Does Rapid Growth Coincide with Innovation? An Examination of Gazelles in Sweden By Jonas Gabrielsson, Diamanto Politis and Natalyia Galan1

Abstract This paper examines the role of gazelles in the innovation economy. We use primary data collected from a survey to Swedish rapidly growing firms that have been classified as gazelles to analyze how these firms work and function in the innovation system in which they are embedded. We also analyze the extent to which they are involved in innovation and other development activities and whether their involvement is associated to continued firm growth.

Introduction The contribution of new and small companies to innovation and economic growth is today widely acknowledged. A landmark study in this stream of research was Birch (1979) who in his pioneering report showed that small companies accounted for the majority of all new jobs in the economy. Later on, Birch and Medoff (1987) argued that most job creation occurs within a relatively small number of companies, and they created the metaphorical term “gazelles” to describe these. Compared to “elephants” – large and often publicly traded companies, and “mice” - small companies that grow very little and hence contribute only marginally to job creation, the “gazelles” are vital small companies that grow quickly and create employment.

The value creating role of gazelles has fueled an interest in high-growth companies among both policy makers and researchers and lead to a growing stream of academic studies (e.g., 1

Jonas Gabrielsson is Associate Professor at CIRCLE, Lund University. Diamanto Politis is Assistant Professor at the School of Business and Engineering, Halmstad University. Nataliya Galan is a post doc researcher at the School of Business and Engineering, Halmstad University. Address correspondence to: Jonas Gabrielsson, CIRCLE, Lund University, PO. Box 117, SE221 00 Lund, Sweden. E-mail: [email protected]. 1

Henrekson and Johansson, 2010). Gazelles are generally recognized as important players in the economy as they account for a disproportionately large share of net employment growth. Moreover, their high growth is fueled by their ability to offer cheaper or better products to customers, and thereby they contribute to evolutionary processes of change and renewal in the economy by squeezing less efficient companies out of the marketplace. Hence, gazelles represent a dynamic element in the capitalist economic system that drives the very system forward.

A general conception is that rapid growth is closely associated with innovation. This conception is partly coming from success stories about pioneering and innovative firms, such as Google, and partly from theories and models that emphasize technological advancement and industrial innovation as engines for economic growth (Romer, 1986; Lucas, 1988). However, even if success stories may provide fascinating and inspiring narratives they are not very good as reference points for broader generalizations as they often represent exceptions rather than the norm. Otherwise, the story would probably lose its „magic‟. Moreover, even if there is a link between innovation and economic growth on aggregated levels in the economy, as empirical evidence often suggest (e.g., Hasan & Tucci, 2010), this does not automatically mean that rapidly growing firms are the most innovative players in the economy. Just as it can be conjectured that rapidly growing firms are among the main carriers of innovations in the economic system (Birch and Medoff, 1987) the aggregated link may also very well be the result of a more complex interaction carried out between multiple players in a complex and increasingly globalized innovation economy (Spilling & Steinsli, 2004). Hence, the overall question of whether rapid growth tends to coincide with innovation remains largely unanswered.

The whole discussion sooner or later boils down to what we mean by innovation. From a Schumpeterian perspective, innovation can be connected to evolutionary processes of „creative destruction‟ where new knowledge is introduced in the economic system through novel combinations of various input factors in an innovative manner and that subsequently destroys or removes old ways of doing things (Schumpeter, 1934). These radical innovations that revolutionize society as we know it could be in the form of new products, new production methods, new markets or new forms of organization. However, even if the Schumpeterian ideal-typical notion of innovation as „creative destruction‟ is strongly embedded in the 2

scholarly discourse many approaches conceptualize innovation as a continuum, ranging from radical to incremental innovations (e.g., Morris, Kuratko & Covin, 2008). The latter end of the innovation scale thus include things that can be classified as mere improvements, such as re-positioning or cost reductions in already existing products or services.

One can maybe criticize the idea of including improvement activities in innovation studies. However, the strength of this „radical-to-incremental‟ view on innovation is that it emphasize that some players contribute to the evolutionary process of innovation, not as pioneers that champions entirely novel ideas and business models, but as a market-sensitive player that copies what others already do and make it better, cheaper or much more efficient (Huse, Neubaum and Gabrielsson, 2005). Thus, it opens up for the idea that firms do not generally “innovate in isolation” (Edquist 1997: 7) and it also calls for the need to have a much more systemic approach for understanding the role of gazelles in relation to the on-going evolutionary process of innovation in broader society.

This call for a broader and more systemic approach is well in line with what has developed into the modern way of understanding innovation (e.g., Freeman, 1987, Lundvall,. 1992, Nelson, 1993, Edquist 1997) where it is largely conceptualized as an interactive process of cocreation embedded in a larger system of multiple actors and institutions. From this perspective, all players in the economy are embedded in the surrounding innovation system and each gradual step of development carried out by independent firms can hence be regarded as a contribution to the on-going evolution of the system as a whole. Hence, the overall question is perhaps not so much if gazelles are involved in innovation or not, but rather to what extent they are involved and which differential roles they may play in the innovation process on the systemic level.

