises operating in emerging economies are often expected to be major providers of technology transfers (Audretsch and Keilbach 2007). However, the presence of a large informal economy deprives the state of the revenues needed to support a strong national system of innovation, further eroding the conditions required for effective knowledge spillovers. Moreover, when the size of the informal economy in an emerging economy becomes larger, multinational enterprises will be faced with greater uncertainty about how their technologies will be used if they are provided to indigenous entrepreneurs, thus reducing access. In fact, the high cost of arms-length contracts in the presence of uncertainty and opportunism is a key factor in the decisions of multinational enterprises to use wholly-owned subsidiaries and joint ventures instead of directly exporting technologies to access markets in the developing world (Beamish and Banks 1987; Oxley 1999; Williamson 1975). Multinational enterprises may thus be more inclined to take additional measures to protect their technologies because they may be more concerned about opportunistic uses. As the size of the informal economy increases, multinational enterprises may engage in various strategies to reduce the potential for piracy, counterfeiting and other competitive uses of their technologies (Goel and Nelson 2009; Rumelt 1997). These may include structures, organizational cultures and control systems that prevent local managers and engineers from gaining access to technology and related knowledge (Huyghe and Knockaert 2015; Markusen 2001). Overcoming these isolating mechanisms is burdensome for indigenous entrepreneurs seeking to use the latest available technologies in their ventures. Thus, we hypothesize:

H5

The size of the informal economy moderates the relationship between inward FDI and the use of latest technologies by early-stage entrepreneurs in emerging economies, such that as the size of the informal economy increases, the negative effect of FDI increases.

3 Methods

Our theoretical framework has two levels—individual-level and country-level. We hypothesize on cross-country direct as well as interaction effects of informal economy, strength of IPR and levels of inward FDI on the likelihood that entrepreneurs would be able to use the latest available technologies in emerging economies. This meant that we needed a data set that is cross-national in nature and that involves measures that are based on definitions that are applied consistently across countries to ensure valid comparisons (Coviello and Jones 2004).

We analyzed survey data for 20 countries for the years 2002–2008 from the publicly available Global Entrepreneurship Monitor (GEM) survey (Reynolds et al. 2005).1 GEM provides data from 2001 to 2008, but since our dependent variable was introduced in GEM’s survey questionnaire in the year 2002 our data set spanned 2002–2008. This comprehensive data set includes responses from 10,431 individuals who were identified as entrepreneurs but may or may not qualify as entrepreneurs using the latest available technologies.

Global Entrepreneurship Monitor (GEM) survey (Reynolds et al. 2005) had been designed and operationalized by the Global Entrepreneurship Research Association (GERA). GERA has been collecting individual-level data alongside country-level data across countries that reflect the incidence as well as determinants and outcomes of entrepreneurial activities in participating nations. Consequently, GEM is the most comprehensive and harmonized data set listing internationally comparative data on individual-level entrepreneurial behaviors making it suitable for studies, such as ours, that gauges the cross-national differences in how country-specific social reference groups’ norms influences individual-level entrepreneurial behaviors. Since 2004, Social Sciences Citation Index (SSCI)—listed journals have published at least 89 academic articles that uses GEM data set for entrepreneurship research (Bosma 2013) further consolidating its credibility. Publicly available GEM data is available only for 2001–2008 however.

Our initial database for emerging economies comprised of 10,431 (un-weighted) interviews of adult-age (18–64 years old) individuals out of which 1,592 (over 15 %) were identified as using the latest technologies in their ventures. This dataset was then complemented with data on national-level institutions for the period of interest: (1) size of a country’s informal economy relatively to official real GDP (Schneider et al. 2010), (2) strength of IPR protection from the Index of Economic Freedom (Miller et al. 2012) and (3) FDI per capita from the World Bank (2012). Descriptive statistics are also shown in Table 1.
Table 1

