Abstract
This study assessed whether and how government–business partnerships (GBPs) offer a unique platform that targets profound environmental impacts via the promotion of radical eco-innovation. It applied transactional cost and complementary logics to explain the rationale of GBP formation for radical eco-innovation, and further assessed the operation of GBPs from governance, learning, and rulemaking aspects. This study applied propensity score matching technique to empirically test these theoretical associations using 225 observations representing 166 U.S. firms’ participation in 192 environmental alliances between 1985 and 2013. The study results confirmed GBPs’ role in channeling public and private efforts in pursuing transformative environmental change via the adoption of radical eco-innovation goals. Results highlight four critical elements of GBP operation—effective governance, exploration learning, cognitive learning, and rulemaking—that enable participants to embrace these radical environmental solutions.
Keywords
In recent decades, the magnitude and urgency of significant environmental challenges—natural resource depletion, the grand scale of species distinction, widespread deforestation, increasing desertification, transnational air pollution, and water usage (all related intimately to the meta-environmental issue of climate change; Heintzman, 2006)—require a call for transformative change that bypasses existing routines and initiates new mechanisms of knowledge sharing and invention (Hart & Milstein, 2003). Radical eco-innovation is a means of achieving such transformative change as it involves the development of new products, the formulation of new markets, and identification of new means of sustainable servicing existing markets (Etzion, 2007; Kemp, 1994). Emerging environmental challenges associated with global sustainability may be catalysts for a new round of radical eco-innovation offering unprecedented business opportunities (Hart & Milstein, 1999). For instance, businesses may seek to redesign current product systems, and launch new products and services with fewer or zero environmental impacts.
These sustainability-related business opportunities, in incorporating environmental and social concerns into development of new markets, create public goods with unpredictable private benefits, the promotion of which thus calls for public sector involvement. In recent years, governments have begun fostering more decentralized approaches in policy and production of scientific knowledge in an effort to better accommodate the complexity of environmental decision making (Lalor & Hickey, 2014). They increasingly act beyond the traditional role of regulator (Manring, 2007), becoming active participants in government–business partnerships (GBPs) in an attempt to produce public goods that foster more significant environmental change (Lin, 2014; Van Tulder, Seitanidi, Crane, & Brammer, 2015). Such GBPs are a form of cooperation between government and business sector, which enable them to share resources, risks, and mutual benefits with an eye toward transforming existing environmental practices through eco-innovation (Lin, 2014; Van Tulder & Pfisterer, 2014).
Despite the significance of GBPs for transformative change, the formation and operation for such an endeavor have limited coverage in the literature, as is their in-depth empirical investigation (Roehrich, Lewis, & George, 2014). The extant cross-sector partnership (CSP) literature has investigated CSP formation for sustainability-related improvements (Selsky & Parker, 2005, 2010). This scholarship, however, has its limitations as it concentrates on the collaborative interface between businesses and non-government organizations (NGOs; Bryson, Crosby, & Stone, 2006; Le Ber & Branzei, 2010; Selsky & Parker 2005), and GBPs received relatively limited attention. While a few innovation-related studies have investigated GBPs by means of a case-study design (Kaiser & Kuhn, 2012; Robin & Schubert, 2013), those studies focused on conventional innovation and a limited European Union (EU) context.
However, the growing body of public–private partnership (PPP) literature involves both GBPs and government–NGO partnership. This stream of GBP literature has traditionally taken two directions: either GBPs as an alternative financial mechanism that attracts private investment in public infrastructure, or GBPs as an alternative public service delivery mechanism with enhanced efficiency (Brinkerhoff & Brinkerhoff, 2011; Lin, 2014; Selsky & Parker, 2005; for an overview, Song, Zhang, & Dong, 2016). These traditional GBP arrangements are normally found in sectors where substantial initial capital investments are required, which offer cost recovery, private finance initiative, and risk-sharing opportunities (Bouman, Friperson, Gielen, & Wilms, 2013; Song et al., 2016).
This literature stream thus assesses GBP performance from an instrumental or efficiency gain aspect (e.g., cost savings; Hartwich & Tola, 2007). It indicates that GBP formation is mostly based on resource mobilization motives rather than for effectiveness reasons (Bouman et al., 2013; Brinkerhoff & Brinkerhoff, 2011).