Interestingly, there is an abundance of studies addressing these broader and much more complex issues despite the potential for generating results with policy relevant implications. Against this background, the aim of this study is to contribute with knowledge about the role of gazelles in the innovation economy. We do this by conducting an examination of how gazelles work and function in the innovation system in which they are embedded as well as the extent to which they are involved in innovation and other development activities, such as university-industry collaborations, R&D, and exporting. We also conduct statistical analyses 3

to examine whether firms that continue to grow rapidly in sales and number of employees are different when it comes to their involvement in innovation activities.

The rest of the paper is structured as follows. In the next section we present a literature review of studies that have examined the issue of innovation in relation to growth companies. Following this is the method section, where we give an overview of the sample and the survey instrument. Thereafter we present and discuss the empirical results. The paper ends with a concluding section where the main findings are highlighted.

Literature review Research on growing firms has probably been around since the dawn of business studies but the area sure received a significant boom when Birch and Medoff (1994) claimed that rapidly growing firms are responsible for most employment growth in the US economy. In his previous work, Birch (1981) had emphasized the role of small firms in creating new net jobs but now it was suggested that most jobs were created by firms that were neither large nor small. Instead, these rapidly growing firms move from being small to large very quickly and the authors accounted that their unique characteristics - great innovation and rapid employment growth – would be missed if they were to be classified by their size (Birch and Medoff, 1994: 163). The rapidly growing firms identified by Birch and Medoff (1994) were termed “gazelles” and the association with an animal species that is famous for its swiftness proved to be a powerful metaphor that quickly diffused among both academics and policy makers. To qualify as a gazelle in the Birch and Medoff (1994) study a firm had to grow at least 20 percent in revenues in each year for four years and from a base of at least 100 000 USD in revenues. This means that the firm at least must double in size over that four-year period.

Later studies have defined gazelles in many various ways. Some studies follow Birch & Medoff (1994) and consequently use the twenty percent growth per year for three or more consecutive years as the qualifying criteria (e.g., Birch, Haggerty, & Parsons, 1995; Fischer & Reuber, 2003; Nicholls-Nixon, 2005). Others studies use fifty percent growth during each of three consecutive financial years (e.g., Autio, Arenius, & Wallenius, 2000; Acs, Parsons, & Tracy, 2008) and there has also been some studies that include the top growing five percent or 4

ten percent of rapidly growing firms in their samples (Kirchhoff, 1994; Parker et. al., 2005; Hölzl & Friesenbichler, 2008). Among other things, the different criteria for identifying gazelles make more detailed comparisons of findings between different studies relatively hard.

Moreover, even if gazelles are a term that is widely diffused there are also variations in how scholars refer to rapidly growing firms in the different studies. Some authors refer to highimpact firms (e.g., Acs, Parsons & Tracy, 2008) due to their disproportionately large impact on employment growth, revenue growth, and productivity. Others simply refer to „fast growing firms‟ or „high-growth firms‟ in their studies. Sometimes these terms are even used interchangeably throughout the different studies.

Given the variation in definitions and approaches in the field we have found it impossible to summarize a coherent and unified body of literature that can serve as the point of departure for our study. In the remainder of this section we will instead provide an overview of relevant research that has studied firm growth from various perspectives. More specifically, we will first provide a summary of studies that has tried to describe the general characteristics of rapidly growing firms. Thereafter we are reviewing studies which have studied the intersection between firm growth and innovation. We will then end our literature review with a reflection of the current state-of-the art with respect to our field of study.

General characteristics of rapidly growing firms Research on rapidly growing firms does not provide a very coherent picture of generalities that characterize them. In fact, there are great variations in results across different studies. One issue that has been under study is whether small and new firms have an above average growth potential. Supporting this view, Bishop, Mason & Robinson (2009) report that the most striking characteristics of UK high-growth firms were their relative youth - half of them were ten years or more. Moreover, Davidsson, Kirchhoff, Abdulnasser & Gustavsson (2002) examine Swedish firms with twenty or more employees and find that younger and smaller firms grow more. As such, both small size and young age seems to have an influence on the likelihood to experience more rapid firm growth

5

These findings can however be contrasted against Acs et al. (2008) who report that rapidly growing firms in the US come in all sizes but they also find that these firms are relatively old, with an average of 25 years. The general conception that rapidly growing firms are primarily young and small does not seem to hold - at least not in the US context. Acs et al. (2008) moreover show that rapidly growing firms are rare, representing between 2 and 3 percent of all firms. At the same time, however, they account for almost all of private sector employment and revenue growth in the economy. These findings corroborates with Almus (2002) study of fast-growing firm in Germany, where it is showed that only a handful of firms increase their employment significantly over time. The findings are also in line with Tether & Massini (1998) who report that a small number of UK firms are responsible for the majority of jobs created.

Overall, the findings reported above by Acs et al. (2008) and others provide a picture of high growth firms being a very small but highly potent segment of the economy. This pattern is however not universal across contexts. Moreover, in a study of Swedish high-growth firms Davidsson & Henrekson (2002) report that their contribution to job creation is quite modest. They were also in their analyses unable to find a small group of “elite” firms that collectively account for a substantial share of total job creation. The findings from countries such as Sweden thus appear to be in sharp contrast with results reported from the US and the UK.