Sample descriptives

Countries

N

% of entrepreneurs using the latest technology

Informal economy

IPR

FDI

Emerging

Argentina

570

9.82

25.65

32.86

201.85

Brazil

751

4.13

41.08

50.00

95.39

Chile

1062

23.35

20.57

90.00

351.60

China

947

18.16

13.72

27.14

60.72

Colombia

1119

18.05

41.97

31.43

108.24

Czech Republic

84

34.52

20.20

70.00

382.94

Hungary

273

3.30

26.12

70.00

1122.02

India

278

32.37

20.13

30.00

11.39

Indonesia

352

24.15

24.48

50.00

9.97

Malaysia

487

8.21

30.38

50.00

171.77

Mexico

154

7.14

31.60

50.00

156.14

Peru

1489

8.19

62.85

32.86

93.18

Philippines

350

21.43

46.00

35.71

17.54

Poland

39

2.56

28.20

55.71

208.15

Romania

67

2.99

37.17

30.00

145.48

South Africa

526

16.16

30.08

50.00

51.67

South Korea

334

14.37

28.65

72.86

75.70

Thailand

1069

16.37

55.87

55.71

90.74

Turkey

250

6.40

33.58

50.00

99.33

United Arab Emirates

230

33.48

27.34

55.71

1447.46

 

10,431

15.26

32.28

49.50

245.06

Nd total number of individual-level response from a given country from 2002 to 2008,  % of entrepreneurs using the latest technology represents the national aggregate measures of individuals who were identified as using the latest available technologies in their ventures (dependent variable = 1, rest = 0) over 2002–2008

Source: GEM data set

3.1 Dependent variables

Our dependent variable is the individual-level—that of an entrepreneur’s—likelihood of using latest technology in entrepreneurship and was obtained from the GEM dataset. GEM identifies (1) nascent entrepreneurs (individuals who are active in the process of starting a new firm during the preceding 12 months and with expectations of full or part ownership, but have not yet launched one) and (2) new entrepreneurs (owners-managers of new firms who have survived for 42 months and have paid wages to any employees for more than 3 months) as early stage entrepreneurs. GEM categorizes established entrepreneurs (owner-managers of firms 42 months old or older) separately. Only nascent and new entrepreneurs are operationalized by GEM as “early stage” entrepreneurs who by definition are still trying to survive such that they are the representative sample of those still engaged in making an “entry”. Latest technologies for use by entrepreneurs would matter more to “early stage” entrepreneurs than those already established and hence the rationale behind combining nascent and new entrepreneurs in our working sample.2

Identified entrepreneurs (from amongst the set of identified nascent or new entrepreneurs) that confirmed that they were using the latest technologies and procedures—that were not available more than a year ago, assumed a value of 1. Others who confirmed that they were using technologies or procedures that were older (either available over a year ago or over more than 5 years ago), were deemed to be entrepreneurs—also involved with use of technology—but were not using the latest ones. This set of entrepreneurs all assumed a value of 0. Our dependent variable was thus operationalized as a dichotomous variable (0 or 1) where all identified entrepreneurs using the latest technologies and procedures (less than a year old) were represented by values of 1 and others using older technologies and procedures (up to 5 years or older) were all represented by values of 0. This sampling strategy is justified by the view suggested by Baumol’s (1990), where every country has its share of entrepreneurs, but how that potential will be used depends on institutions. Moreover, institutions would also exercise a regulatory influence on the use of latest technologies by entrepreneurs.

3.2 Predictor variables

3.2.1 Country-level (level-2) institutions

We used three country-level institutions in our analysis—the size of a country’s informal economy, the strength of a country’s IPR protection regime and the levels of FDI per capita. The data sources span over various years. We used the averages over those years that were common to the GEM data set—2002 to 2008—wherever possible.

We referred to Schneider et al. (2010) assessment of a country’s size of shadow economy relatively to ‘official’ real GDP to obtain a proxy for our variable of interest, namely, the size of the domestic informal economy for each country in our sample. Schneider et al. (2010) used a Multiple Indicators Multiple Causes (MIMIC) model which essentially is a Structural Equation Model (SEM) with one latent variable—shadow economy in their case. Even though considerable controversy still prevails about the definition and quantification of informal economy estimates, among many potential sources we chose to rely on Schneider’s extensive expertise on the topic—more than 640 scholarly publications, many of which on the shadow economy. Schneider et al. (2010) estimated the size of shadow economy across 162 countries. In a Structural Equation Modeling (SEM) approach, the relationships among certain observed variables are first explained in terms of their covariances (which measure the extent to which two or more variables change or vary together) which are then assumed to be generated by a set of a smaller number of unobserved variables. An overview of the SEM approach is available from Bollen (1989). Schneider et al. (2010) first considered the relationships between seven country-level observed variables—tax and social security contributions burden, intensity of regulations, public sector services, official economy, monetary indicators, labor market indicators and the state of the official economy. The covariances among these seven variables were then assumed to be generated by the size of each country’s informal economy (Schneider et al. 2010, p. 447–452). The source lists data from 1999 to 2007. However, we used the average over 2002–2007 for all countries and used those values as measures of shares of informal economy for each country for 2002–2008.

