Infraestrutura de transportes e colaboração universidade-empresa: evidências regionais do Brasil Victor Medeiros (CEDEPLAR-UFMG); Alexandre de Queiroz Stein (CEDEPLAR-UFMG); Rafael Saulo Marques Ribeiro (CEDEPLAR-UFMG); Ulisses Pereira dos Santos (CEDEPLAR-UFMG); resumo: As universidades têm desempenhado um papel importante na promoção da inovação no nível da empresa. As interações universidade-empresa são assunto de crescente interesse nos estudos sobre inovação. A literatura identificou que spillovers de conhecimento são limitados no espaço. Assim, uma infraestrutura de transporte adequada pode amplificar transbordamentos de conhecimento, conectando locais e promovendo a difusão do conhecimento. Neste trabalho, examinamos o impacto de um aumento no provimento de estradas sobre as interações universidade-empresa no Brasil, usando modelos de variáveis instrumentais para tratar possíveis problemas de endogeneidade. Nossos resultados sugerem que rodovias impactam positivamente as interações U-E. Também mostramos que os efeitos das estradas nas interações U-E locais são maiores para empresas e grupos de pesquisa de pequeno porte, grupos de pesquisa de alta qualidade e microrregiões líderes. Além disso, encontramos que a rede de estradas pode estar concentrando espacialmente os fluxos de conhecimento no Brasil. palavras-chave: infraestrutura de transportes; colaborações universidade-empresa; desenvolvimento regional; variáveis instrumentais; Brasil. Código JEL: O18; H54; O31; R40; C20 Área Temática: 4.4 Redes de inovação – alianças de P&D, interações universidade-empresa, outras redes 1. Introduction Universities have been playing a central role in fostering technological progress in firms (Garcia et al., 2015). However, it is also known that knowledge spillover decreases with distance, thus imposing a limit to collaborative efforts between universities and firms (Jaffe et al., 1993; Feldman and Audretsch, 1999; Bottazzi and Peri, 2003). Aspects like the face-to-face contact, the transit of researchers from universities to companies and the possibility of sharing labs and equipment favor the knowledge spillovers among closer locations. That is why spatial agglomeration generally stimulates the maintenance of frequent contacts between academic researchers and firms’ research and development staff (Garcia et al., 2013). Consequently, innovative activity tends to be more concentrated than industrial activity (Carlino and Kerr, 2015). Then efforts for shortening distance barriers among universities and firms enabling the transit of academic and industry researchers are expected to improve spatial diffusion of knowledge. This implies that pervasive knowledge building requires not only more investments in higher education and in academic R&D, but also more investments in the provisioning of adequate means to connect scientific and technological assets in different locations. In this way, the development of road network and efficient transportation systems in order to connect places can be taken as strategy for boosting learning and knowledge diffusion over space (Feldman and Kogler, 2010). In Brazil, 24,646 out of 37,640 research groups in 2016 (around 68.4%) were concentrated in the South and Southeast regions, which are the regions hosting the main productive and innovative hubs in the country (Brazilian Ministry of Science and Technology, 2016). As a consequence, the U-I collaborations are highly concentrated in very few localities, thus exposing the marked spatial discontinuity of the Brazilian innovation system (Figure 1) (Gonçalves and Almeida, 2009; Santos, 2017). The geographical distribution of the country's highway network also follows a similar pattern (Figure 2). Such an agglomeration process in terms of both roads, and productive and innovative activities, creates a vicious circle. If a region has a poor road infrastructure and high transportation costs, firms might not be able to interact with more distant local partners; similarly, an undeveloped transportation infrastructure can discourage the displacement of researchers and workers within and between regions, hence disrupting the process of face-to-face contact, knowledge spillovers and innovation. Since innovation is considered one of the main drivers of economic growth and regions have different infrastructure endowments, investments in road infrastructure may be a key policy measure with the aim of promoting a sustained and regionally balanced economic growth. Figure 1. Territorial distribution of local U-I collaborations in Brazil: 2016 Source: Authors’ elaboration using data from the Brazilian Ministry of Science and Technology. Figure 2. Territorial distribution of highways in Brazil: highways density, 2010 Source: Authors’ elaboration using data from the Ministry of Transport and the National Department of Transport Infrastructure. Several empirical studies have proven the role of infrastructure on economic growth, productivity, employment, trade and poverty (Arbués et al., 2015; Cosci and Mirra, 2017; Deng et al., 2014; Fingleton and Szumilo, 2019; Hong, Chu and Wang, 2011; Jiwattanakulpaisarn et al., 2010; Medeiros, Ribeiro and Amaral, 2020; Wan and Zhang; 2018; Zhang and Ji, 2019). However, few studies have investigated the role of transportation infrastructure in promoting innovation and expanding knowledge flows. Agrawal, Galasso and Oettl (2017) found that roads had a strong knowledge diffusion effect, which encouraged regional innovation in the United States. Similarly, Wang et al. (2018) showed that road development spurs innovation by enlarging market size and facilitating knowledge spillovers in China. Dong, Zheng and Kahn (2019) document a complementarity effect between knowledge production and the transportation network in China. Since high-speed railways reduce cross-city commute times, they reduce the cost of face-to-face interactions between skilled workers who work in different cities. Following these three studies, we examine the impact of road development on local U- I collaborations in Brazil, an issue that has been overlooked by both the innovation and the transportation infrastructure literature. The contribution of this paper to the literature is fourfold. Firstly, we examine the role of inter and intra state roads1 on local U-I linkages in a developing country. Using Brazil’s recent data on U-I collaborations, we document that increases in the highways stock effectively increases local U-I linkages with a lag of six years. We emphasize that the local within-region knowledge flows channel also works through the interaction among firms and universities (Agrawal, Galasso and Oettl, 2017). Better transportation infrastructure may increase the mobility of workers and researchers and consequently the diffusion of knowledge across space, thereby allowing ideas to cross-fertilize. This finding shed light on the “black box” of knowledge spillovers and provides solid evidence on the determinants of U-I collaborations. These results remain under several econometric specifications. Secondly, in order to avoid possible endogeneity issues that may arise from mutual determination between U-I linkages and roads investment decisions, we employ an instrumental variables approach based on the bureaucratic and geographical costs associated with infrastructure projects. The choice of instruments follows the related literature (Duflo and Pande, 2007; Saiz, 2010; Wang et al., 2018) in light of the Brazilian case. This strategy seeks to solve the problem of endogeneity between infrastructure and development outcomes based on the identification of some of the main problems for the design and feasibility of infrastructure projects in developing countries, especially in those with large geographical areas and huge geographic heterogeneities. The first instrument is the share of legally protected areas2 in a micro-region. The greater the proportion of protected areas, the higher the difficulty in constructing highways. The second instrument chosen is the slope of a micro- region, which measures the relative difficulty (cost) of constructing roads there. Similarly, we also try altitude and rain volume as instruments. Thirdly, we also identify some heterogeneous effects of increased highways on local U-I interactions. The first one is related to firm and research group size heterogeneity. Larger firms tend to interact more with universities in order to obtain new knowledge, improving their innovative capacity. In general, larger firms have less financial and educational obstacles to innovate (Bishop et al., 2011). Our findings indicate a significant effect of highways stock for small-sized firms and research groups. Smaller firms and groups are more restricted to their local environment since long distance collaborations require a broader range of capabilities and incur in higher costs (Muscio, 2013). The second heterogeneity is related to the research group quality. High-quality groups tend to engage in collaborations at a larger geographical distance, suggesting that such research groups can attract more distant firms as collaboration partners (Garcia et al., 2015). We provide evidence that better highway connectivity encourages firms to search for higher quality research groups, probably by allowing these firms to interact with more distant local universities. Lastly, our results provide insights on the role of roads on U-I linkages taking into account 1 We study road transportation because of its importance in the Brazilian scenario. The sector has historically concentrated the most part of the country's cargo transportation, being more than 61% in 2019 (National Transport Confederation, 2019). Interstate roads may connect different states and are administrated by the Federal Government of Brazil, while intrastate roads connect different cities within the same state and are administrated by state-level governments. 