Another issue that has been addressed is whether firm growth can be related to specific industries. Given the focus on knowledge-intensive industries in the European policy landscape one could for example expect that it is in such industries where rapidly growing firms could be find. However, Almus (2002) report that technology-intensive manufacturing branches and knowledge-based business-related services were not generating the majority of fast-growing firms in Germany. In a study reported by Bishop et al. (2009) it was moreover shown that high-growth UK firms were showing a disproportionate involvement in business services. Thus, there is not much evidence to back up the claim that knowledge-intensive industries are at the core of growth.

However, Davidsson et al (2002) provide evidence that firms should be in a growing industry in order to grow, which among other things suggest that rapid growth primarily is a market driven process. This can be related to the findings of Littunen & Tohmo (2003) who in 6

their study of small firms operating in metal-based manufacturing and business services in Finland show that the most successful firms in terms of growth were characterized by the ability to be flexible and make changes in their production process to complement an active market development strategy.

Overall, it seems hard to make any strong conclusions with respect to the characteristics of rapidly growing firms across spatial and temporal contexts. The next section will continue with a review of studies in the intersection between firm growth and innovation.

Firm growth and innovation The idea that innovation is associated with firm-level growth is strongly embedded in the literature. The direction between innovation and growth is however double-sided. On the one hand, innovation creates new ideas and competencies that can be commercialized and exploited on the market. On the other hand, successful exploitation provides revenue streams that support necessary knowledge inputs and qualification requirements to carry out further innovation activities. Simply speaking, success may breed success and discussions about firm growth and innovation are thus much like the story about the hen and the egg: what came first?

The potential problems associated with the double-sided relationship between innovation and growth has not hindered scholars to investigate innovation as a driver for growth. For example, Van Reenen (1997) used a sample of firms collected from the London Stock Exchange and from the SPRU innovation database to study growth in British manufacturing firms and he found that innovations have a positive and significant effect on employment growth. Tether & Massini (1998) studied innovative small firms in the UK and found that they have grown at a significantly faster average rate than small firms in general. However, in a ten year period none of the fastest growing firms in his sample had grown into very large companies with over a thousand employees. Thus, although innovation seems to promote growth there seems to be a threshold where

Roper (1997) conducted a multi-country survey-based study of Irish, German and UK manufacturing firms. On the whole, he found that innovative small firms grew significantly faster than non-innovators. The way this was achieved, however, varied among the studied 7

countries. In Germany, growth was achieved by a product innovation strategy which sharply increased productivity but reduced employment. On the other hand, UK and Irish small firms adopted a more balanced approach where their innovative behavior was associated with increases in both employment and productivity. Further comparisons of how product innovation was organized indicated that German small firms adopted a less market-oriented, less risky, and more formally organized approach than their UK and Irish counterparts. Thus, there seems to be country specific variations in

There have also been studies that have studied innovation and growth in relation to firm survival. Parker, Storey & van Witteloostuijn (2010) studied British gazelles and found that innovation based strategies had no effect on their subsequent likelihood of survival, with one exception. Gazelles that developed new products for market introduction after 1996 were less likely to survive and they were moreover less likely to be acquired than liquidated. Thus, innovation based strategies seems to have become increasingly risky.

In another study that have studied innovation and growth in relation to firm survival, Audretsch (1995) report from a longitudinal study of US manufacturing firms that the likelihood of new firms to survive over a decade is lower in industries where innovative activity, and especially the innovative activity of small firms, plays an important role. At the same time, those new firms that were able to survive were the ones that exhibited higher growth rates. Thus, it seems that new firms

Freel (2000) compared innovators with non-innovators in a survey-based study of SMEs in West Midlands in the UK where he finds that innovators are likely to grow more than noninnovators. However, this did not apply to growth in profits where no statistically significant differences were found. In another study, Freel & Robson (2004) report from a survey-based study of firms in Scotland and Northern England and show a positive relationship between novel product innovation and employment growth. There was however notable variations based on type of industry. For firms in manufacturing industries there was a negative relationship between product innovation and growth in sales or productivity, at least in the short term. This was the case for both incremental and novel innovations. By contrast, growing sales and productivity appeared to be positively associated with incremental process introductions in service firms. 8

These findings can be related to Oke, Burke & Myers (2007) who report from a survey to small firms with ambitions to grow in the UK, showing that they tend to focus more on incremental than radical innovations. Furthermore, this focus was related to growth in sales turnover. Interestingly, they point out that their findings corresponds with an earlier study published by Enterprise report (1994) which showed that 88% of the fastest growing and most profitable small firms achieved growth by selling more of their existing products/services to their existing market. Hence, according to these studies fast growth is not primarily fuelled by radical innovation based strategies but rather more incremental approaches.

Hölzl (2009) put further light on the relationship between firm growth and innovation. Using CIS data from 16 countries, he shows that both R&D and the share of products new to the market are much more important for gazelles in countries close to the technological frontier than in countries that are further away from the technological frontier. Hence, high-growth small firms are only more innovative than other small firms in countries close to the technological frontier. For the other country group the results are not statistically significant.