The IPR index was obtained from the Index of Economic Freedom (IEF) published by the Heritage Foundation3 (Miller et al. 2012)—scaled 0 to 100 and available from 1995 to 2012—combines various aspects of the degree to which private property is protected in a given country, the extent to which IPR are respected and citizens are protected against illegal expropriation of property, in addition to the ability of individuals to accumulate private property. Overall, it signals the degree to which a country’s laws protect private property rights. The use of the IPR index published by the Heritage Foundation in entrepreneurship research has been validated by recent scholarly works such as Autio and Acs (2010), McMullen et al. (2008), and Zhao (2006). For our study, we used the average scores of IPR over the 2002–2008 period for each country included in our dataset. A lower score indicates that intellectual property is loosely protected, whereas a higher score indicates it is tightly protected.

FDI per capita values were assessed using World Bank (2012) data. The source lists data since 1960, however for the purpose of our study average scores over 2002–2008 were used. The use of this data set in entrepreneurship research is validated by articles published in several academic journals (Kim and Li 2013; Pathak et al. 2013). We initiated this process by collecting data on FDI net inflows (new foreign investment inflows less disinvestment) as a percentage of Real GDP, and consequently, Real GDP per capita in constant dollars (2000 US$). We then computed FDI per capita in constant dollars by multiplying both variables for each of our countries during the period of interest, and proceed by averaging the values per country.

Given that the scores have been generated separately and come from separate sources, one unit change in these scores would not mean the same thing across all sources. Hence, in order to facilitate consistent interpretation of the analysis, we z-standardized these country-level.

3.2.2 Interaction terms

We created two interaction terms to test our hypotheses, those between (1) informal economy and IPR and, (2) informal economy and FDI per capita in predicting the likelihood of using the latest technology by entrepreneurs, where informal economy is the moderator. Z-scores of all variables were used to generate corresponding interaction terms.

3.3 Country-level and individual-level controls

Neoclassical growth theory suggests that technological progress is the main driver of economic growth. The main reason being that despite the positive direct and indirect effects associated with the fostering of macroeconomic stability, infrastructure, institutions, financial system, labor and capital markets among others, all these ultimately converge to a scenario of diminishing returns in the long run. In this context, a recent growth theory proposed by Parente and Prescott (2005) indicates that barriers to technological adoption are the main reason for the development disparities across nations, given the principle of “common technology”, which assumes that once a technology is created it is available to everyone, though not necessarily adopted. Such barriers to adoption comprise everything that may deter technological progress at the national level, ranging from the rule of law to outright sabotage—“whatever their form, each has the effect of increasing the cost of technology adoption” (Parente and Prescott 1994, p. 299). The ultimate consequence is the lack or buffering of domestic technological shocks, regardless if they were induced exogenously or endogenously, that is, if the technology was created domestically or elsewhere.

Based on the above, we used barriers to technological adoption as a country-level control variable. Country scores for the 2002–2008 period were generated using the Cassou and Xavier-Oliveira (2011) model, which improved the Parente and Prescott models by introducing adjustment costs, thus enabling closed-form solutions and enhancing the overall simulation structure. Positive values of this unbounded macroeconomic indicator mean that the country of interest has higher barriers to technological adoption than the technological leader (United States, by assumption); negative values indicate the opposite.

In addition, we controlled for a number of individual-level variables as well as demographic characteristics, all of which were obtained from the GEM dataset. We controlled for an individual’s familiarity ties with entrepreneurs, their perceptions of opportunity recognition and whether or not an individual was a business angel.