However, GBPs formed in addressing daunting environmental problems may go beyond instrumental aims. GBPs as a deliberated strategy mechanism that champions transformative environmental change are rarely covered in the extant GBP literature. Even though a few studies (Lin, 2014; Stadtler, 2015) have specifically explored the antecedents of GBP formation for environmental improvements and the governance of such partnerships, they have not fully explained how these GBPs are operated, nor whether they encourage partners to pursue more radical innovation goals or simply incremental, and reactive, environmental improvements. The normative aspects of these GBP operations have not been fully analyzed in the extant literature.
Given the complexity and scope of environmental challenges, it is important to assess and understand whether and how GBPs offer a unique platform that targets profound environmental impacts via the promotion of radical eco-innovation. This study attempts to fill this literature gap by systematically assessing the formation and operation of GBPs in pursuit of radical eco-innovation.
In this article, I ask “whether” and “how” GBPs offer a unique platform that targets profound environmental impacts via the promotion of radical eco-innovation. It advances the business sustainability, CSP, and PPP literature by highlighting GBP’s significant and positive role in channeling public and private efforts in pursuing transformative environmental change via the adoption of radical eco-innovation goals. It further advances the alliance and environmental innovation scholarship by explaining “how” GBPs facilitate participants in their pursuit of radical eco-innovation from four operational dimensions—effective governance, exploration learning, cognitive learning, and rulemaking. While these elements have both instrumental and normative aims, my study highlights the normative aspect of GBPs for radical eco-innovation. These elements expand the literature regarding “What is an effective governance” and “What is an effective learning” in alliance pertaining to radical eco-innovation. It highlights the limitation of instrumental means (e.g., financial risk sharing under the traditional GBP infrastructure project), and proposing more normative aspects, such as “reciprocal commitment,” that govern GBP formed for radical eco-innovation, and cognitive learning that help participants build transformational change vision.
The article begins with a literature review of eco-innovation, followed by the application of transaction cost and complementary logics to view the rational of GBP formation for radical eco-innovation. I then illustrate GBP’s four operational elements from instrumental and normative perspective, and came up with five hypotheses. The method session explains the sample, statistic model, measurements, and results. The study concludes by discussing the theoretical and policy implications of the findings.
Theory Development: Formation and Operationalization of GBPs for Radical Eco-Innovation
Typology of Environmental Solutions
Scholars conceptualized eco-innovation as a continuum with reactive and radical eco-innovation anchoring both ends of the spectrum (Aragon-Correa, 1998; Hart, 1995).
Reactive environmental solutions
Reactive environmental solutions respond to changes in environmental regulations and stakeholder pressures via investments in end-of-pipe pollution control measures (Aragon-Correa & Sharma, 2003). End-of-pipe pollution control technologies include a multitude of biological and chemical systems used for treating water, barrier systems used for treating air, and disposal methods for other forms of solid waste (Henriques & Sadorsky, 2005). These solutions usually involve the use of pollution-eliminating or sorting instruments. This is an off-the-shelf compliance-based or legislation-push response (Roome, 1992) mainly utilized to reduce business risks while hoping that the environmental response meets regulatory and community standards (Bansal & Roth, 2000). Such reactive environmental solutions align with, and are enforced by, command-and-control regulatory measures specified by legislation and rule prescribing which environmental technologies are permitted and which are illegal (McManus, 2009).
Incremental environmental solutions
In the 1990s, environmental regulations shifted toward more flexible “cap and trade” schemes. For instance, the U.S. Pollution Prevention Act of 1990 was a national policy decreeing that pollution should be prevented or reduced at the source wherever possible (U.S. Environmental Protection Agency [USEPA], 2012). Pollution prevention technologies extract and use natural resources more efficiently, generate products with fewer harmful components, minimize pollutant releases to air, water, and soil during manufacturing and product use, and involve the design of durable goods that can be reused or recycled (Organisation for Economic Co-Operation and Development [OECD], 1995). Such a scheme encourages better maintenance, material substitution, recycling, and innovative procedures to decrease, or prevent, toxic emissions and waste during manufacturing processes and product use (Hart, 1995). Pollution prevention not only saves costs related to installing and managing end-of-pipe pollution control mechanisms, but it can also increase productivity and efficiency levels because reduced waste translates into a more efficient use of inputs, which results in lower raw material and waste removal costs (Schmidheiny, 1992; Young, 1991).