2 These are conservation units (sustainable use and integral protection), military areas and indigenous lands. spatial issues. Infrastructure effects on growth and productivity might be greater in the initial stages of development, whilst in developed regions these impacts could be lower (Cosci and Mirra, 2018; Chen and Vickerman, 2016; Crescenzi and Rodríguez‐Pose, 2012). Our findings show larger road effects on local U-I linkages in the leading region of the country – the South. Probably the highways are supporting innovative activities in those locations by facilitating the movement of researchers and workers to more distant locations and stepping up the interaction among them. In the laggard regions of the country, the undeveloped road infrastructure may be discouraging the flows of people through highways. Next, we test for spatial effects of the road stock. As argued by the New Economic Geography literature, infrastructure may affect the distribution of firms and workers between and within regions (Ottaviano, 2008), and it will shape the way which firms and universities interact. In order to capture these possible neighborhood effects, we include highways density in neighboring micro-regions in the regressions. Our findings provide evidence of a negatively signed and significant spatial effect of increased highways stock on U-I linkages, thereby indicating that the greater the roads stock in the neighboring regions, the lower the U-I collaborations in the micro-region. Nevertheless, the overall effects of transportation infrastructure on U-I collaborations are still positive. The paper is organized as follows. Section 2 depicts the related literature. Section 3 describes the data and methods. Section 4 reports the estimation results and the underlying heterogeneities. Section 5 concludes. 2. Related literature Our paper is associated to the literature on the determinants of regional innovation and knowledge flows and the effects of transportation infrastructure on regional development. We focus on the role of roads network in stimulating local university-industry collaborations, especially in developing and more regionally unequal economies. Universities are an essential source of knowledge and may boost innovative activities of firms. Knowledge produced by academic R&D generally serves as basis for relevant technological innovations (Mazzucato, 2013). In addition, universities contribute to the formation of new and skilled professionals and favor the rising of spin off companies (Trippl et al., 2015). In this way, higher education institutions are seen as agents for economic development in both regional and national levels. Furthermore, universities can play an important role as agents of social development (Arocena and Sutz, 2005), especially in laggard countries or regions at where the productive and innovative activities are weak and not based on high-technology industries compared to the leading economies. However, evidence regarding the spatial distribution and the co-location of university-industry linkages is mixed. On the one hand, a number of works in the regional innovation literature has convincingly advocated the importance of spatial proximity in generating knowledge spillovers (Feldman, 1994; Feldman and Audretsch, 1999; and Jaffe et al., 1993; Gonçalves and Almeida, 2009). Pecuniary knowledge externalities emerge from interactions among local agents, which tends to increase the knowledge sharing, technological learning and its dissemination (Antonelli, 2008). Firm’s R&D activities, skilled labor and academic research are examples of sources of local knowledge spillovers (Garcia et al., 2013). Many others studies have pointed out the importance of these knowledge spillovers bounded in space (Breschi and Lissoni, 2001; D’Este and Iammarino, 2010; Garcia et al., 2015; Laursen et al., 2011; Muscio, 2013; Rodríguez-Pose and Crescenzi, 2009; Varga, 2000). In general, those works found that the smaller the spatial distance between universities and firms, the greater the interactions among them. On the other hand, a number of studies have called into question the argument that the geographical proximity between universities and firms significantly increase the possibility of a firm to achieve a successful pattern of innovation. Breschi and Lissoni (2009) shows that the effect of spatial proximity on knowledge diffusion is not so strong. There are other factors that may influence innovation and knowledge flows rather than geographic proximity, including social and cognitive proximity, institutional and infrastructure aspects, and market opportunities (Guastella and van Oort, 2015). Recent studies have shown that firms often search for high quality, geographically distant universities that can solve their innovation problems (D’Este and Iammarino, 2010; De Fuentes and Dutrénit, 2014; Garcia et al., 2015; Laursen et al., 2011; Muscio, 2013). Manifold studies have investigated the importance of infrastructure on economic growth, poverty, productivity, trade and innovation in a regional approach (Agrawal, Galasso and Oettl , 2017; Arbués et al., 2015; Cosci and Mirra, 2017; Fingleton and Szumilo, 2019; Hong, Chu and Wang, 2011; Jiwattanakulpaisarn et al., 2010; Medeiros, Ribeiro and Amaral, 2020; Wan and Zhang; 2018; Zhang and Ji, 2019). Our study sheds some light on the role of roads on local university-industry collaboration. We argue that an adequate transportation infrastructure amplifies knowledge spillovers by connecting places and promoting the exchange of ideas. By contrast, even if two places are geographically close, but lack the support of transportation infrastructure, knowledge spillovers will take place at a lower magnitude than expected (Feldman and Kogler, 2010). In this sense, highways might play a central role in stimulating knowledge creation and dissemination. Some recent works have investigated the relationship between transportation infrastructure, innovation and knowledge flows. Agrawal, Galasso and Oettl (2017) evaluated the impacts of the stock of interstate roads on regional innovation in the U.S. using patent data. The authors’ main results show that in regions where the stock of highways is larger, innovators build on local knowledge that is geographically more distant, insofar as this infrastructure facilitates the circulation of local knowledge. Similarly, Wang et al. (2018) examined the effects of roads on innovation at the firm level in China. In addition to the circulation of local knowledge channel, they find that improved roads expand market size, which in turn leads to more innovation. Dong, Zheng and Kahn (2019) evaluated the impacts of China’s high-speed rail network on the interaction among high skilled teamwork, and found that this type of transportation infrastructure increases the production of academic papers and facilitates flow of ideas between two high-speed rail connected cities. Although the literature on transportation infrastructure and knowledge flows has advanced, there are still umpteen open points. We focus on the role of highways in encouraging U-I linkages, an issue that might be crucial in stimulating innovative activity in lagging countries and regions. 3. Data and empirical strategy 3.1. Data In order to evaluate the role of transportation infrastructure on university-industry linkages, we used data from the Brazilian Ministry of Science and Technology3 which provides a broad dataset covering the activities of academic research groups in Brazil at the regional level. Then, we merged this dataset with detailed information of firms’ location and size collected from the Brazilian Ministry of Labor database. This way, we were able to combine information about the location of both firms and research groups. Next, we constructed local U-I measures at the micro-regional level, which can be associated with the European Union NUTS-3 (Garcia et al., 2015). According to Santos (2017), there are several advantages in using the micro-regional scale compared to other aggregations in the Brazilian case. A state-level analysis tends to exhibit a high level of heterogeneity, not allowing us to capture local economic dynamic. On the other hand, the municipal scale was not deemed the most appropriate one for this analysis either because the technological and economic structure of a municipality serves residents in neighboring municipalities as well. The highway data was obtained from the Ministry of Transport and the National Department of Transportation Infrastructure. Following the literature (Agrawal, Galasso and Oettl, 2017; Dong, Zheng and Kahn, 2018; Wang et al., 2018), we construct a proxy for the stock of roads. First, we used the length of paved roads (in km)4 and multiplied it by the number of road lanes5. Next, we divided it by the micro- region area (km²) and used the log form. 3 We exploit the Directory of Research Groups provided of the National Council for Scientific and Technological Development using the Lattes platform. These data were organized by the research group on Economics of Science and Technology of the Center for Development and Regional Planning of the Federal University of Minas Gerais. 