Ties with entrepreneurs indicates vicarious exposure and was measured by asking whether or not the individual knew someone personally who had started a business in the past 2 years (1 = yes, 0 = no). Familiarity ties with entrepreneurs have been suggested as an important source of vicarious experience that affects the entrepreneurial intentions of individuals (Davidsson and Honig 2003) as they learn and replicate actions by observing others.

Opportunity recognition was operationalized by asking individuals whether or not they felt that in the next 6 months there would be good opportunities in their countries to start up their own businesses (1 = yes, 0 = no). Opportunity recognition is a skill highly relevant in the field of technology where some breakthrough product innovations have largely involved the transfer of a ‘low-value’ technology from one business sector to another where it then becomes ‘high-value’ (Christensen 1997). Only those entrepreneurs who are alert to such opportunities reap the benefits of new technology use.

Business angel was operationalized by asking individuals whether or not they had, in the past 3 years, personally provided funds for a new business started by someone else excluding any purchases of stocks or mutual funds (1 = yes, 0 = no). Informal investors or business angels serve as important source of financing especially when external financing is restricted. Being a business angel signals the availability of personal funds and confidence to support an inherently risky proposition.

Further, an individual’s gender, age, education level and household income (Arenius and Minniti 2005) have been recognized to exercise an important influence on entrepreneurship. Hence, we controlled for them. Individuals’ level of education had five levels—0 = none; 1 = some primary; 2 = primary; 3 = secondary and 4 = graduate), and socioeconomic status represented by household income tier had 3 equally large strata in each country—1 = lower income tier; 2 = middle income tier and 3 = upper income tier).

3.4 Estimation methods

Since we combined individual-level observations with country-level measures of institutions, the data was analyzed using hierarchical linear modeling methods. Since our dependent variable was dichotomous, we carried out our outcome regressions using random-effect logistic regression4 to estimate the influence of country-level factors (level-2) on individuals’ likelihood of using latest technology in entrepreneurship.

Subsequently, we adopted a four-step testing strategy to estimate the likelihood of use of the latest technology by entrepreneurs in emerging economies. First, we estimated between-country variance that existed in the dependent variable by including no predictors or controls in our random-effect logistic regression model. We observed significant country-level variance in our dependent variable suggesting that county-level factors could be responsible in explaining this variance in the dependent variable. This finding mandated multi-level analyses since country-level variance could only be accounted for by country-level factors. This regression model was called the “null model” (Model 1 in Table 4).5 As our second step, we added individual-level as well as country-level controls prior to the addition of the three country-level predictors (Model 2). As our third step, we included the three country-level variables (informal economy, IPR and FDI per capita) to the model in step two to investigate the main effects of the three country-level (Model 3). Finally, as our fourth step, we tested the interactions of informal economy with IPR and FDI per capita (Model 4).

4 Results

Descriptive statistics and correlation matrix are shown in Tables 2 and 3 respectively. Table 4 shows the effects on the use of latest technology by entrepreneurs in emerging economies.
Table 2

Descriptive statistics

Variables

N

20 Emerginging Countries

Mean

SD

Min

Max

Individual-level variables

Use of latest technology by entrepreneurs

10,431

0.15

0.36

0

1

Age

10,431

36.41

11.51

18

64

Gender

10,431

1.44

0.50

1

2

Education level

10,431

2.18

1.06

0

4

Household income

10,431

1.92

0.81

1

3

Ties with entrepreneurs

10,431

0.65

0.48

0

1

Business angel

10,431

0.13

0.33

0

1

Opportunity recognition

10,431

0.62

0.49

0

1

Country-level variables

Barriers to technology adoption

20

6.80

5.04

0.14

24.30

Informal economy

20

32.28

15.48

13.72

62.85

IPR

20

49.50

17.57

27.14

90.00

FDI per capita

20

245.06

386.08

9.97

1447.46

Table 2 reports values for only those years that the country has participated in the GEM survey (unbalanced panel). Table 2 sample statistics were the ones used in all our regression models

Table 3

Correlation matrix for emerging economies

Variables

20 Emerging economies

1

2

3

4

5

6

7

8

9

10

11

12

1. Use of latest technology by entrepreneurs

1.00

           

2. Age

−0.01

1.00

          

3. Gender

0.00

0.01

1.00

         