Another related issue is the association between R&D and firm growth. Stam & Wennberg (2009) conducted mail survey and telephone interviews with start-ups in the Netherlands. They found that initial R&D increased levels of inter-firm alliances in the first post-entry years, enabled the exploitation of external knowledge and stimulated new product development later on in the life course of high-tech firms. However, initial R&D did not affect the growth rate of low-tech firms, which rather seemed to be driven by the growth ambitions of the founding entrepreneur. Moreover, initial R&D spurs the growth only for the 10% fastest growing firms.

In a longitudinal study of Italian manufacturing firms, Del Monte and Papagni (2003) report that the sales growth rate was higher for firms with R&D than firms without R&D. The growth were moreover positively correlated with research intensity and the effect of research on firm growth were greater in the traditional sectors than in the sectors with high research intensity

9

Wöhrl, Hüsig & Dowling (2009) conducted a study of technology-based growth companies that went public in Germany between 1997 and 2002. They report a positive relationship between R&D intensity and firm sales growth. The relationship is however curvilinear, where there is a negative impact of high R&D spending on returns on sales. Over time the positive effect of R&D on sales growth decreases and even turns negative, and the negative impact of R&D intensity on return on sales fade with increasing company age.

Burgel, Fier, Licht & Murray (2002) use survey data collected from high-tech start-ups in Great Britain and Germany to examine the relations between internationalization and R&D. In their study they report that early internationalization has positive effects on sales growth and competitiveness, but not on employment growth. International sales are not a necessary condition for fast growth. R&D firms are estimated to have a 5-6% higher sales growth compared to both non R&D firms and firms who only occasionally conduct R&D. As with internationalization, R&D thus has large positive effects on sales growth but no effect on employment growth.

Summary and Specification of Research Questions Our literature review has shown that although substantial research exists that have addressed the issue of firm growth in general and rapidly growing firms in particular, we have not been able to summarize a coherent and unified body of literature that can serve as a theoretical point of departure for our study. Another reflection is that there has been very little attention to the role and contribution of rapidly growing firms in the broader innovation system in which they are embedded. Given our focus in this study, it thus seems fair to argue that the current state of the research field does not allow us to formulate any strong hypotheses. Instead, guided by our research interest and the literature just reviewed the following issues have been selected for further investigation in order to contribute with development at both the theoretical and empirical level:

Q1: What are the characteristics of Swedish gazelles?

Q2: How do Swedish gazelles work and function in the innovation economy?

10

Q2: To what extent is involvement in innovation activities among Swedish gazelles associated with continued rapid growth?

Methodology Sample and data collection To create a sample of gazelles we used a list of firms that have been recognized as experiencing rapid growth by the Swedish business newspaper Dagens Industri. To be included in this list a company must meet the following criteria: i) at least four full years of operations with a positive net result every year, ii) total sales of more than 10 million SEK, iii) at least 10 employees, iii) doubled their sales the past three years, iv) have “healthy” finances, and v) have grown organically (not through M&As). These criteria for qualifying as a rapidly growing firm are widely acknowledged in the Swedish business world.

To further limit the sample size, and control for spatial and temporal conditions, we included only companies that operate in the Öresund region in the period 2001 to 2007. The selection criteria led to a total sample of 544 companies. Contact information was found on the website of each company.

The survey instrument The primary instrument for data collection was a questionnaire survey. The measures used in the questionnairewere derived from a careful review of previous theoretical and empirical work on innovation and growth in small companies and their role and contribution in the broader innovation system in which they are embedded (e.g., Spilling and Steinsli, 2004). Secondary data was collected from annual reports, web sites and other documents describing the status and development of the companies.

Following the first send-out of the questionnaire and two reminders we received questionnaires with complete responses from 206 companies, which correspond to a response rate of 37.9 percent. Chi -square and t-tests were then conducted to assess whether there were any significant differences between respondents and non-respondents. We could not find any differences with respect to variables such as industry belongingness, firm age, firm size, growth rate and year when they were identified as gazelles. Thus, the sample of respondents 11

appears to be representative of the population from which it was drawn and, on these criteria, we have no reason to suspect that there are any response biases in our sample.

Results General characteristics of the firms in the study A description of the gazelles is presented in table 1 below. The responding firms in the study have a mean age of 19.8 years, which corresponds well with previous studies that has reported that rapidly growing firms are relatively old (Acs et al., 2008). Mean sales in 2007 were 87 MSEK but as can be seen in Table 1 this value is highly skewed. The same goes for the number of employees where the mean is 40.4. There is thus a great variety in terms of size among the gazelles both in terms of total sales and employees. In all, we can find no support for that gazelles on average are young and small.

Interestingly, the gazelles have on average a greater part of their sales coming from the local market rather than from international sales. This suggests that many gazelles to a large extent are oriented towards serving local market needs. Moreover, further examination of the data show that regional sales and international sales are negatively associated, at p > .001. This suggests that the gazelles that focus on international markets are less oriented towards the local market.