4 We tried three measures: total road length; road density (total road length divided by the micro-region area in km²), and; road per capita (road length divided by the micro-region population). The results were quite similar regardless of the variable used. 5 If the road has one lane, we multiplied its length by one. If the road is duplicated (two-lane), we multiplied its length by two. 3.2. Empirical framework Our baseline model focuses on the relationships between interstate and intrastate highways in micro-region m in 2010, Highwaysm,2010, and local university-industry linkages in micro-region m in 2016, Ym,2016. The idea of using the road variable with a lag of six years seeks to take into account that the realization of U-I connections at full potential may require some time until investments in the provisioning of transportation infrastructure reach maturity and specific new knowledge in both firms and universities is created. In other words, use of a six-year lag is justified since the provisioning of new roads will only come into productive use by both firms and universities in some future period6. Thus, our baseline model goes as follows: Ym,2016 = α + βHighwaysm,2010 + θXm + εm (1) We use two measures for university-industry collaborations. The first is a dummy variable that equals one if a micro-region had at least one local U-I collaboration in 2016 and zero otherwise. The second variable is the log of total local U-I collaborations, “log(U-I interactions + 1)” as proposed by Wang et al. (2018) and Agrawal, Galasso and Oettl (2017)7. In order to capture the local dynamic of the interactions among universities and firms, we consider only those interactions that occur between firms and research groups established in the same micro-region. In this sense, our dependent variables allowed us to evaluate the partial effects of highways on local U-I linkages. The term Xm is a vector of control variables, including the educational level, gross domestic product (GDP) per capita, population, demographic density, a dummy variable indicating the existence or not of a paved airport, the innovate dynamic of the micro-region measured by the number of patents and a dummy variable capturing regional heterogeneity. A more detailed description of the variables can be found at Table A1 in Appendix A. We use Probit and Poisson8 models to estimate the impacts of highways on U-I collaborations. These models are suitable9 when using binary and count dependent variables, respectively (Cameron and Trivedi, 2005). The parameter of interest is β, which describes the impact of highways provision on U-I linkages. The main empirical challenge in estimating equation (1) is the possible bias coming from endogeneity issues. It is possible that the error term εm is correlated with the stock of highways. For instance, in areas with high growth potential, local governments may invest more in infrastructure there. At the same time, those micro-regions may have a greater fiscal capacity to improve its universities (Dong, Zheng and Kahn, 2019). If these situations exist, then the observed rise in micro-region innovative activity is likely driven by unobserved factors rather than road development. In this case, conventional Probit and Poisson would yield biased estimates of the causal effect of highways on U-I linkages. In order to avoid the problem of omitted variables, we employ an instrumental variables approach. 3.3. Instrumental variables: a legal and geographical cost approach We employ two types of instruments based on the transportation infrastructure literature in the light of the Brazilian reality. The first instrumental variable used is the proportion of legally protected areas10 in a micro-region. The greater the proportion of protected areas, the more difficult it may be to constructing highways. Building roads in these areas requires incurring in heavy bureaucratic costs including environmental licensing and long delays in permit issuance by local authorities. This instrument has a high negative correlation with the road stock. It seems to indicate the deep problems related to the Brazilian bureaucracy and its consequences for the elaboration and conclusion of infrastructure projects. National survey demonstrates excessive bureaucracy as one of the main problems in the construction and infrastructure sector - 30.5% of the country's entrepreneurs answered that they spent considerable time and resources in complying with legal requirements to set up, obtain licenses and authorizations) (CNI, 2019). The second set of instruments are related to geographical issues, which may directly affect the 6 Infrastructure investments can be expected to take a long time to mature (Straub, 2011). 7 In this case, the log form was applied to patent data. 8 It could be argued that the Zero-Inflated model might be more appropriate in this case. However, studies (Naya et al., 2008; Staub and Winkelmann, 2013) show that the Poisson and Zero-Inflated models generate, in most cases, similar estimates. In addition, the Zero-Inflated model does not yet offer reliable options for the application of an instrumental variables approach, an essential test in our work. On the other hand, the Poisson model has been widely used in models that account for endogeneity issues. In this sense, we have opted for the Poisson model. 9 In these cases, linear models could generate biased and inconsistent estimates. 10 These are conservation units (sustainable use and integral protection), military areas and indigenous lands. costs of constructing roads. The first instrument to be tested is the slope of a micro-region, following Duflo and Pande (2007), Saiz (2010) and Wang et al. (2018). Our slope variable measures the proportion of the micro-region area with slope above 20%, which corresponds to hilly areas. The greater the value of this variable, the higher the cost of building roads. In steeper areas, a stringent road design would lead to a less winding construction. To conform this type of project, it is essential to build several special artworks such as tunnels and bridges. Those roads have higher economic costs and higher environmental requirements, which in some cases may lead to the unfeasibility of their execution. Similarly, we use geographical features such as altitude and the rain volume as instrumental variables. We argue that the instruments affect University-Industry interactions only through the infrastructure variable. The low correlation among measures of U-I interaction and the chosen instruments show that steep terrains or protected areas do not directly limit (or expand) local U-I linkages (Table A2 in Appendix A). On the other hand, these characteristics are crucial in determining the feasibility of an infrastructure project, which in turn will change the trade costs and travel times intra and inter cities. We have also included several control variables in order to mitigate potential omitted variable problems. The validity of the instrumental variable estimation hinges on the orthogonality of the dependent variable and the instrument conditional on control variables, not on unconditional orthogonality (Duranton and Turner, 2012; Wang et al., 2018). The summary statistics are described in Table 1. Table 1. Summary statistics Variable Mean Std. dev. Min Max Highways 12.40 2.149 1.101 15.13 Log (U-I interactions + 1) 0.245 0.718 0 4.635 Having at least one U-I interaction 0.142 0.349 0 1 GDP per capita 2.883 0.611 1.635 5.045 Population 12.19 0.954 7.983 16.50 Demographic Density 7.984 1.509 3.478 13.35 Airport 0.618 0.486 0 1 Patents 0.527 1.185 0 7.158 Instruments Slope 0.069 0.103 0.001 0.835 Protected areas 0.074 0.162 0.000 0.972 Altitude 382.1 276.9 2.75 1171 Rain 88.19 31.01 20.36 182.9 Source: Authors’ elaboration. 