Radical eco-innovation
Pollution prevention allows firms to maintain their core practices through incremental environmental solutions. Such measures may, however, face constraints when dealing with grand environmental challenges such as deforestation, biodiversity loss, and climate change. Such complex environmental issues have driven demands for clean technologies or radical eco-innovations that produce cleaner fuels and power, developing new methods of sustainability within existing markets, or driving a major transition that alters existing markets or creates new ones (Christmann, 2000; Etzion, 2007; Hoffman, 2005; Sroufe, Curkovic, Montabon, & Melnyk, 2000). This is the most radical environmental solution and is variously referred to as a “leading edge” approach (Roome, 1992) or “business redefinition” (Sharma & Henriques, 2005). There is growing voice within the business community and civil society for decisive corporate changes via the adoption of radical eco-innovation (Heintzman, 2006). Within the government authority, the scope of complex environmental issues has made instituting multi-jurisdictional controls a challenge (Lin & Darnall, 2015). Increasingly, many governmental agencies are going beyond the limitations inherent in regulatory system requirements and collaborating with businesses to foster more radical solutions.
A Risk Perspective of Eco-Innovation
Environmental problems (and the societal expectations that come with them) create decision-making risks. The structure, consequences, and probabilities of many environmental problems are not fully known (Baird & Thomas, 1985). Firm managers are often not fully aware of the extent of the risks, the available options to address them, and how large and immediate the outcomes would result from the impact of those risks (Baird & Thomas, 1985). In the meantime, firms are facing increased societal pressures to adopt radical eco-innovations that will require identification of alternative energy and material, redesigns of their current product systems, and/or the launching of new products and services.
For instance, climate change issues increase business risks, insurance costs, and the possibility of more stringent regulation and public sanctions (Kolk & Levy, 2001). As public concerns about destructive environmental degradation related to climate change (e.g., floods, hurricane, drought) increase, public and civil society (especially environmental NGOs) play increasingly important roles to mobilize public sentiment, alter accepted norms, and impose new roles on the business to phase out fossil fuel economy that radically addresses climate change issues. Yet, undertaking such a transition from fossil fuel to renewable energy (e.g., use fuel cells to replace fossil fuels such as coal, oil, and natural gas as the world’s primary energy source) generates great uncertainty and financial risk for business, as such radical innovations are costly to develop and are limited to commercialization (Lin & Darnall, 2015).
In this context, unresponsive firms may face public scrutiny and regulatory sanctions, while responsive firms may increase their investment risk when undertaking measures to address these environmental problems. These consequences make most firms cautious of pursuing environmental solutions. Consequently, firms may invest in reactive (end-of-pipe) pollution control technologies and only tackle environmental problems after they have occurred, or cautiously invest in pollution prevention approaches when the technologies have proved to be technically and financially suitable for commercial adoption. Firms confront increased investment uncertainty when they advance toward the adoption of radical eco-innovation. Such a radical approach requires significant research and development (R&D) investment for new technologies or products; however, the investment outcomes are distant or highly uncertain, and frequently do not produce the intended payoffs. This radical eco-innovation goal has thus a high chance of termination as a result of “perceived lack of fit with the current strategy and high risk” (Greve, 2007, p. 950).
Comparing the three types of environmental solutions (specifically reactive, incremental, and radical ones), one point of departure is the distinction between reactive/incremental solutions, “which advance existing technology, and radical innovations, which develop new technology” (Greve, 2007, p. 947). The distinction lies in whether the environmental innovation requires extensive and radical modification to firms’ current business domain, operation, and the underlying values, assumptions, and cognitive frameworks. Reactive/incremental eco-innovations focus on maintaining and refining extant business practices and grabbing low-hanging fruits that ameliorate immediate environmental concern. In contrast, radical eco-innovation “anticipates future regulations and social trends, and designs or alters operations, processes, and products to prevent” (rather than merely ameliorate) negative environmental impacts (Aragon-Correa & Sharma, 2003, p. 73). Such a distinctive feature results in different environmental impacts for adopting firms.
The required scale of change (reactive/incremental vs. radical) related to the continuum of eco-innovation extends different levels of risks and cost–benefit structures to focal firms. In this context, pursuing these varied environmental solutions reflects firms’ stance toward the natural environment and the extent of risks they are willing to embrace in addressing environmental problems. Transaction cost and risk-sharing logic become relevant in justifying the rationale of GBP formation for radical eco-innovation.