4. Education level

0.02

−0.12

−0.10

1.00

        

5. Household income

0.03

0.00

−0.10

0.28

1.00

       

6. Ties with entrepreneurs

0.00

−0.09

−0.09

0.15

0.13

1.00

      

7. Business angel

0.02

−0.02

−0.05

0.10

0.07

0.12

1.00

     

8. Opportunity recognition

0.02

−0.04

−0.02

0.03

−0.02

0.13

0.06

1.00

    

9. Barriers to technology adoption

−0.01

−0.03

0.05

−0.12

−0.07

−0.03

−0.04

0.11

1.00

   

10. Informal economy

−0.09

−0.05

0.12

−0.02

−0.12

−0.09

−0.03

0.07

0.28

1.00

  

11. IPR

0.04

0.08

−0.04

0.13

−0.01

−0.01

−0.02

−0.02

−0.34

−0.12

1.00

 

12. FDI per capita

0.01

0.04

−0.07

0.14

0.01

−0.02

0.02

−0.04

−0.28

−0.17

0.28

1.00

Table 4

Effects on use of latest technology by entrepreneurs in emerging economies

 

1

2

3

4a

Fixed part estimates

Individual-level

 Age

 

0.99 (0.00)**

0.99 (0.00)**

−0.00 (0.00)**

 Gender (1 = male as baseline)

    

 2 (Female)

 

0.91 (0.06)

0.91 (0.06)

−0.09 (0.05)

Education level (0 = none as baseline)

 1 (Some secondary)

 

0.35 (0.30)

0.35 (0.29)

−1.13 (0.84)

 2 (Secondary)

 

0.39 (0.33)

0.39 (0.33)

−1.02 (0.84)

 3 (Post-secondary)

 

0.33 (0.28)

0.33 (0.27)

−1.19 (0.84)

 4 (Graduate)

 

0.39 (0.33)

0.39 (0.33)

−1.03 (0.84)

Household income (1 = Lower as baseline)

 2 (Middle tier)

 

1.02 (0.07)

1.04 (0.07)

0.03 (0.07)

 3 (Upper tier)

 

1.14+ (0.08)

1.18 (0.09)*

0.13+ (0.07)

Ties with entrepreneurs

 

0.91 (0.05)

0.91 (0.05)

−0.11 (0.08)

Business angels

 

1.06 (0.08)

1.08 (0.09)

0.06 (0.08)

Opportunity recognition

 

1.07 (0.06)

1.08 (0.06)

0.05 (0.06)

Country-level

 Barrier to technological adoption

 

0.54 (0.08)***

0.47 (0.07)***

−1.00 (0.20)***

 Informal economy

  

0.83 (0.30)

−3.55 (1.29)***

 IPR

  

0.68 (0.25)

−2.31 (0.81)**

 FDI per capita: (H3)

  

0.69 (0.08)**

−1.07 (0.33)***

Interaction terms

 Informal economy × IPR: (H4)

   

2.09 (0.82)*

 Informal economy × FDI: (H5)

   

0.91 (0.38)*

Random part estimates

 Variance of intercept

0.73 (0.14)

0.69 (0.14)

0.46 (0.13)

0.26 (0.10)

 % of variance explained or Rho

17.2 (0.05)

15.2 (0.04)

12.4 (0.04)

7.4 (0.03)

Model fit statistics

 Number of observations

10,431

10,431

10,431

10,431

 Number of countries

20

20

20

20

 Degrees of freedom (number of variables)

0

12

15

17

 Chi square

26.54

33.52

50.33

 Probability > Chi square

**

***

***

 Log likelihood

−4814

−4408

−4305

−4200

 Likelihood ratio test of Rho

***

***

***

***

Bold values indicate hypotheses that were statistically supported

Columns represent odds ratio (OR) instead of regression estimates. OR values greater than 1 signal positive association. OR values smaller than 1 signal negative association

Standard errors in parentheses; *** p < 0.001; ** p < 0.01; * p < 0.05; p < 0.1+; 2-tailed significances for hypotheses

aModel 4 reports the beta-coefficients of the logistic regressions and not the odds ratios. Interaction terms and their graphical representation are generated using these beta-coefficients (not the odds ratios)

4.1 Intra-class correlation (ICC)

Significant between-country variance in the dependent variable necessitates multi-level analysis (Hofmann 1997). To check this, we estimated a multi-level logistic regression (Models 1 Table 4). This yielded intra-class correlation coefficients (ICC or rho) of 17.2 % in the emerging economies set.