4. Results and discussion 4.1. Regional U-I collaborations: benchmark results We begin our analysis by finding a positive and significant impact of intrastate and interstate highways stock on local U-I linkages in Brazil. Table 2 presents the econometric estimation results based on the specification in Equation (1). By using the full sample, it becomes difficult to disentangle two distinct transmission channels through which road networks affect innovation. The first one refers to the fact that highways might increase U-I interactions by facilitating the flow of researchers and workers between and within micro-regions. Better transportation infrastructure accelerates the mobility of people and the diffusion of knowledge across space, allowing ideas to cross-fertilize (Agrawal, Galasso and Oettl, 2017; Glaeser and Gottlieb, 2009). The second channel is related to the agglomeration economies (Duranton and Turner, 2012; Gibbons et al., 2018; Holl, 2016). Developed infrastructure may attract firms and researchers to a particular location, expanding local economies and its market potential. In this paper, we emphasize the first mechanism. To this end, our estimations refers to the impacts of interstate and intrastate highways stock in 2010 on local U-I linkages in 2016 between firms and research groups that already existed in 2010. In order to give robustness to our results, we include several control variables, beginning with no control variables, then including some control variables related to the micro-region development, demographic control variables, other modes of transportation and finally regional dummies. We note a positive and significant effect of roads on U-I collaborations based on conventional Poisson models11. However, in the most robust specification (Column 5), the highway coefficient is not significant. This insignificant effect may stem from endogeneity issues as we argue above. We expect unobservable factors to be correlated with both the levels of highways and the knowledge flows in a micro-region for a number of reasons (Agrawal, Galasso and Oettl, 2017; Wang et al., 2018). Table 2. Highways stimulate University-Industry interactions: Poisson regressions. Dependent variable: Log (University-Industry linkages + 1) 1 2 3 4 5 Highways 0.926*** 0.101* 0.183*** 0.132** 0.157 (0.20) (0.05) (0.07) (0.06) (0.12) Higher Education 0.011 0.267 0.067 0.440 (0.32) (0.26) (0.24) (0.30) GDP per capita 0.387** 0.488** 0.488** 0.336 (0.17) (0.21) (0.21) (0.25) Patents 0.597*** 0.326*** 0.343*** 0.310** (0.05) (0.12) (0.11) (0.13) Population 0.679*** 0.467*** 0.520*** (0.12) (0.10) (0.16) Population density -0.280 -0.137 -0.123 (0.18) (0.16) (0.19) Airport 1.166** 1.097** (0.59) (0.56) Northeast -0.034 (0.49) Southeast -0.520 (0.59) South 0.152 (0.61) Midwest -0.051 (0.47) Constant -13.594*** -4.818*** -12.522*** -11.026*** -12.092*** (2.77) (0.91) (2.13) (1.90) (1.99) Pseudo R2 0.14 0.39 0.42 0.43 0.44 Log likelihood -338 -240 -230 -224 -220 Source: Authors’ elaboration. Note: Values between square brackets are standard deviations. Clustered sandwich estimator option was used in order to allow for intragroup correlation (state-level). Significance: ***=1%; **=5%; *=10%. 11 We have also tested Probit models using as dependent variable “Having at least one local U-I interaction” as well as Zero-Inflated models using the same dependent variable as the Poisson models as robustness checks. The estimates are in Tables B1, B2 and B3 in Appendix B. Regarding the conventional Probit model, the results indicate a non-significant road effect on U-I interaction. When we move to the IV Probit regressions, the results indicate a positive and significant road effects in three of five specifications. This may suggest that our dependent dummy variable is not as suitable as the continuous Log (U-I interaction + 1). The results for the Wald test of exogeneity allow us to reject the null hypothesis of no endogeneity, which supports our choice of using models that control for endogeneity. Table 3. Highways stimulate University-Industry interactions: IV Poisson regressions. Dependent variable: Log (University-Industry linkages + 1) 1 2 3 4 5 Highways 0.290 0.767 1.125* 1.081* 1.054*** (0.20) (0.74) (0.59) (0.58) (0.40) Higher Education 0.383 1.042* 0.898 1.224*** (0.47) (0.56) (0.56) (0.45) GDP per capita 0.148 0.177 0.177 0.616* (0.28) (0.30) (0.28) (0.37) Patents 0.416** 0.181 0.184 0.233* (0.18) (0.15) (0.15) (0.13) Population 0.852*** 0.721*** 0.674*** (0.20) (0.21) (0.23) Population density -0.607*** -0.507*** -0.404** (0.18) (0.19) (0.19) Airport 0.808 0.914 (0.58) (0.73) Northeast -0.712 (0.73) Southeast -2.158** (0.97) South -1.422 (0.94) Midwest -1.065 (0.86) Constant -5.073* -13.429 -24.876*** -23.889*** -22.155*** (2.60) (9.59) (7.17) (6.99) (3.88) Hansen's J statistic 0.28 0.20 0.05 0.10 1.39 Source: Authors’ elaboration. Note: Values between square brackets are standard deviations. Clustered sandwich estimator option was used in order to allow for intragroup correlation (state-level). Instruments: 1- protected areas; 2- slope. Significance: ***=1%; **=5%; *=10%. In order to avoid this endogeneity, we turn to an instrumental variable estimation. The IV estimations using the protected area and slope instruments are described in Table 312. These results can be interpreted as the causal impact of the 2010 level of interstate and intrastate highways in the micro-region on micro-region U-I linkages in 2016. We find that more roads in 2010 led to more U-I collaborations six years later. These estimation results are in line with the previous literature that has shed some light on the positive effect of the stock of highways on innovation (Agrawal, Galasso and Oettl, 2018; Dong, Zheng and Kahn, 2019; Wang et al., 2018). This result indicates that increases in the provision of highways might have accelerated the knowledge flows between firms and research groups that already existed in 2010, which goes beyond the expected agglomeration effects coming from better infrastructure. Our findings are in line with the results found by Agrawal, Galasso and Oettl (2017) using patent data. In relation to the control variables, we have verified the expected signs. The higher the human capital the greater the U-I linkages in the microregion. The same result holds for the GDP per capita, population and population density controls. The Southeast regions dummy has presented a positive and significative coefficient, indicating that the microregion has more U-I interactions in comparison with the North region. This result was expected as the Southeast is the most developed and innovative region in Brazil. 12 In unreported estimates, we observed a high and significant correlation between the highways stock in 2010 and the instruments. The simple correlation among the variables can be seen in Table A2 in the Appendix A. 4.2. Robustness checks 4.2.1. Instruments In order to test the validity of our estimations, we have tried several specifications varying the set of instruments used (Table 4). We run different models with different instruments such as protected area, slope, altitude and rain and report all in the results that satisfied the Hansen’s J test of overidentifying restrictions. Independent of the model specification, we have obtained quite stable econometric parameters varying between 1.054 and 1.281, all suggesting a positive and significative relationship among the transportation infrastructure and the U-I interactions. Table 4. Highways stimulate University-Industry interactions: robustness check, instruments. Dependent variable: Log (University-Industry linkages + 1) Protecte d areas Protecte d areas and altitude Protecte d areas and rain Protecte d areas and slope Protecte d areas, slope and altitude Protecte d areas, slope and rain Protecte d areas, altitude and rain Protecte d areas, slope, altitude and rain Highway s 1.102** * 1.271** * 1.109** * 1.054** * 1.220** * 1.054** * 1.281** * 1.224** * (0.41) (0.44) (0.41) (0.40) (0.43) (0.40) (0.44) (0.43) Hansen's J statistic 0.00 1.09 0.04 0.84 2.39 0.84 1.14 2.39 Source: Authors’ elaboration. Note: Values between square brackets are standard deviations. Clustered sandwich estimator option was used in order to allow for intragroup correlation (state-level). Instruments: 1- proportion of area with slope above 20 degrees; 2- percentage of protected areas. Significance: ***=1%; **=5%; *=10%. All regressions control for the micro-region human capital, GDP per capita, population, population density, existence of an airport, regional dummies and level of patents. 