GBPs Formation for Radical Eco-Innovations: A Transaction Cost Logic
Public goods are non-rival and non-excludable, in that individuals cannot be effectively excluded from use and where use by one individual does not reduce availability to others (Varian, 1978). However, private goods are rival and excludable, and they yield positive private benefits (Walter, 2004). In between private and public goods are mixed goods (or impure public goods) that have elements of both public and private goods (Ramesh, Nagadevara, & Naik, 2010). They include common goods that are rival but not excludable. Its supply can be depleted, but people are not restricted in their use of the good, which are susceptible to the Tragedy of the Commons (Hardin, 1968). Example includes fish in the ocean. Mixed goods also include club goods that are non-rival but are excludable. Examples include highway and membership of an industry club.
Environmental issues create negative externalities, and the states and the markets are two of society’s main mechanisms for coordinating activity related to the treatment of these hazards (Coase, 1960). As addressing these issues provides both public and private benefits, Coase (1960) used the proper allocation of cost and benefits related to the treatment of environmental externalities to determine whether the market or state should be in charge of the treatment (Ramesh et al., 2010). Building on Coase (1960), Ramesh and colleagues (2010) suggest that when environmental treatment generates public goods, it will be the responsibility of public organizations. For instance, the tackling of many complex transboundary pollution problems such as chlorofluorocarbons (CFCs) produces public goods, as the benefits (costs) derived from limiting (not limiting) the pollutant are non-rival and non-excludable (e.g., to a set of nations; Murdoch & Sandler, 1997). Therefore, addressing these issues will be the responsibility of public sector as it is difficult to charge for individual’s welfare gain associated with the CFCs reduction (Elsig & Amalric, 2008).
However, if private goods are derived out of the environmental treatment, private organizations are expected to be in charge (Ramesh et al., 2010). For instance, reactive eco-innovation tends to create private goods as firms that “fail to adhere to command-and-control regulatory mandates risk obtaining unwanted negative media attention, community scrutiny, consumer complaints, and public boycotts” (Lin & Darnall, 2015, p. 551). In such instances, firms mainly focus on their own environmental hazards, risks, and threats when pursuing reactive eco-innovation approach, which derives direct compliance-related private benefits. According to Coase (1960), it is possible to let private sectors handle such reactive environmental solutions directly due to their stronger connection to private interests. For instance, in the late 1980s and early 1990s, many inter-firm alliances were formed in the United States to spread pollution control product and services in response to the command and control legislations (SDC Platinum, 2013).
Nevertheless, in some environmental solutions the distinction between public and private goods is not very clear. These are deemed mixed goods, and hence to be provided by GBPs to meet dual public and private benefit purposes (Ramesh et al., 2010). Radical eco-innovation tends to generate such mixed goods, which could be club goods that are non-rival but are excludable (e.g., clean energy generated through wind-turbine that benefits certain users). It could also be common goods when it generates positive externalities (e.g., research spillovers) and alternative environmental solutions that are rival bur not excludable to the public.
The mixed goods generated by radical eco-innovation have both public and private goods elements. In the existence of negative externalities, the cutting-edge solutions created through radical eco-innovation may allow firms to give back to society at large by driving significant ecological and societal change (public goods). In the meantime, these solutions may offer tremendous private benefits as they manage to increase brand value and shift environmental and social concerns into new markets. They enable businesses to outperform their rivals and achieve first-mover advantage (private goods). However, the social value of radical eco-innovation, partially due to the existence of research spillovers, may lie above its private return. As such, compared with reactive/incremental eco-innovation, radical eco-innovation requires greater resource commitments, is more explorative in nature, is riskier to pursue, and is thus hard to fully internalize the private benefits of their R&D efforts. This external effect and the enhanced business risk leads to underinvestment in eco-innovation from a social point of view, and thereby justifies governmental intervention and involvement.
Building on the transaction cost framework articulated by Coase (1960) and Ramesh and colleagues (2010), GBP is superior to both the government bureaucracy and competitive markets in promoting mixed goods such as radical eco-innovation, where public actors will be involved because of the positive externalities but would not do it alone because of internal inefficiencies. Similarly, private sectors will be involved, provided they receive the necessary returns, but would be reluctant to undertake these activities on their own because they might not give them sufficient private benefits (Ramesh et al., 2010).
