Introduction: the empirical puzzle The global automotive industry is dominated by a limited number of multinational vehicle manufacturers. Within these firms, there is a continuous battle over where to locate R&D and production. For reasons such as national import restrictions and logistics, production tends to be located in several different countries. By contrast, automotive R&D tends to be more concentrated. This is because it is based on a systemic knowledge base, comprising both tacit and codified pieces of knowledge (Breschi and Malerba, 1997). Effective coordination under such circumstances relies on territorial proximity. Still, the global automotive manufacturers must ensure that their R&D maintains a connection with regional market requirements. Being a small country with a limited domestic market, Swedish car production is marginal from a global perspective: in 2007, the production of the Swedish car manufacturers Volvo Cars and Saab Automobile amounted to only 0.6 per cent of the total production worldwide (Faguert et al., 2009). By contrast, the Swedish heavy vehicle industry (trucks, buses, etc.) has a remarkably strong position. In 2007 the Swedish heavy vehicle manufacturers Volvo and Scania had as much as 28 per cent European market share and controlled about 10 per cent of the world’s production of heavy commercial vehicles. The automotive industry is very important for the Swedish economy. In 2008 its exports amounted to 13 per cent of total Swedish exports (BIL Sweden, 2009) and besides ICT (information and communication technology), the automotive industry is the largest sector in Sweden in terms of R&D spending (Norgren et al., 2007). Combining combustion engines with electric drives, hybridization has been proposed as one of the most promising routes to reduce vehicular CO2 emissions in a short-to mid-term perspective (Hekkert et al., 2005; Schäfer et al., 2006). Hybrid-electric vehicle technology includes a range of technological options, e.g. in terms of different configurations (series, parallel, power-split), degrees of hybridization (micro, mild, full, plug-in) and in different key subsystems technologies (e.g. energy storage, electric machines, power electronics). In particular, hybrid-electric vehicle technology enables significantly reduced fuel

consumption in drive cycles characterized by frequent acceleration/deceleration, such as urban traffic. Hybridization may also pave the way for further electrification of the transport system, which may result in additional CO2 emission reduction. During the last decade, hybrid-electric vehicle technology has established itself as a viable alternative for passenger cars, a development that has mainly been promoted by Japanese car manufacturers. In contrast, the application of this technology in heavy vehicles is still in an infant stage. However, since many heavy vehicles operate in urban traffic (e.g. inner-city buses, distribution trucks, refuse trucks), there is a significant potential for fuel efficiency gains in these applications. Based on a case study of the development of hybrid-electric vehicle technology in Sweden over the period 1990-2010, this chapter analyses how governance arrangements on different levels can interplay to facilitate innovation systems conducive to sustainable vehicle technologies. The discourse begins in the theoretical foundations on governance and technological innovation systems (TIS). The research design is then explained, followed by our three analytical steps: a historical exposé of the TIS development; scrutiny of how various governance arrangements have influenced the TIS; and the analysis of how different levels of governance have interacted. The final section draws conclusions related to the governance of innovation systems for sustainable automotive technologies.