4.2.2. The role of internet Another issue that has been pointed by Agrawal, Galasso and Oettl (2017) is the role of internet in shaping the U-I interactions. The authors argue that the access to Information and Communication Technology (ICT) may amplify or reduce the effect of roads depending on whether face-to-face interactions and ICT are complements or substitutes in knowledge production. Also, the road infrastructure variable might be just capturing the level of infrastructure development in a region, as infrastructure variables are likely to be highly correlated. In order to test this issue, we include the log of internet access in each microregion as a proxy to the telecommunications infrastructure. Table 5 shows the econometric results. Table 5. Highways stimulate University-Industry interactions: robustness check, the role of internet. Dependent variable: Log (University-Industry linkages + 1) Protecte d areas Protecte d areas and altitude Protecte d areas and rain Protecte d areas and slope Protecte d areas, slope and altitude Protecte d areas, slope and rain Protecte d areas, altitude and rain Protecte d areas, slope, altitude and rain Highways 0.955** 1.069** 0.988** 0.851** 0.932** 0.851** 1.089** 0.949** (0.43) (0.44) (0.42) (0.37) (0.39) (0.37) (0.43) (0.38) Broadband infrastructur e 1.114** * 1.125** * 1.132** * 1.068** * 1.035** * 1.068** * 1.133** * 1.042** * (0.36) (0.38) (0.36) (0.33) (0.34) (0.33) (0.38) (0.35) Hansen's J 0.00 1.24 0.12 1.72 3.78 1.72 1.28 3.77 statistic Source: Authors’ elaboration. Note: Values between square brackets are standard deviations. Clustered sandwich estimator option was used in order to allow for intragroup correlation (state-level). Instruments: 1- proportion of area with slope above 20 degrees; 2- percentage of protected areas. Significance: ***=1%; **=5%; *=10%. All regressions control for the micro-region human capital, GDP per capita, population, population density, existence of an airport, regional dummies and level of patents. The positive and significant effect of road infrastructure remains. The inclusion of the internet variable slightly decreases the magnitude of the coefficient of the transport variable. In addition, the expected positive sign of the internet variable coefficient indicates its importance for the promotion of U- I interactions. Our findings are robust to the inclusion of an ICT variable as a control, thus reinforcing the key role played by road infrastructure on the U-I linkages at a more local level. 4.2.3. Infrastructure quality Other issue that has been raised by the literature on infrastructure is the heterogeneous effect of the quality of infrastructure services (Medeiros and Ribeiro, 2020; Medeiros, Ribeiro and Amaral, 2020). Poor quality highways may increase the number of accidents and force individuals to adopt other modes of transportation, changing the expected effect of highways, for example. In order to control this effect, we include the log of the number of transit accidents in each microregion as a proxy for road quality. Table 6. Highways stimulate University-Industry interactions: robustness check, road quality. Dependent variable: Log (University-Industry linkages + 1) Protecte d areas Protecte d areas and altitude Protecte d areas and rain Protecte d areas and slope Protecte d areas, slope and altitude Protecte d areas, slope and rain Protecte d areas, altitude and rain Protecte d areas, slope, altitude and rain Highway s 0.838** 0.981** 0.929** 0.856** 0.913** 0.856** 1.066** 0.986** (0.41) (0.43) (0.41) (0.38) (0.41) (0.38) (0.44) (0.41) Traffic accidents - 0.137** * -0.118** -0.134** - 0.141** * -0.132** - 0.141** * -0.117** -0.126** (0.05) (0.05) (0.06) (0.05) (0.05) (0.05) (0.06) (0.05) Hansen's J statistic 0.00 1.04 0.52 1.34 2.67 1.34 1.41 2.87 Source: Authors’ elaboration. Note: Values between square brackets are standard deviations. Clustered sandwich estimator option was used in order to allow for intragroup correlation (state-level). Instruments: 1- proportion of area with slope above 20 degrees; 2- percentage of protected areas. Significance: ***=1%; **=5%; *=10%. All regressions control for the micro-region human capital, GDP per capita, population, population density, existence of an airport, regional dummies and level of patents. Even controlling for the road quality, the positive and significant road effect on U-I linkages remains. The inclusion of the road quality variable also slightly decreases the magnitude of the coefficient of the transport variable. Thus, the poorer the quality of the road network, the smaller the U-I interaction in a microregion. 4.2.4. Spatial autocorrelation The classical econometrics does not consider the spatial dimension explicitly. If not taken into account, the spatial dependence may cause bias in estimates or influencing the inference process, thus casting doubts on the results of regressions performed. This section re-estimates the IV regressions of Table 4 employing spatial econometric techniques in order to check the robustness of our main results. In addition, the estimation results described so far have focused only on the direct impacts of increases in highways stock on U-I interactions. However, another key aspect that must be taken into account while assessing the economic impact of infrastructure provisioning at the local level is that regions may benefit disproportionally from road improvements elsewhere (Agrawal, Galasso and Oettl, 2017; Wang et al., 2018). As argued by the New Economic Geography literature, infrastructure may affect the distribution of firms and workers between and within locations (Ottaviano, 2008). A first possible effect (“straw effect”) occurs when better connectivity between two regions causes less attractiveness to the poorer region. This effect occurs because economic activities are “sucked up” by the richer region due to better infrastructure conditions and establishment facilities for firms and families (Behrens et al., 2007). Another possible effect (“shadow effect”) happens when improving infrastructure in a region does not make it more attractive. In this case, the expansion of transportation infrastructure in the poorest region would be mostly used as an additional economic support for the richest region, thus causing resources to shift from the poorest to the richest region. A part from these unwanted effects, infrastructure may also reduce regional disparities, by promoting knowledge transmission from developed to less developed locations. In order to capture those possible spatial effects, we include highways density in neighboring micro-regions in the regressions. To create the spatial lags, queen matrices of first-order were created. The spatial weight matrix was constructed by contiguity, wherein the micro-regions that have a common border were considered neighbors (LeSage and Pace, 2009). Before proceeding with the estimations, we tested for spatial autocorrelation using the Moran’s I statistic. We observed a significant and positive spatial autocorrelation for the road stock in 2010, indicating that micro-regions with high (low) levels of road networks are surrounded by other micro-regions with high (low) road networks. Given that there are important local specificities in Brazil, we also tested for local clusters using the local Moran’s I statistic. As expected, we note a great cluster of micro-regions with poor transportation infrastructure in the North and part of the Midwest region (see Figure B1 in the Appendix B). On the other hand, there are “high- high” clusters in the Southeast and South regions. Table 7 describes the estimation results. Even controlling for spatial autocorrelation, the positive direct effects of the highways stock on U-I interactions remained. The magnitude of the net effect (direct minus indirect effects) is quite similar to our estimates in Table 4. Our findings evidence a negative and significant indirect effect of highways stock in 2010 on U-I linkages in 2016, which indicates that the greater the roads stock in the neighboring regions, the lower the U-I collaborations in the micro-region. Table 7. Highways stimulate University-Industry interactions: robustness check, spatial autocorrelation. Dependent variable: Log (University-Industry linkages + 1) Protecte d areas Protecte d areas and altitude Protecte d areas and rain Protecte d areas and slope Protecte d areas, slope and altitude Protecte d areas, slope and rain Protecte d areas, altitude and rain Protecte d areas, slope, altitude and rain Highways 2.513** 2.523** 2.454** 2.435** 2.502** 2.435** 2.456** 2.457** (1.21) (1.21) (1.15) (1.16) (1.22) (1.16) (1.13) (1.16) Highways , spatial lag -1.478* -1.487* -1.448* -1.452* -1.499* -1.452* -1.450* -1.471* (0.79) (0.79) (0.76) (0.77) (0.80) (0.77) (0.75) (0.76) Hansen's J statistic 0.44 0.44 1.03 1.55 1.05 1.55 1.02 1.56 Source: Authors’ elaboration. Note: Values between square brackets are standard deviations. Clustered sandwich estimator option was used in order to allow for intragroup correlation (state-level). Instruments: 1- proportion of area with slope above 20 degrees; 2- percentage of protected areas. Significance: ***=1%; **=5%; *=10%. All regressions control for the micro-region human capital, GDP per capita, population, population density, existence of an airport, regional dummies and level of patents. In the literature on roads and innovation using patent data, Wang et al. (2018) found a positive spatial spillover effect, while Agrawal, Galasso and Oettl (2017) obtained a not significant spatial feedback effect. However, our findings appear to indicate the existence of a negative spatial externality of roads stock on U-I linkages in Brazil. It is possible that more adequate transportation networks are intensifying the innovation dynamic of well establish micro-regions in terms of U-I interactions, which might exacerbate regional disparities through the occurrence of regionally unequal knowledge flows. This result is in line with the positive link between innovation and spatial inequality found by Lee and Rodríguez- Pose (2013) for European regions. Another explanation may be related to the immaturity of the Brazilian innovation system: