The Impact of FDI and Group Firms on Innovation.
John Hudson, Rudolf Sivak and Anetta Caplanova
ABSTRACT
We analyse regional spillover effects from FDI oninnovation in other firms. We find evidence for a nonlinear impact of foreign firms on innovation. An increasing regional foreign presence at first reduces innovation and then beyond a certain point increases it. This is consistent with a literature which suggests either effect may be plausible. However, potentially more important than being ‘foreign owned’ is being part of a group of firms, whether domestic or foreign. The impact is again nonlinear, but this time is such that increasing regional group presence consistently increases innovation. Finally, with respect to foreign owned firms themselves, they are more likely to engage in product innovation and license products from other firms. The same is again true of group firms. In addition, group firms are also more likely to engage in R&D. We also find evidence for the positive impact of bank credit and good governance on the innovation activities of firms.
JEL: D83, O31, F21.
Key Words: innovation; FDI; knowledge diffusion, group firms
Professor John Hudson, Department of Economics, University of Bath, Claverton Down, Bath, BA2 7AY, United Kingdom.
Professor Rudolf Sivak, Department of Finance, University of Economics in Bratislava, Dolnozemska cesta 1, 85235 Bratislava, Slovakia.
Assoc. prof. Anetta Caplanova, Department of Economics, University of Economics in Bratislava, Dolnozemska cesta 1, 85235 Bratislava.
Corresponding author: John Hudson,Email: , Tel: 0044 1225 385287, 0044 1225 383423
We are grateful for the extremely helpful comments of an anonymous referee as well as the helpful comments of the referee at the Midwest Finance Association Annual Meeting 2011, where earlier version of the paper was presented.
The Impact of FDI and Group Firms on Innovation.
1Introduction
Innovation is critical to the global economy, individual nations and regions within nations, as the world seeks to emerge from the effects of the global economic crisis. It can occur in all industries, even the oldest such as agriculture, where substantial progress is being made with the introduction of new crops. Innovation was at the heart of Schumpeter’s fundamental contributions to economics. Yet the argument can be made that it is an under-researched subject. Under-researched in the sense that the empirical work related to innovation has been in general either indirect, relating for example to productivity, or to aspects of innovation such as patents or R&D expenditure which do not reflect innovation in all its guises. This the OECD (2010) describes as a ‘serious distortion’; which omits much that plays central roles in innovation. Patents relate more to inventions rather than innovations, are country non-specific and do not cover all innovations. R&D expenditures may also be inappropriate because not all innovations are generated by R&D expenditures (Gorodnichenko et al. 2010).Even more important with respect to emerging economies, much innovation relates to imitation and adaptation of already created and tested innovations (World Bank, 2008).
Innovation can be viewed as a two stage process, whereby first knowledge is generated and diffused and secondly knowledge is turned into innovation. Key to the second process is the entrepreneur, although this does not mean that knowledge generation and innovation does not take place in large firms or long established firms, it clearly does. There are a number of barriers to knowledge generation and diffusion which include skills, distance, and finance. Most knowledge is generated in a limited number of countries[1], with developing countries generating little and thus shouldfocus on knowledge absorption (World Bank2008) The countries in our analysis, fall somewhere between these two cases, but are probably nearer to being knowledge adaptors and absorbers rather than generators. Apart from knowledge diffusion, there are also potential barriers to entrepreneurship which the literature suggests include governance, competition policies, skill deficits, finance, demand, culture and attitudes.
The empirical literature as well as a related theoretical literature have reached something of a consensus that FDI is an important potential facilitator of knowledge transfer. That is,itmay mitigate some of the barriers particularly to knowledge diffusion. There is, however, some disagreement on to what extent it is fulfilling this potential. There is also the possibility that FDI in increasing competition impacts adversely on domestic firms, thus reducing innovation. Compared with the literature on FDI, there has been less analysis of the impact of being part of a larger group of firms, whether foreign or domestic, on innovation. But in many ways they present the same potential for access to knowledge as do foreign firms per se, and being domestic they may be more likely and able to diffuse that knowledge to other domestic firms. The literatureon the impact of FDI has, until recently at least, beenprimarily focused on advanced countries in western Europe and North America and also based on limited measures of innovation. The extent to which any conclusions actually apply to other countries, emerging and developing, is open to question as is the extent to which they are applicable to more general measures of innovation.Over the past decade these limitations are gradually being addressed by a body of research, much of which makes use of the World Bank Enterprise surveys (WBES). There is also an increasing volume of work using other data bases, e.g. on China (Sun and Du 2010).