The ICC (or rho) value represents the proportion of variance in the dependent variable that resides between countries owing to country-level characteristics. Since the observed ICC values represent significant variance, they necessitated multi-level analyses, hence warranting looking into country-level factors that could explain this variance.

4.2 Country-level effects

Random-effect logistic regression models are reported in Table 4, namely Models 2, 3 and 4. These models report estimates for the fixed part (estimates of coefficients) and random part (variance estimates), as well as model fit statistics. Model 2 includes all individual-level as well as country-level controls, Model 3 include the three country-level variables used in the regression models and Model 4 includes the interaction terms.

4.2.1 Country-level effects in emerging economies

Model 3 of Table 4 accommodates for the four country-level variables, namely the size of a country’s informal economy, IPR and FDI. All estimates are reported as odds ratios (exponential of the beta coefficients obtained from logistic regressions), with ratios greater than one representing positive association (percentage increase) and those less than one representing negative association (percentage decrease). The effects of an increase of one standard deviation in country-level size of the informal economy and the strength of IPR were observed to decrease the likelihood of using the latest technology by entrepreneurs by 17 % (1 − 0.83) and 32 % (1 − 0.68) respectively, these effects were observed to be statistically insignificant. An increase of one standard deviation in the levels of FDI per capita in emerging economies decrease the likelihood of the use of latest technology by entrepreneurs by 31 % (1 − 0.69). This effect was observed to be statistically insignificant (p < 0.004). Combined we find support for one out of the three main-effects hypothesized in relation to emerging economies—that for hypothesis H3.

The variance component decreased from 0.69 in Model 2 to 0.46 in Model 3, suggesting that the addition of the three country-level predictors collectively explained a significant 33 % of the remaining variance in the dependent variable after the country-level and individual level controls have been accounted for.

4.3 Moderation effects

We tested two interaction effects for each of the two sets of countries. These are the interactions between (1) Informal economy and IPR and (2) Informal economy and FDI per capita in predicting the likelihood of using the latest technology by entrepreneurs in emerging economies. Informal economy is the moderator in all the two terms. The results are reported in Model 4 of Table 4 for the set representing 20 emerging economies. The two hypotheses (H4 and H5) concerning the moderation effects were tested. Given that the effect size as well as the causal directionality of the interaction terms could be best explained with graphical support, we corroborated the discussion below with Figs. 1 and 2.6
Fig. 1

Interaction between informal economy and strength of IPR regimes in emerging economies. (Low and high informal economy in Figs. 1 and 2 represent (Mean − 1.5 SD) and (Mean + 1.5 SD))

Fig. 2

Interaction between informal economy and the levels of FDI per capita in emerging economies (Low and high informal economy in Figs. 1 and 2 represent (Mean − 1.5 SD) and (Mean + 1.5 SD))

4.3.1 Informal economy as moderator in emerging economies

Figure 1 shows the interaction between the size of a country’s informal economy and the strength of IPR regime (significant at p < 0.02) and offers three important insights.7 First, we observe that in the context of emerging economies, the likelihood of using the latest technology by entrepreneurs [p(DV = 1)] is the highest when the size of the informal economy is smaller and where the IPR is weakly protected. This likelihood is about 0.49. However as the size of informal economy increases, the negative effect of IPR increases. Second, even when the size of the informal economy remains smaller in emerging economies, a strongly protected IPR regime reduces this likelihood significantly [from a p(DV = 1) = 0.49 at weaker IPR regime to a p(DV = 1) = 0.34 at stronger IPR regime]. Third, the likelihood of using the latest technology by entrepreneurs is invariably lower when the size of the informal economy is larger as opposed to when it is smaller for all levels of the strength of the IPR regime. As the size of the informal economy increases, it stifles entrepreneurs’ likelihood of using the latest technology when the strength of the IPR protection regimes is stronger. Overall, this supports hypothesis H4, which argued that use of latest technology by entrepreneurs in emerging economies would be negatively affected by the high levels of informal economy and stronger IPR.