because of the immense disparities in terms of urban, productive, and scientific and technological infrastructure, the spatial innovative agglomerations tends to be concentrated in historically more developed regions (Gonçalves and Almeida, 2009). In addition, we observe a huge discontinuity in the Brazilian territory, wherein great urban centers are surrounded by poor and small-population regions, which intensify the concentration of productive and innovative activities in few central places. A similar phenomenon is observed in the Chinese case (Crescenzi, Rodríguez-Pose and Storper, 2012). Our results suggest that laggard regions are not able to benefit from the knowledge spillovers stemming from the dynamism of surrounding regions due to their several economic and social constraints. This is entirely consistent with theories of economic development that suggest there is divergence in the earlier stages of development (Chen and Vickerman, 2016), and also corroborates the view that knowledge spillovers depend on a region’s absorptive capabilities, which very poor areas may lack (Zhang and Ji, 2019). It is important to emphasize that the total effects of transportation infrastructure on U-I collaborations are still positive, given that the magnitude of the direct effects exceeds the value of the indirect effects. Therefore, even with adverse spatial effects, larger road networks in 2010 caused an increase in U-I interactions in 2016. 4.3. Additional transmission channels and heterogeneities 4.3.1. Firms and research groups heterogeneities Having investigated the impact of roads stock on total U-I collaborations, we further consider the impact on U-I linkages by firm’s size and re-estimate eq. (1). Larger firms tend to seek such collaborations more often in order to obtain new information, enhance their professional recruitment, and facilitate the application of external knowledge in their innovation activities (Bishop et al., 2011). In Brazil, larger firms invest disproportionally more in innovative activities, have less obstacles to innovate, get more resources and incentives from government agencies to develop innovative activities, innovate more and have formal and informal methods that tend to be more effective in protecting new technologies and knowledge compared to their smaller competitors (Rapini, Chiarini and Santos, 2018). In order to test the existence of possible heterogeneities linked to the size of firms, we test the impact of increases in the highways stock on local U-I linkages for large, medium and small-sized firms13. The estimates by firm size are described in Table 8. This new set of estimations sheds some light on the presence of heterogeneities associated with firm and group size. Although we did not see significant effects of highways on local U-I interactions of larger firms, column "Small" indicate a significant effect of increases in highways stock for medium and especially small-sized firms. Small firms are more likely to engage in collaborative efforts with research groups located at a close distance to avoid incurring in substantial costs (Muscio, 2013). Small firms face worse conditions to innovate compared to big firms due to difficulties in attracting skilled workers, low access to credit and absorptive capacity, which might limit them to search for distant and high-quality universities. Due to their competitive disadvantage, small and medium-sized firms tend to become more dependent on local knowledge flows and hence interact with nearby universities and research groups. Our findings suggest that improving transportation infrastructure may stimulate local interactions between research groups and small and medium-sized firms, probably by cutting costs and expanding firms’ access to more distant local knowledge. Table 8. Highways stimulate University-Industry interactions: group and firm size heterogeneity. Dependent variable: Log (University-Industry linkages + 1) Firms Research groups Large Medium Small Large Non-large Highways 0.747 0.628 1.039** 0.833 1.127*** 13 We followed the IBGE classification based on the number of workers. We consider as small firms those with up to 99 employees; medium-sized firms those with 100 to 499 employees, and; large firms those with more than 500 employees. (0.47) (0.57) (0.41) (0.51) (0.42) Hansen's J statistic 0.00 0.63 0.02 3.47 0.11 Source: Authors’ elaboration. Note: Values between square brackets are standard deviations. Clustered sandwich estimator option was used in order to allow for intragroup correlation (state-level). Instruments: 1- protected areas; 2- slope. Significance: ***=1%; **=5%; *=10%. All regressions control for the micro-region human capital, GDP per capita, population, population density, existence of an airport, regional dummies and level of patents. We have also tested the effects of highways according to the size of the research group14. Research groups with more researchers have more accumulated capabilities, which overcome barriers to collaborating with industry partners (De Fuentes and Dutrénit, 2012). Similar to the previous result on firms, we have found that the road stock benefits the smaller research groups (Table 8). Next, we estimated the effects of highways on U-I collaborations by research group quality. The lack of a high-quality local partner tends to be associated with more geographically distant collaborations (Laursen et al., 2011). The quality of academic research was measured as the number of published papers per researcher following Garcia et al. (2015). We created a dummy variable that assumed the value one if the research group was among the 75% higher quality research groups. Then, we sliced the sample into “High” and “Other” quality interactions by micro-region. The results are described in Table 9. Table 9. Highways stimulate University-Industry interactions: research group quality heterogeneity. Dependent variable: Log (University-Industry linkages + 1) High Low Highways 1.907** 1.015** (0.97) (0.40) Hansen's J statistic 0.00 0.00 Source: Authors’ elaboration. Note: Values between square brackets are standard deviations. Clustered sandwich estimator option was used in order to allow for intragroup correlation (state-level). Instruments: 1- protected areas; 2- slope. Significance: ***=1%; **=5%; *=10%. All regressions control for the micro-region human capital, GDP per capita, population, population density, existence of an airport, regional dummies and level of patents. The highways stock positively affects local U-I linkages for both high quality and non-high-quality research groups, being this impact quite higher for the U-I interactions of high-quality groups. This finding reinforces the important role of roads in stimulating local U-I collaborations, and appears to indicate that better highway connectivity encourages firms to search for higher quality research groups, probably by allowing these firms to interact with more distant local universities. By the research group’s side, it is likely that more developed transportation infrastructure facilitates the flows of researchers within the micro-region, facilitating face-to-face contact with firms. This closer contact may be leading to better meeting the firm’s demands, making the U-I interaction more attractive. 4.3.2. Regional heterogeneity Another important aspect of some developing economies – as the Brazilian case - is the marked regional heterogeneity in terms of economic and social conditions. On the one hand, we observe some regions with high levels of infrastructure coverage and technological dynamism. On the other hand, we also have a number of localities with poor transportation systems and weak or inexistent U-I linkages. As we have showed earlier, Brazil presents a high number of micro-regions without any U-I linkage. The U- I collaborations are extremely concentrated in the Southeast and South regions, which exhibits the incomplete and immature nature of the Brazilian system of innovation (Suzigan et al., 2009). In addition, the highways networks in sufficiently good conditions are most notably concentrated in high-income regions such as the South and Southeast states as well as in the coastal micro-regions. 