This is a literature we will be adding to in using one of the more recent waves of the WBES survey carried out in 2009, which we use to analyse innovation in 30 emerging/transition countries in Euro-Asia, which include rapidly developing ones, now members of the EU, and also ones less well progressed along the development path. These represent a specific group of countries, different to the developing countries of e.g. Africa and different again from more developed countries. The determinants of innovation and the potential role of FDI are crucial issues for their development. We focus on regional spillovers, i.e. the impact of foreign firms on domestic ones within 150 regions of the 30 countries. This regional approach is also relatively new to the literature, but is critical in allowing an analysis of spatial spillovers. It is the only way we can find such information on the different regions. It might be argued that some regional variables, e.g. views on corruption, are based on the subjective views of firms. It is common in the literature on corruption etc. to use such subjective variables and have been used in a regional contextbefore (Sivak et al.2011). However in this context Weber Abramo (2007)has argued that survey data relating to corruption may, particularly in wealthier countries, reflect institutional quality at large and not just corruption. There are two points to note on this. Firstly, Weber Abramo’s paper related to household perceptions and it may not carry over to firm based perceptions. Secondly, in this paper the issue is not corruption per se but rather institutional quality. Nonetheless this is a point to bear in mind when interpreting the results.
A feature of our study is that, unlike much of the rest of the literature, we actually analyse variables pertaining to innovation. Specifically we analyse four aspects of innovation: (i) the introduction of a new product or service, (ii) incremental innovation, or product/service upgrading, (iii) R&D (in-house or outsourced) and (iv)the licensing of technology from a foreign-owned company.In what follows we use the term ‘product innovation’ to describe the first of these and the term innovation to refer more generally to all four concepts. The paper is thus one of relatively few to deal with actual firm based measures of innovation, rather than proxies for innovation.
The paper also contributes to the literature in other ways. From a policy perspective it is an important question whether inward FDI really drives economic growth and an important aspect of this is any impact on innovation. These issues are particularly important for firms in transition economies, where many firms are trying to gain a foothold in export markets, and the effects on their economy could be relatively large compared to the more marginal effects on developed countries. In addition, the traditional research approach focuses on the impact of FDI in a region or industry on the degree of innovation or productivity in host country firms. Other approaches used in previous literature trace innovation or productivity benefits from foreign ownership,or highlight the quality of institutions as drivers or deterrents for innovation. Combining all these aspects into a more general study of FDI and innovation, as we do, is thus to an extent novel. The paper proceeds as follows. In the next section we will review the literature relating to the process of innovation, the impact of FDI on innovation and other factors, which impact on innovation. We will then introduce the data and discuss the econometric specification. The results will follow in the following section and finally we conclude the paper.
2The Literature on Innovation
2.1 The Process of Innovation
Innovation begins with knowledge, entirely new knowledge, or knowledge new to the market, i.e. new to a country, region or town. R&D is a critical factor in this. R&D can take place in many forms, either ‘in house’ by firms or by research institutes in collaboration with firms. Once generated, knowledge is then diffused to other firms and other countries and regions. Bartlesman and Doms (2000) note an S shaped diffusion pattern through the economy. The starting point for much of the knowledge diffusion literature is that cities are the focal points for knowledge spillovers (Henderson 2007). This knowledge is then diffused to others, with those who are a long way from a major town or city being at an informational disadvantage. A key barrier to knowledge diffusion is lack of appropriate technical skills in firms and countries to be able to absorb and if necessary adapt knowledge for local use.
The entrepreneur plays a key role insuccessfully converting knowledge into new (perhaps new to market) products. That describes product innovation, on which this paper and much of the literature is focused. But there is also process, management/organisational and marketing innovation on which we have less to say. Factors, which the literature suggests impact on the entrepreneur, include finance and governance. There are a number of factors, which facilitate innovation, by helping overcome the barriers to both entrepreneurship and knowledge diffusion/generation. An obvious role for FDI lies in facilitating the diffusion of knowledge to countries other than where the knowledge was generated, and it is to this we now turn.
2.2 The Impact of FDI on Innovation
Assuming that foreign firms employ a higher level of technology than the average domestic firm, then FDI should almost by definition increase the average level of technology in the host country. The key question is whether there are spillovers to other firms. Do they help to facilitate the diffusion of advanced knowledge to local firms? The literature suggests that spillovers can arise in different ways:(i) when trained workers move to domestic firms or set up their own enterprise (Kaufmann 1997;Fosfuri et al. 2001), (ii) by observation, by watching and learning about new products, equipment, marketing techniques, and management practices. (Ben Hamida and Gugler 2009), (iii) by impacting on the supply chain(Hoekman et al.2005), (iv) by increased competition,forcing local firms to be more competitive (Ben Hamida and Gugler 2009). There is perhaps also a fifth channel, relating outsourcing tolocal specialist service providers, e.g. IT and finance, which can then benefit other firms.
There is, however, also the possibility of a negative effectof FDI ondomestic firms. Aitken and Harrison (1999) find such an effect, in that an increase in foreign investment causes a decline in the productivity of domestic firms. They attribute this to a ‘market stealing’ effect and it links in with the literature on the impact of competition. Their argument was that if imperfectly competitive firms faced fixed costs of production, a foreign firm entering the market would have an incentive to increase production relative to its domestic competitor. The productivity of domestic firms declines as fixed costs are spread over a smaller market, forcing them up their average cost curves. It then seems possible that part of this cutting back process will involve a reduction in innovative activity.