With respect to industry belongingness, we find that the gazelles in the sample are spread across a wide range of different industries. Although firms all industries to some extent exploit knowledge the bulk of firms can hardly be classified as technology or knowledgeintensive firms according to the definition used by Eurostat. Only six percent of the gazelles operate in industries that can be classified as medium-to-high or high technology intensive manufacturing. Moreover, twenty two percent, or about 1/5 of the gazelles, operate in knowledge-intensive services. This leaves seventy two percent of the firms in other industries. We could not find any significant association between knowledge-intensiveness and their market orientation.

Studies of innovation systems have emphasized the important role of corporations, universities or other R&D-intensive organizations as sources of potential growth firms in the regional economy (Oakey 1995; Lindholm Dahlstrand, 1999). However the link between 12

existing organizations and the gazelles in our study is relatively modest. Sixteen percent of the gazelles are spin-offs from other companies and only two percent are spin-offs from universities. An examination of these latter university spin-offs revealed that they originated from Lund University (three cases) and Uppsala University (one case).

About half of the respondents, or 49 percent, could be classified as family businesses. A closer examination of these firms showed that they were negatively associated with international sales, at p > .05, and negatively associated with company spin-offs, at p > .05. They were moreover negatively associated with knowledge-intensive services, at p > .01. Table 1: Characteristics of responding firms

Firm age No of employees in 2007 Sales in 2007 (MSEK) % of regional sales % of international sales Family business Technology intensive manufacturing Knowledge intensive services Company spin-off Academic spin-off

Mean 19.8 40.4 87.0 49.8 16.9 .49 .06 .22 .16 .02

Std dev 13.7 50.1 132.5 38.2 27.9 .50 .25 .42 .37 .14

How Swedish gazelles work and function in the innovation economy One way of analyzing how Swedish gazelles work and function in the innovation economy is to see how they collaborate with knowledge-intensive organizations in the region, such as universities. In Table 2 below we report the percentage of responding firms that were involved in collaborations with a university in the region. Here we can observe that the majority of the firms in our sample have no collaborations at all. Among the different types of collaborations with local universities that are reported it is evident that informal collaboration types dominates. The most common thing is to recruit new employees, where 15.8 percent claims that they do this. Only just above ten percent claim that they are involved in discussions with personnel or students from universities in the region. As can be seen, more formal collaborations such as assigned R&D tasks, consulting tasks and joint R&D projects are much rarer and lay around five percent. In all, collaborations between gazelles and universities are hence quite modest. 13

Table 2: University-industry collaborations No collaboration Collaboration, but not in region Assigned R&D tasks Consulting tasks Discussions Joint R&D projects Recruitments

Percent 65.2 4.9 3.4 6.4 11.3 3.9 15.8

A closer examination of the data shows that it firms that operate in knowledge-intensive service industries that can be associated with recruiting new employees and discussing with personnel or students, both at p > .01. Moreover, firms that operate in technology intensive manufacturing are more likely to be involved in collaborations with universities outside the region, at p > .01.

Another way of analyzing how Swedish gazelles work and function in the innovation economy is to see how they compete on the basis of products or services which are in various stages of their “life cycles”. Our life cycle analysis is based on the work of Covin & Slevin (1990) and Klepper (1996) and is based on the idea that firms may offer products or services which, for example, are new and innovative, familiar and widely used, or old and possibly approaching obsolescence. It should be acknowledged that the notion of a life cycle is an industry-level concept. This means that if a young firm introduces a new product that is in the maturity stage of its life cycle it would fall into the maturity category. Accordingly, following Covin & Slevin (1990, p. 127) the firms in our sample were assessed based on the percentage of their current total sales that were accounted for by products or services in each of the following life cycle stages:

1) Introduction stage, defined as when products or services are unfamiliar to many potential users and industry-wide demand for these products or services is just beginning to grow.

2) Growth stage, defined as when total industry-wide demand for products or services is growing at a rate of 10% or more annually.

14

3) Maturity stage, defined as when products or services are familiar to the vast majority of prospective users and industry-wide demand for these products or services is relatively stable.

4) Decline stage, defined as when total industry-wide demand for products or services in is decreasing at a more or less steady rate.

While we find gazelles in all industry life stages, we could in our analysis observe a concentration of sales from products or services in the maturity stage among the gazelles. The average sales in the different life cycle stages across all firms were 7.8 percent from products or services in the introduction stage, 22.9 percent from the growth stage, 62.1 percent from maturity stage and 5.9 percent from decline stage. A closer examination of the data shows no significant associations between sales in any of the industry life stages and the age or size of the gazelles. Interestingly, the only association was that firms in technology intensive manufacturing industries were more likely to receive sales from products in the decline stage, at p > .01.