14 Following Garcia et al. (2015), we used the number of researchers as a proxy for research group size. We generated a dummy variable that assumed the value one if the research group was among the 75% higher groups in terms of researchers. Then, we divided the sample into “Large” group – local U-I linkages occurring between firms and large- sized research groups- and “Other” group – local U-I collaboration occurring between firms and not large-sized research groups. In the literature on infrastructure and development, some findings point out different highways effects on development depending on the level of development of the country or region (Calderón and Serven, 2014; Chen and Vickerman, 2016; Deng et al., 2014; Hong, Chu and Wang, 2011). Some studies also argued that the benefits associated to the increased provision of highways are unevenly distributed across sectors and space (Cosci and Mirra, 2018; Holl, 2016; Liu, Wan and Zhang, 2020; Qi et al., 2020). Infrastructure effects on growth and productivity might be greater in the initial stages of development, than in mature economies (Crescenzi and Rodríguez‐Pose, 2012). In the attempt to capture possible heterogeneous effects of U-I linkages by income level, we interact the road variable with a dummy variable that assume the value 1 if the microregion are in the South, Southeast and Midwest (the Southern regions). While the Northern region consists of the low- income states, the Southern region is constituted by the high-income states of the country. Table 10 presents our estimates: Table 10. Highways stimulate University-Industry interactions: regional heterogeneity. Dependent variable: Log (University-Industry linkages + 1) Protecte d areas Protecte d areas and altitude Protecte d areas and rain Protecte d areas and slope Protecte d areas, slope and altitude Protecte d areas, slope and rain Protecte d areas, altitude and rain Protecte d areas, slope, altitude and rain Highways*Sou th 0.443** 0.392** 0.522** 0.432** 0.468** 0.391** 0.499** * 0.462** (0.18) (0.17) (0.21) (0.17) (0.20) (0.17) (0.19) (0.19) Hansen's J statistic 0.00 1.42 1.46 0.09 3.50 1.42 1.65 3.54 Source: Authors’ elaboration. Note: Values between square brackets are standard deviations. Clustered sandwich estimator option was used in order to allow for intragroup correlation (state-level). Instruments: 1- protected areas; 2- slope. Significance: ***=1%; **=5%; *=10%. All regressions control for the micro-region human capital, GDP per capita, population, population density, existence of an airport, regional dummies and level of patents. Our findings show a positively signed effect of the interaction variable, indicating that the more developed regions of the country are benefiting more from road development. One possible explanation is related to the more developed infrastructure in those regions, which may have been facilitating the knowledge flows between local universities and firms. Moreover, those leading regions are also characterized by higher levels of income and education compared to the laggard regions U-I interactions and by hosting more developed transportation infrastructure, which may place them in a better position to reap the benefits from local-specific policies. Conversely, in the laggard regions of the country, the poor condition of the roads networks may lead to increased transportation costs, thus substantially undermining the flows of people engaged in the innovation sector in those micro-regions. In some cases, it may be more cost effective for innovative firms based in low-income regions to use other modes of transportation such as air transportation in order to go after more distant and highly ranked universities and research groups located in high-income states. 5. Concluding remarks Using a research group database merged with highways information both at the micro-regional scale, we estimate the causal effect of interstate and intrastate roads on local U-I collaborations. The empirical strategy is based on models for binary and count dependent variables that are robust to reverse causality. Our findings point out that better roads in a micro-region rise local U-I linkages over a six-year period, indicating that the “local within-region knowledge flows” channel found by Agrawal, Galasso and Oettl (2017) also works through the interaction among firms and universities. Better transportation infrastructure accelerates the mobility of workers and researchers and the diffusion of knowledge across space, allowing ideas to cross-fertilize. We tried several robustness checks in order to provide reliable results. The positive effects of road stock on U-I interactions seem to be robust to several instrument specifications and control variables, including broadband infrastructure, road quality, neighboring road stocks, regional heterogeneity, demographic and development variables. We also investigate the possible existence of heterogeneous effects by firm and research group size and stages of regional development. Our additional estimates find larger collaborations effects for smaller research groups and firms and high-quality research groups. We also find a larger road effect on local U-I linkages in the high-income regions of the country. This result appears to indicate that roads are stimulating U-I connections only in more economically and socially developed states, thus possibly reinforcing a vicious circle of regional disparities across the country. Also, we find a negative spatial externality of roads stock on U-I linkages in Brazil. This result may be partially attributed to the huge discontinuity in the Brazilian innovation system, wherein great urban centers are surrounded by poor regions, concentrating productive and innovative activities in few central places. The neighboring less developed regions may not be endowed with the necessary economic and social conditions to effectively benefit from the knowledge spillovers stemming from more dynamic regions. As stated by Wang et al. (2018), when designing innovation policy, the role of infrastructure should be included in the toolkit. It is shown here that transportation infrastructure endowments may shape the way that regions benefit from innovation spillovers. In addition, our estimates also suggest that road networks may increase regional gaps through U-I collaborations as firms and universities interact more in the developed regions than in the less developed regions. Given the existence of infrastructure’s negative spatial effects on U-I linkages, coordinated policies might be needed in order to avoid competition among local governments using highway infrastructure investment to attract firms and workers. Road policies seem to be an important tool in promoting a more balanced economic development, since it benefited more small firms and research groups, who have less resources to seek distant partners. Also, complementary policies aimed at improving human capital and other absorptive capabilities may also be important to amplify knowledge diffusion (Zhang and Ji, 2019). Our study has some limitations. The lack of longitudinal data at the micro-regional scale prevent us from studying the time heterogeneity (Straub, 2011). The effects of roads on the different types and sectors of the U-I collaborations have not been studied as well. We leave those as future research topics. Transportation infrastructure and University-Industry collaborations: regional- level evidence from Brazil Abstract: Universities have played a major role in fostering innovation at the firm level. The university- industry (U-I) linkages have become a subject of growing interest in studies on innovation. Literature has identified that knowledge spillovers are bounded in space. Thus, an adequate transportation infrastructure may amplify potential knowledge spillovers by connecting places and promoting learning and knowledge diffusion. In this work, we examine the impact of an increase in roads provisioning on U-I linkages in Brazil using instrumental variables econometric models to account for possible endogeneity issues. Our results suggest that highways positively impact U-I interactions. We also show that the effects of roads on local U-I collaborations are greater for small-sized firms and research groups, higher-quality research groups and leading micro-regions. Also, we find a negatively signed spatial effect of roads on U-I linkages, thus suggesting that the road network may be spatially concentrating knowledge flows in Brazil. Keywords: transportation infrastructure; U-I collaborations; regional development; instrumental variables; Brazil Referências bibliográficas AGRAWAL, A; GALASSO, A; OETTL, A. Roads and innovation. Review of Economics and Statistics, 2017, 99(3): 417-434. ANTONELLI, C. Pecuniary Knowledge Externalities: the Convergence of Directed Technological Change and the Emergence of Innovation Systems. Industrial and Corporate Change, 2008, 17(5), pp. 1049-1070. ARBUÉS, P, BAÑOS, J. F., Mayor M. The spatial productivity of transportation infrastructure. Transportation Research Part A 75 (2015) 166–177, 2015. AROCENA, R.; SUTZ, J. Conhecimento, inovação e aprendizado: sistemas e políticas no Norte e no Sul. In: Conhecimento, sistemas de inovação e desenvolvimento. Rio de Janeiro: Editora UFRJ/Contraponto., 2005. BEHRENS, K.; LAMORGESE, A. R.; OTTAVIANO, G. I.; and TABUCHI, T. Changes in transport and non-transport costs: Local vs global impacts in a spatial network. Regional Science and Urban Economics, 2007, 37(6), 625-648. BISHOP, K.; D’ESTE, P.; and NEELY, A. Gaining from interactions with universities: Multiple methods for nurturing absorptive capacity. Research Policy, 2011, 40, 30–40. BOTTAZZI, L., & PERI, G. (2003). Innovation and spillovers in regions: Evidence from European patent data. European Economic Review, 47(4), 687–710. https://doi.org/10.1016/S0014-2921(02)00307- 0BRESCHI, S; LISSONI, F. Knowledge Spillovers and Local Innovation Systems: a Critical Survey. Industrial and Corporate Change, 2001, 10, pp. 975–1005. BRESCHI, S.; LISSONI, F. Mobility