GBPs Formation for Radical Eco-Innovations: A Complementary Logic
In addition to transaction cost logic, other GBP scholars justify GBP formation from a complementary logic, where the multi-actor partners from business, government, and civil society sectors need to join forces to, integrated resources and solutions required by the scope and nature of the problems being addressed (Brinkerhoff & Brinkerhoff, 2011). Bouman and colleagues (2013) recognized the comparative advantages as well as the differences in incentives between public and business sectors that, whereas business partners “tend to be more capable in providing incentives to maintain productivity (compared to their public counterparts), the public sector is usually better equipped to account for collective externalities” (p. 48). GBPs may become a preferred option to balance these mixed motives in a cost-effective manner.
Moreover, GBPs operate in between the market and state to undertake deliberate actions that initiate, shape, and/or enforce a radical innovation path diverging from current practices and routines. Given the complexity and scope of environmental challenges, GBPs have the potential to generate rapid and far-reaching environmental change impacts, partially due to the complementary logic when the public and private partners share critical resources (e.g., power, funding, technical expertise) that cannot be easily exchanged by means of market transactions. This setting thus offers unique advantages as government partners can provide firm partners better access to decision makers and opinion makers, and they have better bargaining skills, reputations, coalition building abilities, and possess political entrepreneurship (Frynas, Mellahi, & Pigman, 2006). Government partners may provide incentives and infrastructures in a way that can help firms develop, deliver, promote, and evaluate various policies, programs, or initiatives, which may accelerate the search for and adoption of radical innovation.
In this context, firms aiming to launch radical eco-innovation increasingly turn to governments for additional policy endorsement, financial subsidies, and technical support. Developed nations, especially Organization for Economic Cooperation and Development (OECD) countries, tend to provide public support for innovative activity in the business sector through an appropriate mix of direct and indirect instruments such as tax credits, direct support, and well-designed GBPs (OECD, 2007). GBPs for inventions, since legalized by the EU and Northern American policy makers in the mid-1980s, have become popular over time (Caloghirou, Ioannides, & Vonortas, 2003), and gradually evolved to the principal form of research cooperation (Hagedoorn, 2002; Kaiser & Kuhn, 2012). Related to the R&D for radical eco-innovation, many developed nations have increasingly leveraged GBPs to promote clean technology and renewable energy.
The Operation of GBP for Radical Eco-Innovation
The dual purpose (public and private) of GBP brings both comparative advantage and challenge. In addition to GBP acting as a catalyst for attracting private funding and the skills of the private sector, such a cross-sector collaboration brings complexity to partnership design and operation beyond the metrics of efficiency (transaction cost) and synergy (complementarity; Bouman et al., 2013). Common challenges with dual structure and purpose of GBP relate to developing a GBP design that allows aligning the partners’ different interests and priorities (Huxham & Vangen, 2000; Stadtler, 2015). To this end, it is important for GBP partners to negotiate and design the operational elements of the partnership based on the dual partnership structure, manifesting through informal and formal agreements on partnership objectives, missions, activities, funding, governance, resource and risk sharing, and distribution of benefits (Bouman et al., 2013; Bryson et al., 2006; Hartwich & Tola, 2007).
The dual structure and purpose of GBP suggest that “the rationales for government partnerships with the for-profit private sector encompass both instrumental and normative aims” (Brinkerhoff & Brinkerhoff, 2011, p. 5). From an instrumental perspective, GBP may follow a governance structure that channels resource commitment and risk sharing. For infrastructure GBPs that are contracted under build–operate–transfer (BOT) or build–own–operate–transfer (BOOT) 1 agreement, gaining access to private financing and public–private risk sharing is the main drivers (Brinkerhoff & Brinkerhoff, 2011). Such GBP setting also allows complementary resource sharing and technical expertise and established networks access (Brinkerhoff & Brinkerhoff, 2011) that enable exploration learning.
In addition, governing GBPs formed for radical environmental innovation may require more normative aspects that “contribute to a broader, more strategic vision of the public good” (Brinkerhoff & Brinkerhoff, 2011, p. 5). This aspect should go “beyond the metrics of efficiency, effectiveness, and synergy” to act on principles, accountability, ethical behaviors, and paradigm that respond to the question of “effectiveness for whom?” (Provan & Kenis, 2007, p. 229). I thus hypothesized the operation of cognitive learning in GBP that helps build shared vision that is consistent with its ultimate social environmental goals. Another normative dimension of GBP operation relates to rulemaking, associated with public good purpose of GBP that seeks to scale up radical environmental change to the industry and system level. The following hypothesis development explains these operational elements in detail.