Firms can also be classified according to the main functions they perform system, which signals their role in the innovation system from a value chain perspective (Spilling & Steinsli, 2004). The idea underlying this analysis is that different firms tend to specialize in certain fields. Table 3 below report the results. Table 3: Firms classified according to functional roles Standard products to end-user Specialised products to end-user Subcontracting R&D-services Software and systems development Data and information services Consulting services Wholesale

Percent 57.4 52.5 43.6 3.9 6.9 6.4 29.4 17.2

From the table it can be observed that standard products to end-users, specialized products to end-users, and subcontracting are the three most common functional roles performed. This is followed by consulting services and wholesale, with 29.4 percent and 17.2 percent respectively. Although a few cases exist, the bulk of gazelles are not involved in providing 15

business services such as R&D-services, software and systems development, and data and information services. Not surprisingly, it is firms that operate in knowledge intensive industries that perform most of these functions. Technology intensive firms were found to be overrepresented in the provision of R&D-services, at p> .05. Moreover, firms that operate in knowledge-intensive service industries are significantly overrepresented in providing software and systems development as well as data and information services, at p> .05. Adding to this, these firms are also significantly overrepresented in providing consulting services, at p> .01.

A characteristic feature of the economic system that often is emphasized in literature and research on innovation is that it is in constant evolution (Edquist, 2005). For example, Schumpeter (1934) characterized the capitalist economy as a dynamic system that undergoes constant change as innovations are introduced and diffused, which disturbs its “circular flow”. Accordingly, based on the identification of evolutionary processes in Spilling & Stensli (2004) we analysed the extent to which the gazelles in our sample has been involved in, or contributed with, different kinds of evolutionary processes in the innovation system. The share of gazelles reporting different types of evolutionary processes is presented in Table 4 below. Table 4: Share of firms reporting different type of evolutionary processes Merged with other company Aquired other companies Licensed in the right to other production Developed and licensed out production rights Contributed to a new start-up Employees have left to start a new business

Percent 18.1 8.3 20.6 32.8 7.8 24.0

Our data does not show a clear overall pattern. Rather, the table illustrates a relatively complex pattern of development, constituted by a number of different evolutionary processes. From the table we can for example observe that almost one fifth of the gazelles have merged with another company in the past. Moreover, the phenomenon of licensing in or out production rights has been the case for between twenty and thirty percent of the gazelles. Furthermore, while few of the gazelles has contributed directly to new start-ups we can see that as many as 24 percent of the firms have had employees that have left to start up a new business on their own. Hence, on the aggregate the pattern is complex and fairly varied.

16

Another issue related to our focus in this study is the extent to which gazelles are involved in different innovation and developmental activities during the past year. This is reported in Table 5 below. Table 5: R&D and innovation activities Own R&D activity Aquired R&D-services New production process Developed new products or services Improved existing products or services Improved existing production process Innovation in marketing/sales Developed new markets Applied new technology Apllied new software or system solutions

Percent 18.1 5.9 16.7 41.2 51.2 25.5 37.4 51.2 20.1 15.7

Here it can be observed that gazelles to a relatively high extent are involved in innovation activities. A majority of gazelles have for example been involved in improving existing products or services and developing new markets, both activities involving more than 50 percent of the responding firms. This corresponds well to findings in the literature that rapidly growing firms are market sensitive players that often exploit their ability to offer incremental innovations to expanding segments of customers (Littunen & Tohmo, 2003; Oke, Burke & Myers, 2007).

Moreover, we classified the gazelles by examining the degree of novelty in their product or service offer. Here we found that the large majority, 74.7 percent of the firms, were offering products or services that are similar to other offerings on the market. Another 22.8 percent were offering products or services that represent significant developments, so called "next generation" products or services. The remaining 2.5 percent were offering products or services that have never previously been sold and that could contribute to new markets or lead to changes in the character of existing markets. On the whole, it corroborates our conclusions that gazelles are primarily focused on growing by selling more of their existing products or services.

17

Innovation and continued firm-growth One of our addressed research questions is whether we can observe any association between certain innovation activities performed and continued firm-growth. For this, we calculated growth in sales and employees for the three consecutive years following their period of data collection through the questionnaire. Firms that had been acquired or been put inactive during the period were removed from the analysis. Following the generally accepted criteria of 20 percent growth for the following three consecutive years (e.g., Birch and Medoff, 1987) we classified firms as rapidly growing firms both in terms of total sales and in number of employees. Of the total sample, we identified 34.6 percent of the firms as having achieved high sales growth and 28.5 percent of the firms having achieved high employee growth the following three consecutive years. These two groups were then independently compared to the corresponding group with the rest of the active firms by Chi2-tests. The results are shown in Table 6 below. Table 6: Innovation activities and firm-growth

Own R&D activity Aquired R&D-services New production process Developed new products or services Improved existing products or services Improved existing production process Innovation in marketing/sales Developed new markets Applied new technology Apllied new software or system solutions

Sales growth High Low 0,11 0,20 0,06 0,07 0,16 0,18 0,39 0,40 0,49 0,51 0,21 0,51 0,36 0,38 0,31 0,31 0,16 0,25 0,15 0,17

Chi2-value 2.376 0.069 0.191 0.036 0.069 1.339 0.044 0.009 2.076 0.132

Employment growth High Low 0,08 0,21 0,04 0,08 0,13 0,19 0,36 0,43 0,50 0,51 0,21 0,28 0,40 0,37 0,26 0,33 0,17 0,23 0,15 0,17

Chi2-value 4.853* 0.881 0.831 0.770 0.019 0.988 0.101 0.785 0.899 0.132

As can be seen, we are unable to observe any differences between the high and low-growth groups with respect to innovation activities, with one exception. Firms in the low-growth group was to a higher extent conducting own R&D activity. However, this is even contrary what could be expected from the literature (e.g., Del Monte & Papagni, 2003; Wöhrl, Hüsig & Dowling, 2009). Hence, except for this anomaly we could find no evidence of any association between innovation activities and continued firm-growth.