Figure 2 plots the interaction between a country’s size of informal economy and its incoming FDI (significant at p < 0.02). We make three important observations. First, we observe that in the context of emerging economies, the likelihood of using the latest technology by entrepreneurs [p(DV = 1)] is the highest when the size of the informal economy is smaller and when the levels of FDI is low. This likelihood is about 0.48. Second, even when the size of the informal economy remains smaller in emerging economies, higher levels of FDI reduce this likelihood significantly [from a p(DV = 1) = 0.48 at lower levels of FDI to a p(DV = 1) = 0.35 at higher levels of FDI]. Third, the likelihood of using the latest technology by entrepreneurs is invariably lower when the size of the informal economy is larger as opposed to when it is smaller for all levels of FDI. As the size of the informal economy increases, it stifles entrepreneurs’ likelihood of using the latest technology for all levels of FDI. Overall, this supports hypothesis H5, which argued that FDI’s influence would negatively affect use of latest technology by entrepreneurs in emerging economies with large sized informal economy.

We also note that the variance component of the random intercept, decreased from 0.46 in Model 3 to 0.26 in Model 4, suggesting that the addition of the two interaction terms explained 43 % (((0.46 − 0.26)/0.46)*100) of the remaining country-level variance that existed in the dependent variable across 20 emerging economies after all controls and the four predictors have been accounted for. This supports that a significant proportion of the variance in the likelihood of using the latest technology by entrepreneurs could be explained based upon exclusively the moderation effects of the informal economy on national institutions across our emerging economies.

5 Discussion

Our article has the potential to address an important and underlying question—how do activities occurring outside formal institutional boundaries positively and negatively influence overall economic growth? Our study provides answers by looking into how the informal economy interacts with other institutions. While we found a significant main effect for only FDI (odds ratio 0.69), the presence of the three predictors (informal economy, IPR and FDI) in a regression model collectively explained 33 % of the remaining variance across the 20 countries included in our study after the individual-level and country-level controls had been accounted for, thereby consolidating our choice of institutions and rendering them salient predictors of the likelihood of entrepreneurs’ using the latest technologies in emerging economies. While the size of a country’s informal economy was not observed to exercise significant main effects, we observed its moderating effect on the two institutions to be statistically significant. In this regard, our results indicate that as the size of the informal economy increases in emerging economies, it exacerbates the potential negative effects of IPR and FDI.

Since institutions are expected to be distinctively different in emerging economy contexts, the findings of our study would be more meaningful and insightful if explained in relation to what we might expect in more developed countries, which typically have much smaller informal economies than emerging economies when considered as a proportion of overall economic activity (Aidis 2005; Schneider 2000). In particular, the effect of IPR and FDI may have positive effects in developed economies due to increased access and fewer crowding effects, and more effective knowledge spillovers. Similarly, the attraction of the informal economy should be weaker in developed countries, reducing the need for voluntary avoidance. In short, this study corroborates that institutions may have different, even opposite effects on technology use by early-stage entrepreneurs in developed and emerging economies, and the presence of a larger informal economy increases these differences. This may go a long way toward explaining the conflicting findings about the effect of institutions on entrepreneurship found in the extant literature. For example, Estrin et al. (2013) have found negative effects for FDI whereas Kim and Li (2012) have reported positive associations between FDI and entrepreneurship. Similarly, Acs and Sanders (2008) have suggested that IPR may have both positive and negative effects on entrepreneurs.

Our study makes several other contributions. First, we depart from the commonly used definition of technology entrepreneurship—new technology development by entrepreneurial firms—and examine the determinants of a rather under-researched yet significant phenomenon: the use of latest technologies by entrepreneurs. Second, by using a multi-level theoretical and empirical design, we examined the regulatory influence of country-level institutions on the use of latest technology by early-stage entrepreneurs, thus contributing to the literature that looks into the contextual influences on entrepreneurship. Third, we contribute by situating our study specifically in the under-researched context of emerging economies.