Effective governance
One aspect of GBP operation relates to governance that enables GBP partners to coordinate and monitor behavior through a set of rules, norms, and trust (Bryson et al., 2006). GBP provides scope for deriving substantial benefits including for sharing risks more efficiently. For infrastructure GBPs that access private financing, public–private risk sharing is one of the drivers (Brinkerhoff & Brinkerhoff, 2011; Song et al., 2016). To embark on the highly risky radical eco-innovation, GBP needs effective governance that ensures both resource commitment and flexibility in innovation search. Strategic alliance governance literature has outlined a continuum of governance structure, with formal equity ownership and informal trust anchoring both ends of the spectrum. This spectrum of governance encompasses a trade-off between degree of control and flexibility that are both important for radical eco-innovation.
There are variances of the contractual mechanism within the formal governance, which range from equity alliances (e.g., joint venture) to non-equity alliances (e.g., joint R&D; Dacin, Oliver, & Roy, 2007; Gulati, 1995b). A formal equity structure takes more time to negotiate, more resource to maintain, and their rigid governance structure is seldom feasible for innovation development (Linnarsson & Werr, 2004). An increasing number of studies have related firms’ innovation (e.g., new product development) to their choices of non-equity structure (Kogut, 1988; Kok & Creemers, 2008), due to the flexibility this type of structure allows for firms to engage in exploration. Therefore, GBP for eco-innovation may choose a non-equity alliance structure to maintain firm partners’ independence, innovativeness, and flexibility, so that they can respond quickly to changing market conditions in pursuing innovation.
While the non-equity structure provides the flexibility needed for GBPs to pursue radical eco-innovation, such structure also increases opportunism risks due to its relatively weak governance control (Lin & Darnall, 2015). In this context, alliance literature suggests an investment-based (Williamson, 1983) and a trust-based (Gulati, 1995a) mechanism as alternative self-enforcing safeguards.
According to Williamson (1983) and Dyer and Singh (1998), investment-based mechanisms are examples of effective governance created intentionally to control opportunism “by aligning the economic incentives of the transactors” (p. 669). In case of GBP, a mutual resource commitment or “economic hostages” from the government and business partners could act as an effective governance structure. These hostages may include “the commitment of nonrecoverable, symmetric investments in specialized or cospecialized assets, which constitute a visible collateral bond that aligns the economic incentives of exchange partners” (Dyer & Singh, 1998, p. 669). Specifically, such economic hostages may incentivize government and business partners to collectively engage in value-creation initiatives (radical eco-innovation) as opportunism will decrease the common investment in value, whereas proper collaboration creates collective value (Dyer & Singh, 1998). Moreover, such a governance model diminishes the need of an expensive and rigid equity control, while providing the flexibility and commitment needed for radical eco-innovation.
For example, in the United States, the National Renewable Energy Laboratory, which is part of the U.S. Department of Energy, has signed more than 300 cooperative research development agreements with companies to research wind, solar, and energy conservation, as well as other initiatives. Its partnerships with Boeing, Shell Solar Industries, and Siemens Solar are engaged in leading edge research to develop thin-film solar electricity technologies. 2 Such radical innovation makes it faster and cheaper to create films that gather sunlight to produce electricity compared with conventional solar panels. To properly incentivize and govern these alliances, these GBPs establish a visible collateral bond between the government and business partners: For every dollar of public investment from a government source, these GBPs receive US$6 in business sector resources to jointly accelerate the technology. Such a governance structure mobilizes government and business investment, as well as direct R&D support toward clean technology projects that hedge the business sector’s investment risks related to the exploration of eco-innovation.
However, GBP formed for radical innovation may require a governance form that is beyond resource sharing, as investment hostage in such projects may not be the indispensable means or may not work as effectively as the infrastructure-oriented GBP projects. Referring to Vieira and Hartwich (2002), Hartwich and Tola (2007) argue about the difficulty in ensuing “real sharing of resources, knowledge, risk, and funding” in practice “because each partner’s benefits depend on the other partners’ commitment and input.” For this kind of GBP, a trust-based, reciprocal commitment could be a more important governance mechanism that helps increase the synergy resulting from the joint use of complementary resources (Hartwich & Tola, 2007).