Summary and conclusions In this study we have examined the role of gazelles in the innovation economy. Relying on primary data collected from a survey to Swedish rapidly growing firms that have been 18

classified as gazelles we have analyzed how these firms work and function in the innovation system in which they are embedded. We have also analyzed the extent to which they are involved in innovation and other knowledge-intensive development activities and whether their involvement can be associated with continued firm growth.

Our results indicate that the gazelles in the sample are relatively old with a mean age of 19.8 years. Moreover, mean sales were 87 MSEK and the mean number of employees was 40.4. However, these mean values were highly skewed and should be interpreted with care. Our results also show that the gazelles in the sample are spread across a wide range of different industries. Only a handful of firms could be classified as operating in technology intensive manufacturing industries while about one fifth of the firms could be classified as operating in knowledge intensive industries. We also found that collaborations between gazelles and universities are quite modest.

With respect to their offerings there was a concentration of sales among the gazelles from products or services in the maturity stage among the gazelles. Here, were standard products to end-users, specialized products to end-users, and subcontracting the three most common functional roles performed. The analysis of the role of gazelles in economic evolution moreover shows a relatively complex pattern of development, constituted by a number of different evolutionary processes. Moreover, we observed that gazelles to a relatively high extent are involved in innovation activities. The majority of the studied firms had been involved in improving existing products or services and developing new market. The innovations can however be classified as incremental rather than radical ones as the majority were offering products or services that are similar to other offerings on the market. Furthermore, we cannot find any associations between innovation activities and continued firm-growth, something which may suggest that there are other factors at play in this process. In all, our findings provide unique insights into the role of gazelles in the innovation economy which provide avenues for further studies.

We believe the current study to be original in at least four senses. First, we have identified and sketched out the issue of how rapidly growing firms work and function in the innovation system in which they are embedded as a neglected area of research. Studies of rapidly growing firms has in this respect traditionally been conducted by entrepreneurship scholars 19

while innovation scholars dealing with innovation systems has failed to identify rapidly growing firms as a group of firms that may deserve special attention. Hence, we believe this area to be a highly potent avenue for future research as it has the potential to combine two strands of literature which so far have been conducted in parallel. Second, we present primary data collected from an original research instrument. As such, the study is presenting unique data that is not available in existing databases or other public records. Third, using this research instrument we examine issues such as R&D behavior, technology transfer and innovation activities in Gazelles. As such, we provide an insight into how high-growth companies perform with respect to such developmental activities. Fourth, we connect our results to current discussions of rapidly growing firms as well as to literature and research on innovation systems. To the best of our knowledge, this has not been done in any systematic sense in prior research. In sum, we hope the above mentioned contributions will inspire scholars to conduct further studies about the role of gazelles in the innovation economy.

References Almus, M. (2002) What characterizes a fast growing firm? Applied Economics, 34 (12): 1497–1508. Audretsch, D.B. (1995) Innovation, growth and survival, International Journal of Industrial Organisation, 13 (4): 441–457. Autio, E., Arenius, P. & Wallenius, H. (2000) Economic impact of gazelle firms in Finland, Helsinki University of Technology. Acs, Z., Parsons, W., & Tracy, S. (2008) High-impact firms: gazelles revisited, in S.O.O. Advocacy (Ed.), SBA Reports. Washington, D.C.: SBA Office of Advocacy. Available at www.sba.gov/advo/research Birch, D. L. (1979) The job generation process, Cambridge, MA: MIT program on neighborhood and regional change, Massachusetts Institute of Technology. Birch, D. L. & Medoff, J. (1994) Gazelles, In L. C. Solmon & A. R. Levenson (Eds.), Labor markets, employment policy and job creation. Boulder, CO: Westview. pp. 159–167. Bishop, K, Mason, G. & Robinson, C. (2009) Firm growth and its effects on economic and social outcomes: Literature and statistical review, Report to the National Endowment for Science Technology and the Arts. London: NIESR. Coad, A., & Rao, R. (2008) Innovation and firm growth in high-tech sectors: A quantile regression approach, Research Policy, 37, 633–648. 20