of Skilled Workers and Co-Invention Networks: an Anatomy of Localized Knowledge Flows. Journal of Economic Geography, 2009, 9, pp. 439-468. CALDERÓN, César; SERVÉN, Luis. Infrastructure, growth, and inequality: an overview. The World Bank, 2014. CAMERON, A. Colin; TRIVEDI, Pravin K. Microeconometrics: methods and applications. Cambridge university press, 2005. CHEN, Chia-Lin; VICKERMAN, Roger. Can transport infrastructure change regions’ economic fortunes? Some evidence from Europe and China. Regional Studies, v. 51, n. 1, p. 144-160, 2017. COSCI, Stefania; MIRRA, Loredana. A spatial analysis of growth and convergence in Italian provinces: the role of road infrastructure. Regional Studies, 2018, v. 52, n. 4, p. 516-527. CRESCENZI, Riccardo; RODRÍGUEZ‐POSE, Andrés. Infrastructure and regional growth in the European Union. Papers in regional science, v. 91, n. 3, p. 487-513, 2012. CRESCENZI, Riccardo; RODRÍGUEZ-POSE, Andrés; STORPER, Michael. The territorial dynamics of innovation in China and India. Journal of economic geography, v. 12, n. 5, p. 1055-1085, 2012. D’ESTE, P. AND IAMMARINO, S. The Spatial Profile of University-Business Research Partnerships. Papers in Regional Science, 2010, 89(2), pp. 336-350. DENG, T., SHAO, S., YANG, L., & ZHANG, X. (2014). Has the transport-led economic growth effect reached a peak in China? A panel threshold regression approach. Transportation, 41(3), 567-587. DE FUENTES C., and DUTRÉNIT G. Geographic proximity and university–industry interaction: The case of Mexico. The Journal of Technology Transfer. Retrieved August 15, 2014 P. 1–20. DONG, Xiaofang; ZHENG, Siqi; KAHN, Matthew E. The Role of Transportation Speed in Facilitating High Skilled Teamwork Across Cities. Journal of Urban Economics, p. 103212, 2019. DUFLO, E., and PANDE, R. Dams. Quarterly Journal of Economics, 2007, 122(2), 601–646. DURANTON, G., and TURNER, M. Urban growth and transport. Review of Economic Studies, 2012, 79, 1407–1440. ETZKOWITZ, H. Research groups as ’quasi-firms’: the invention of the entrepreneurial university. Research Policy, Elsevier Science Publishers B.V. (North-Holland), v. 32, n. 1, p. 109–121, 2003. ETZKOWITZ, H.; LEYDESDORFF, L. The dynamics of innovation: from National Systems and “Mode 2” to a Triple Helix of university–industry–government relations. Research Policy, North-Holland, v. 29, n. 2, p. 109–123, 2 2000. ISSN 0048-7333. FELDMAN, Maryann P. The Geography of Innovation. Dordrecht: Kluwer, 1994. FELDMAN, Maryann P., and David B. AUDRETSCH, “Innovation in Cities: Science-Based Diversity, Specialization and Localized Competition,” European Economic Review 43, 1999, 409–429. FELDMAN, Maryann P.; KOGLER, Dieter F. Stylized facts in the geography of innovation. In: Handbook of the Economics of Innovation. North-Holland, 2010. p. 381-410. FINGLETON, Bernard; SZUMILO, Nikodem. Simulating the impact of transport infrastructure investment on wages: A dynamic spatial panel model approach. Regional Science and Urban Economics, v. 75, p. 148-164, 2019. GARCIA, R., ARAUJO, V., and MASCARINI, S. The role of geographic proximity for university- industry linkages in Brazil: An empirical analysis. Australasian Journal of Regional Studies, 2013, 19(3), 433. GARCIA, R., ARAUJO, V., MASCARINI, S., GOMES SANTOS, E., and COSTA, A. Looking at both sides: how specific characteristics of academic research groups and firms affect the geographical distance of university–industry linkages. Regional Studies, Regional Science, 2015, 2(1), 518-534. GIBBONS, S., LYYTIKÄINEN, T., OVERMAN, H. G., & SANCHIS-GUARNER, R. New road infrastructure: the effects on firms. Journal of Urban Economics, 2019, 110, 35-50. GLAESER, E. L., and GOTTLIEB, J. D. The wealth of cities: Agglomeration economies and spatial equilibrium in the United States. Journal of Economic Literature, 2009, 47(4), 983–1028. GONÇALVES, Eduardo; ALMEIDA, Eduardo. Innovation and spatial knowledge spillovers: evidence from Brazilian patent data. Regional Studies, v. 43, n. 4, p. 513-528, 2009. HOLL, Adelheid. Highways and productivity in manufacturing firms. Journal of Urban Economics, v. 93, p. 131-151, 2016. HONG, J., CHU, Z., & WANG, Q. (2011). Transport infrastructure and regional economic growth: evidence from China. Transportation, 38(5), 737-752. JAFFE, A.B., TRAJTENBERG, M. AND HENDERSON, R. Geographic Localization of Knowledge Spillovers as Evidenced by Patent Citations. Quarterly Journal of Economics, 1993, 63, pp. 577- 598. JIWATTANAKULPAISARN, Piyapong; NOLAND, Robert B.; GRAHAM, Daniel J. Causal linkages between highways and sector-level employment. Transportation Research Part A: Policy and Practice, v. 44, n. 4, p. 265-280, 2010. LAURSEN, K., REICHSTEIN, T., and SALTER, A. Exploring the effect of geographical proximity and university quality on university–industry collaboration in the United Kingdom. Regional Studies, 2011, 45, 507–523. LEE, Neil; RODRÍGUEZ-POSE, Andrés. Innovation and spatial inequality in Europe and USA. Journal of economic geography, v. 13, n. 1, p. 1-22, 2013. LESAGE, James; PACE, Robert Kelley. Introduction to spatial econometrics. Chapman and Hall/CRC, 2009. LIU, Shuli; WAN, Yulai; ZHANG, Anming. (2020). Does China’s high-speed rail development lead to regional disparities? A network perspective. Transportation Research Part A: Policy and Practice, v. 138, p. 299-321, 2020. LUNDVALL, B. Å., JOHNSON, B., ANDERSEN, E. S., and DALUM, B. National systems of production, innovation and competence building. Research policy, 2002, 31(2), 213-231. MAZZUCATO, M. The Entrepreneurial State: Debunking Public vs. Private Sector Myths. Anthem Press, 2013. MEDEIROS, Victor; RIBEIRO, Rafael Saulo Marques. Power infrastructure and income inequality: Evidence from Brazilian state-level data using dynamic panel data models. Energy Policy, v. 146, p. 111734, 2020. MEDEIROS, Victor; RIBEIRO, Rafael Saulo Marques; DO AMARAL, Pedro Vasconcelos Maia. Infrastructure and household poverty in Brazil: a regional approach using multilevel models. World Development, v. 137, p. 105118, 2020. MUSCIO, A. University–industry linkages: What are the determinants of distance in collaborations? Papers in Regional Science, 2013, 92, 715–739. NAYA, H., Urioste, J. I., CHANG, Y. M., RODRIGUES-MOTTA, M., KREMER, R., & GIANOLA, D. (2008). A comparison between Poisson and zero-inflated Poisson regression models with an application to number of black spots in Corriedale sheep. Genetics Selection Evolution, 40(4), 1-16. NELSON, R. Capitalism as an engine of progress. In: NELSON, R. (Ed.). The sources of economic growth. Massachusetts: Harvard University Press, 1990. OTTAVIANO, Gianmarco IP. Infrastructure and economic geography: An overview of theory and evidence. EIB papers, v. 13, n. 2, p. 8-35, 2008. QI, G., SHI, W., LIN, K. C., YUEN, K. F., & XIAO, Y. (2020). Spatial spillover effects of logistics infrastructure on regional development: Evidence from China. Transportation Research Part A: Policy and Practice, 135, 96-114. RAPINI, Márcia Siqueira; CHIARINI, Tulio; SANTOS, Ulisses Pereira. Interação de grandes empresas com universidades no Brasil: Evidências a partir da Pesquisa “Sondagem da Inovação”. In: Experiências de interação universidade-empresa no Brasil. Cedeplar, Belo Horizonte, Brazil, 2018. RODRÍGUEZ-POSE, Andrés; CRESCENZI, Riccardo. Research and development, spillovers, innovation systems, and the genesis of regional growth in Europe. Regional Studies, v. 42, n. 1, p. 51-67, 2008. SAIZ, A. The geographic determinants of housing supply. Quarterly Journal of Economics, 2010, 125(3), 1253–1296. SANTOS, Ulisses Pereira. Spatial distribution of the Brazilian national system of innovation: an analysis for the 2000s. CEPAL Review, 2017. STAUB, Kevin E.; WINKELMANN, Rainer. Consistent estimation of zero‐inflated count models. Health economics, v. 22, n. 6, p. 673-686, 2013. STRAUB, Stéphane. Infrastructure and Development: A Critical Appraisal of the Macro-level Literature, The Journal of Development Studies, 2011, 47:5, 683-708, DOI: 10.1080/00220388.2010.509785. SUZIGAN, W., ALBUQUERQUE, E., GARCIA, R., and RAPINI, M. University and industry linkages in Brazil: some preliminary and descriptive results. Seoul Journal of Economics, 2009, 22, 591–611. TRIPPL, M.; SINOZIC, T.; LAWTON SMITH, H. The Role of Universities in Regional Development: Conceptual Models and Policy Institutions in the UK, Sweden and Austria. European Planning Studies, v. 23, n. 9, p. 1722–1740, 2 set. 2015. VARGA, A. Local Academic Knowledge Transfers and the Concentration of Economic Activity. Journal of Regional Science, 2000, 40(2), pp. 289-309. WAN, Guanghua; ZHANG, Yan. The direct and indirect effects of infrastructure on firm productivity: Evidence from Chinese manufacturing. China Economic Review, v. 49, p. 143-153, 2018. WANG, X., XIE, Z., ZHANG, X., and HUANG, Y. Roads to innovation: Firm-level evidence from People's Republic of China (PRC). China Economic Review, 2018, 49, 154-170. ZHANG, Yin-Fang; JI, Shengbao. Infrastructure, externalities and regional industrial productivity in China: a spatial econometric approach. Regional Studies, v. 53, n. 8, p. 1112-1124, 2019.