The literature is growing rapidly and there is some evidence of FDI induced knowledge spillovers, but the evidence also suggests that such an impact is variable and mixed for both developed and developing countries. Karpaty and Lundberg (2004) found positive evidence for the existence of spillover benefits from FDI for Sweden,while, using the same measure of spillovers, Aitken and Harrison (1999) and Castellani and Zanfei (2003) reported significant negative spillovers for Venezuela andSpain respectively (Castellani and Zanfei also found positive results for Italy). Girma (2005)found positive result for the UK, dependent upon absorptive capacity,and Marin and Bell (2006) found insignificant results for Argentina. Sabirianova, Svejnar and Terrell (2005) analysed the effects of firm entry on firm performance in Russian and Czech industrial firms. There was a positive effect on the productivity of incumbent foreign firms but a negative effect on the productivity of domestic firms. FDI has been shown to have spillover effects on patent rates in US states (Ford and Rork 2010),on the supply chain (Javorcik and Spatareanu 2009) andon manufacturing in Switzerland, particularly for demonstration-related spillovers (Ben Hamida and Gugler 2009).Inzelt (2008) presents the evidence for the importance of labour mobility, relating to highly qualified workers, in spillover effects of FDI through facilitating tacit knowledge flows.Much of the positive evidence comes from analysing vertical spillovers, namely those taking place through contacts between multinationals and their local suppliers of intermediate inputs (Moran 2001; Javorcik 2004, using Lithuanian data; Javorcik and Spatareanu 2008, for Romania; Blalock and Gertler 2008, for Indonesia). Apart from vertical spillovers, there is evidence that domestic firms in general benefit from a foreign presence in their sector (Haddad and Harrison 1993 for Morocco; Aitken and Harrison 1999 for Venezuela; and Djankov and Hoekman 2000, for the Czech Republic).
A great deal of work has been done on China, with again mixed results. Buckley, Clegg, and Wang (2002) found positive evidence for the existence of spillover benefits from FDI, but Tian (2007) negative results. Cheung (2010) concludes that China's domestic firms benefit from FDI induced spillovers and from export activity of both domestic firms and foreign-invested enterprises. However, Xiao et al. (2009) conclude that FDI has a positive impact on the capacity of independent innovation of China primarily in the eastern regions, but not so much in the rest of the country.
Hence, the evidence is mixed, but perhaps the weight of the evidence is beginning to come down in favour of some positive, if qualified, impact. One of the reasons for these mixed results may lie in the difficult nature of testing for spillover effects, another may liein the differences in the dependent variable used to proxy innovation. There are other factors too, relating to different types of spillover effects and differing levels of absorptive capacity within domestic firms which limit their capacity to learn.
2.3 Other Variables which Impact on Innovation
Although the focus is on the impact of FDI, we need to allow for the impact of other variables in our analysis. Caselli and Coleman (2001) analyse the diffusion of computers across the world. They find human capital and a country's openness to manufacturing imports to positively impact on innovation adoption. Skills are also emphasized by the OECD (2010) in impacting on knowledge diffusion through their impact on absorptive capacity (Cohen and Levinthal 1989). With respect to the determinants of entrepreneurship, finance and governance have been emphasized (OECD 2010; the World Bank 2008; Ayyagari et al. 2007). On the theory side, there have been suggested linkages between governance and institutions such as property rights and innovation (e.g. Grossman and Helpman 1991). However, according to Gorodnichenko et al. (2010) there is, their own study apart, no firm level empirical work on the impact of the business environment on innovation. By this they mean, e.g., the impact of governance factors, such as corruption, on innovation.
As already indicated there isrelatively little on the determinants ofinnovation per se, with more emphasis on firm efficiency (Wagner 2007). For example, Henry et al. (2009) model the production frontier for 57 developing countries and find an important influence of trade and trade policy in raising output both through technology improvements, embodied in imported capital goods, and by inducing efficiency improvements.There is a relatively large literature relating to the impact of theexporting and importing activities of domestic firms on their efficiency – presumably through innovation that is induced by the exposure of the domestic firms to more advanced practices and technologies (see Wagner 2007, for a survey).
Many economists have argued that competition provides incentives for the efficient organization of production, putting downward pressure on costs, and motivating innovation (e.g. Arrow 1962; Gilbert and Newbery 1982). Dixit and Stiglitz (1977) and Aghion and Howitt (1992), among others, stress that product market competition reduces monopoly rents that induce innovation. Recently, Aghion et al. (2005) have shown that competition can have different effects on firms’ willingness to innovate depending on their level of efficiency. In particular, firms close to the efficiency frontier are expected to be spurred by competition to innovate and increase their efficiency, while firms that are far from the frontier are expected to be discouraged by competition from innovating. Recent work has expanded this to make the impact dependent upon industry conditions (Cohen and Levinthal 1989). Much of the work is based on advanced economies, but it does not always travel well. Thus, Carlin, Schaffer and Seabright (2004) use, as we do, the (1999) Business Environment and Enterprise Performance Survey to analyse transition countries, finding that innovation is higher in monopolistic industries.