5.1 Limitations and scopes for future research

In spite of several contributions, the present study may have limitations that could be addressed by future research. First, while we pursue an important research question—the influence of informal economy and other institutions on the use of latest technologies in entrepreneurship—our study does not provide direct insight into the influence of the informal on the popularly held notion of technology entrepreneurship, that of development of new technology. In addition, although the main effects of informal economy and IPR were observed to be in the proposed direction, the fact that these were not statistically significant poses a limitation. With a bigger sample of emerging economies, this issue could be potentially resolved. Second, although other formal institutions such as the rule of law, governance structure, regulatory quality, financial institutions, division of labor, etc., or informal institutions such as societal-level cultural orientations, desirability of entrepreneurship and income inequality, ethnic fractionalization and polarization, immigration, etc., could have important influences on rates of entrepreneurship across countries in general and on technology use in particular, our multi-level study involving just 20 countries poses limitations on the number of higher-level (country-level institutions in our case) predictors that could be used (Mass and Hox 2005; Snijders and Bosker 1993). Third, because annual data on our predictors were missing for some years and countries (making it an even narrower and unbalanced panel) our study was limited to provide ‘non-transient’ analyses. Future studies could collect more annual data points enabling a more elaborate transient assessment that could accommodate and account for any social, economic and technological change that may have occurred between the periods of interest in this study—2002–2008. Fourth, we limited our analysis to Schneider’s measure of the informal economy. This measure has the advantage of covering a broad number of countries, thus ensuring overlap with the countries that participated in GEM. However, future research could examine other measures of informal economic activity and compare and contrast with the proxy used here to investigate if any differences arise. Finally, future research might examine the effect of the informal economy on persistence in entrepreneurship. Entrepreneurs who would have survived past the ‘early-stage’ would have formed a more realistic opinion on what to expect and how to operate in environments where the size of informal economy is relatively large. Consequently, the effects of informal economy on established entrepreneurs may not be as detrimental as are on early-stage entrepreneurs.

6 Conclusion

The use of latest technology by early-stage entrepreneurs as a dependent variable differentiates this study from most others in entrepreneurship research which tend to examine rates of entrepreneurial entry in general without much attention to the type of entry. Investigating the regulatory influences of institutions as well as of the interactions between them on use of latest available technology by entrepreneurs is timely and appropriate given that it is this type of recombination that is most likely to facilitate new technology development that eventually contributes to economic growth (Shane 2009). Not all types of entrepreneurship contribute equally (Baumol 1990) and technology-driven development is more likely to come from the application of the latest technologies. Hence, policy makers should consider that without the use of latest technologies, the economic potential of entrepreneurship may be buffered.

Footnotes
1

Data from only 20 countries were usable for all the variables and controls included in the regression models.

 
2

In unreported robustness checks, we isolated nascent and new entrepreneurs following which we replicated our analyses on each of the two samples. We did not observe significant loss of generalizability of our findings. We chose to adhere to GEM’s operationalization of “early stage” entrepreneurs as nascent or new entrepreneurs and used them in a combined sample in all our analyses.

 
3

In unreported robustness checks, we replaced the IPR index obtained from the Heritage Foundation with those from the World Economic Forum’s (2011) Global Competitive Index (GCI) Report, Fraser institute’s “Economic Freedom of the World” report (Gwartney et al. 2012) and GEM National Expert Survey. Our findings were generalizable across all of these sources but we chose to use IPR index from the Heritage Foundation because it offered data on IPR for the maximum number of countries common to the GEM survey.

 
4

We used xtlogit—STATA 12.0 as the statistical software tool.

 
5

Temporal variance in dependent variable was insignificant relative to country specific variance.

 
6

To be able to meaningfully interpret and express graphically the interaction terms we use beta coefficients of the logistic regressions and not the odds ratios.

 
7

Low and high informal economy in Figs. 1 and 2 represent (Mean − 1.5 S.D.) and (Mean + 1.5 S.D.).

 

Copyright information

© Springer Science+Business Media New York 2015

Authors and Affiliations

  • Saurav Pathak
    • 1
  • Emanuel Xavier-Oliveira
    • 2
  • André O. Laplume
    • 2
  1. 1.College of Business AdministrationKansas State UniversityManhattanUSA
  2. 2.School of Business and EconomicsMichigan Technological UniversityHoughtonUSA

Personalised recommendations