Keizo Mizuno (p), Kazuhiko Mikami

We examine firms' incentives for joint ventures and their performances in a simple theoretical framework. Our model sheds light on the following four aspects. First, the likelihood of a joint venture is endogenously determined by comparing the profit under a joint venture with that in a non-cooperative investment regime. Second, we allow firm heterogeneity in terms of production efficiency and all member firms have to agree upon the size of a facility. Third, each member firm has an opportunity to make a technology choice before or after the establishment of a joint venture. Fourth, when we consider the relationship between a joint venture and firms' technology choices, the timing of these two investment decisions matters because the decision made in the first stage plays a role as a strategic commitment device. Two scenarios are analyzed with respect to the timings of two kinds of investment.
To shed light on the above four aspects, we build a model in which firms have an opportunity to undertake a joint venture to build a new upstream facility that has a demand-enhancing effect with a fixed facility cost. Moreover, firms can choose a downstream production technology non-cooperatively; this affects the cost share of upstream investment each firm must incur under the joint venture. The cost share of upstream investment is determined by a cost-sharing rule under which neither firm disagrees on cost sharing or on the size of a new upstream facility. Then, firms can choose their downstream technology non-cooperatively before or after undertaking a joint venture to build a new upstream facility.
Analyzing this model, we first show that when firms undertake a joint venture for demand-enhancing investment, they have less incentive to adopt an efficient technology than they have in a non-cooperative investment regime. Second, we show that a joint venture is more likely to occur in the scenario of strategic technology choice (i.e., the production technology is strategically chosen before the decision to undertake a joint venture) than in the case in which a firm's technology is exogenously given. By contrast, the likelihood of a joint venture does not depend on the magnitude of the fixed facility cost in the scenario of strategic cooperation (i.e., when the decision on the undertaking of a joint venture is strategically determined before a firm's technology choice).

Vinko Mustra (p), Blanka Šimundić, Zvonimir Kuliš

Analysing the current state of the EU's economic, social and territorial cohesion, the last Cohesion Report (EC, 2017a) has raised the importance of belonging to different “development clubs” characterised not only by different income levels but also by different structural features (Iammarino et al, 2017). On this matter, the paper focuses on the group of the regions facing a particular challenge, so-called “middle income trap”.
More precisely, authors after the identification of these regions in EU analyze the relevance of the policy that combines innovation with specific strengths of the regional economy, smart specialization policy. Such specialisation, that is more about fostering human capital formation and the knowledge diffusion processes and avoiding automatically prioritizing high-technology sectors (Mccann and Ortega-Argiles,2015) may fit more to the specific characteristics and needs of the “middle income trap” regions.
Considering that measuring smart specialization outcomes is quite problematic (e.g. Foray, David and Hall, 2011, Boschma 2014, Morgan 2015, Iacobucci and Guzzini, 2016; Santoalha, 2016, Balland et al. 2017) in empirical part of the paper we have decided to introduce the new approach. Instead of trying to do “impossible task” and construct the measure for smart specialization outcomes, we focus on constructing the measure which integrates important features of smart specialization logic (Gianelle et al , 2016, EC, 2017b)
By introducing the new proxy for smart specialization the results show significantly different effects of the smart specialisation logic among different “development clubs”, especially for “middle income trap” regions. On this matter, the results should stimulate policy makers in raising their consciousness about the perspective of “middle income trap” regions.

Francesco Perugini (p), Valentina Giannini , Donato Iacobucci

The main aim of the paper is to analyze recent trends of innovation strategies in the EU textile and clothing industry (TCI) and the role of local variety in the adoption of open innovation strategy.
Despite the importance of the TCI in many European countries, there is a lack of studies on the innovation performance in this industry. Indeed, it is recognized that firms in the TCI are considered as ‘suppliers dominated’ for their innovation model, being characterized by low investment in R&D and little capabilities for autonomous innovation. However, this situation is rapidly changing given the increasing importance of so-called transversal technologies (such as ICT) and the diffusion of open innovation models. Moreover, literature on agglomeration economies suggests that there is a close relationship between related variety at local level and the innovation performance of firms. As a result some of the questions, such as the role of variety for regional innovativeness and firm capacity to innovate, are still unexplored.
Our work directly bears on these questions, and provides an empirical analysis of the role of variety and intensity of internal and external knowledge on firms’ innovative capability in the EU TCI. The main hypotheses is that open innovation strategies are gaining importance for the innovation performance of firms and that the characteristics of the local system influence their ability to innovate, as a result of the relations with firms in different sectors.
We use information on patent activities of European firms from the ORBIS database and we exploit information from the REGPAT database in order to characterize the technological capabilities at regional level. Data refers to about 1,900 firms in the TCI owing about 10,400 patents. Our methodological approach builds upon the large literature analyzing the impact of variety on firm patent activities. We estimate the impact of related and unrelated variety of European regions’ knowledge structure on the patenting activity of firms, and we analyze how the characteristics of the local context determines the innovation and financial performance of firms.
Preliminary results are in accordance with main hypotheses: 1) there is an increasing role of innovations based on the absorption of knowledge from outside the TCI; 2) the ability of firms to use different technological knowledge depends also on the characteristics of the local system (the variety of the local system seems more important for the innovative performance of firms than the belonging to a specialized cluster).

Adriana Carolina Pinate (p), Claudio Di Berardino

Over last years, the evolutionary economic geography literature has confirmed that variety in related activities enhance employment growth and innovation. Since the concept related variety (Frenken et al., 2007) was introduce a crescent number of scholars has analyzed the effect of related variety on regional growth, leaving rather implicit the mechanism ‘how’ variety in related activities leads with growth and export specializations (Content & Frenken, 2016). Nevertheless, recent findings suggest that the effects of related variety on regional growth differ across industries (Bishop & Gripaios, 2010; Mameli et al., 2012; Cortinovis & Van Oort, 2015; Caragliu et al., 2016, Innocenti & Lazzeretti, 2017) being limited to a high-tech sectors only (Hartog et al., 2012; Content & Frenken, 2016). However, empirical studies that support this theoretical line are yet scarce.
The present study provides insights into this issue by adding evidence to understand how related variety sectorial differentiation impact on regional employment growth in Italian province over the period 2001 and 2011 by a cross-sectional regression model, focusing on manufacturing sector and paying particular attention to control variables. Inspired by the approach of Hartog et al. (2012), we decomposed related variety among the four categories of Pavitt taxonomy ‘science based’, ‘specialized supplier’, ‘scale intensive’ and ‘supplier dominated’ industries, to assess whether they have a different impact on regional employment growth.
This work present important novelties since by the first time it’s implement Pavitt taxonomy to analyze the difference effects on growth of related variety. Also, the introduction of trademarks as control variable output of innovation (Mendonça et al., 2004), being the first empirical evidence on literature that use of this propriety right a regional level and to measure the effect on regional employment growth.
The findings show that after decomposed related variety the ‘specialized supplier’ and ‘supplier dominated’ industries have a positive and significant impact on employment growth. Instead, ‘science based’ industries had not effect on employment growth. This result indicate that related variety is not limited to high-tech sectors, as Hartog et al. (2012) work found. The use of trademarks as control variable show a positive an significant effect on employment growth, this result indicate an interesting opportunity for using trademarks as indicators of innovation and regional growth. The findings are robust ones controlled by spatial autocorrelation.