Covin, J.G. & Slevin, D.P. (1990) New venture strategic posture, structure, and performance: An industry life cycle analysis, Journal of Business Venturing, 5(2): 123-135. Davidsson, P., & Henrekson, M. (2002). Determinants of the prevalence of start-ups and high-growth firms, Small Business Economics, 19(2), 81-104. Haggerty, A & Parsons, W. (1995) Who’s Creating Jobs?, Cognetics Inc., Boston. Davidsson, P., Kirchhoff, B., Abdulnasser, H-J. & Gustavsson, H. (2002) Empirical analysis of business growth factors using Swedish data, Journal of Small Business Management, 40, 332–349. Del Monte, A. & Papagni, E. (2003) R&D and the growth of firms: empirical analysis of a panel of Italian firms, Research Policy, 32, 1003–1014. Edquist, C. (2005) Systems of innovation: Perspectives and challenges, in Fagerberg, J., Mowery, D.. & Nelson, R. (red.). The Oxford Handbook of Innovation, Oxford University Press, New York, pp. 181-208. Entreprise Report (1994) Growth in the 1990s: winners and losers, 3i/Cranfield European Enterprise Centre, Cranfield. Freel, M. (2000). Do small innovating firms outperform non-innovators? Small Business Economics, 14 (3), 195–210. Freel, M. S., & Robson, P. J. (2004). Small firm innovation, growth and performance. Evidence from Scotland and Northern England. International Small Business Journal, 22 (6), 561–575. Freeman, C. (1987) Technology policy and economic performance: lessons from Japan. Pinter, London. Fischer, E. & Reuber, A.R. (2003), Support for rapid-growth firms: A comparison of the views of founders, government policymakers, and private sector resource providers, Journal of Small Business Management, 41, 346–365. Edquist C. (1997) Systems of innovation approaches - their emergence and characteristics, in Edquist, C. (ed.) (1997) Systems of innovation: technologies, institutions and organizations, London: Pinter/Cassell . Hasan, I. & Tucci, C.L. (2010) The innovation-economic growth nexus: Global evidence, Research Policy, 39(10), 1264-1276. Henrekson, M., & Johansson, D. (2009). Gazelles as job creators: a survey and interpretation of the evidence. Small Business Economics, 35 (2), 227-244. Huse, M., Neubaum, D.O., & Gabrielsson, J. (2005). Corporate Innovation and Competitive Environment. International Entrepreneurship and Management, , 1(3), 313-333.). 21

Hölzl, W. (2009). Is the R&D behaviour of fast-growing SMEs different? Evidence from CIS data for 16 countries, Small Business Economics, 33(1), 59-75. Hölzl, W., & Friesenbichler, K. (2008). Final sector report gazelles. In Europe Innova (Ed.), Sector Reports. Vienna: WIFO. Kirchhoff, B.A. (1994) Entrepreneurship and dynamic capitalism, Westport, Conn.: Praeger. Klepper, S. (1996) Entry, exit, growth, and innovation over the product life cycle, American Economic Review, 86(3), 562-583. Lindholm Dahlstrand, Å. (1999) Technology-based SMEs in the Göteborg region: Their origin and interaction with universities and large firms, Regional Studies, 33(4): 379-389. Littunen, H., & Tohmo, T. (2003). The high growth in new metal-based manufacturing and business service firms in Finland. Small Business Economics, 21 (2), 187–200. Lundvall, B-A., (Ed) 1992 National Systems of Innovation. Towards a Theory of Innovation and Interactive Learning. Pinter, London. Morris, Kuratko & Covin, J.G. (2007) Corporate entrepreneurship and innovation: entrepreneurial development within organizations, Fort Worth, Tex.: Harcourt College Publishers. Nelson, R. (Ed.) 1993 National innovation systems. A comparative analysis. Oxford Univ. Press, New York. Oakey, R (1995) High technology new firms: Variable barriers to growth, London, Paul Chapman Publishing Ltd. Oke, A., Burke, G., & Myers, A. (2007). Innovation types and performance in growing UK SMEs. International Journal of Operations & Production Management, 27 (7), 735–753. Parker, S. C., Storey, D. J., & van Witteloostuijn, A. (2010). What happens to gazelles? The importance of dynamic management strategy. Small Business Economics, 35, 203–226. Romer, P.M, (1986) Increasing returns and long-run growth, Journal of Political Economy, University of Chicago Press, vol. 94(5), pages 1002-37 Roper, S. (1997). Product innovation and small business growth: A comparison of the strategies of German, UK and Irish companies. Small Business Economics, 9 (6), 523–537. Lucas, R. (1988) On the mechanics of economic development, Journal of Monetary Economics, 22(1), 2-41. Spilling, O. R. & Steinsli, J. (2004) On the role of high-technology small firms in cluster evolution, in Oakey, I R. During, W. & Kauser, S. (eds.) New technology based firms in the new millennium, Vol. III, Oxford, Elsevier. 22

Schumpeter, J.A. (1934) The theory of economic development, Cambride: Harvard University Press. Stam, E., & Wennberg, K. (2009). The roles of R&D in new firm growth. Small Business Economics, 33, 77–89. Tether, B.S., & Massini, S. (1998). Employment creation in small technological and design innovators in the UK during the 1980s. Small Business Economics, 11, 353–370. Van Reenen, J. (1997). Employment and technological innovation: evidence from UK manufacturing firms. Journal of Labor Economics, 15 (2), 255–284. Wöhrl, R., Hüsig, S., & Dowling, M. (2009). The interaction of R&D intensity and firm age: Empirical evidence from technology-based growth companies in the German “Neuer Markt”. Journal of High Technology Management Research, 20, 19–30

23