Based on the above analysis, GBP may lend itself to an informal self-enforcing mechanism—a non-equity, investment or trust-based governance structure—to ensure efficiency (transaction costs reduction) and effectiveness in mobilizing mutual commitment toward radical eco-innovation goals. I thus propose the following:
Strategic alliances provide platforms for learning (Kogut, 1988) through which partners may enhance their collective ability to accept, make sense of, and respond to internal and external change (Cyert & March, 1963). Such learning may be reflected in organizational behavior change in routines, procedures, processes, and actions, and/or cognitive change in understanding and beliefs (Arya & Salk, 2006; Fiol & Lyles, 1985). I used exploration and cognitive learning to capture the co-learning and capacity building in GBPs in their pursuit of radical eco-innovation goals.
Exploration learning
In view of the uncertainty involved in developing innovative eco-friendly technology, products, and services, radical eco-innovations are a form of organizational exploration. Exploration learning, emphasizing the development of the new and unknown, may offer firms a chance to break away from the extant knowledge path and a shift to a different technological trajectory (Benner & Tushman, 2002, p. 679; March, 1991). Exploitation learning, in contrast, emphasizes things already known, thereby helping the partners maximize their returns from knowledge developed in the past (March, 1991). Compared with returns from exploitation, returns from exploration are highly “uncertain,” “more remote in time,” “more distant from the locus of action” (Lavie, Stetter, & Tushman, 2010, p. 116; March, 1991).
GBPs are likely to be leveraged for exploration learning in their pursuit of radical eco-innovations. GBPs bring partners together from the government (e.g., lab) and business (e.g., corporate research center) sector, thereby increasing their capabilities’ complementarity regarding exploration. This partner structure allows the combination of heterogeneous competencies and perspectives to stimulate organizational learning, and create the “technology fusion” that has become increasingly important in the search of innovation (Sakakibara, 1997). During this process, multiple actors are involved in technological paths emerging in real time (Pinch & Bijker, 1987). These actors generate learning while experimenting with new technologies (Garud & Karnoe, 2003). GBPs provide firms with the framework under which to undertake such experiments.
Furthermore, in GBPs, government and business partners are less concerned about competitive risks and knowledge leakage (Gimeno, 2005), which may boost the exchange of tacit knowledge and enhance GBP partners’ likelihood to conduct exploration learning. As such, partners are likely to share valuable information, knowledge, technology, people, and other critical resources. For example, the Advanced Energy Centre (AEC) is a Canadian GBP that works with Canadian clean-tech start-ups and SMEs with the mission to foster innovative energy technologies in Ontario and Canada. Within this GBP, partners extensively shared knowledge and expertise; and the human resources were highly mobile. Staff members communicated intensively, acting as both a bridge and a think tank by pooling their public knowledge and business expertise to explore alternative solutions for complex environmental problems. Such intensive interaction and communication increase the likelihood of knowledge spillovers and explorative learning in GBPs. Based on the above analysis, I suggest the following:
Cognitive learning
Compared with reactive/incremental eco-innovation, radical eco-innovation targets complex environmental problems, is more difficult to design, and exhibits greater market risks and political uncertainty. This creates conceptual barrier for business managers who may view the pursuing of radical eco-innovation a risky and expensive initiative. Government sector may not understand business sector’s financial insecurity and private interest. Differing cultures, values, missions, and cognitive framework cross-government and business partners add to the difficulties in reaching a shared vision and common understanding. Facilitation of radical eco-innovation thus calls for cognitive learning, which involves the facilitation of partners to develop new understandings of surrounding events, and develop different interpretations of new and existing information (Sharma & Vredenburg, 1998). Such learning may help involved organizations to build awareness and understand the urgency to launch transformative change in addressing grand-scale environmental challenges. Such a learning with purpose may eventually help develop shared vision across partners to reach a common goal, resulting in the development of knowledge and problem-solving skills related to radical eco-innovation.
Cognitive learning in GBPs can take multiple means, including education, training, and awareness enhancement workshops and seminars. Moreover, cognitive learning can be conducted over regular alliance activities. Building on Lavie and colleagues’ (2010) findings, maximizing the value of existing resources in an exploitation alliance means that there is less need for intensive interaction regarding knowledge creation and transfer. In contrast, the joint knowledge and capability building in exploration alliances require close interaction, which exposes firms to their partners’ norms, values, cognitive thinking, and mind-set. This intensive interaction, communication, and co-learning between government and business partners help develop strong personal connections and relationships. Such interactions facilitate cognitive learning, wherein managers, through their association with their government partners, gradually develop different interpretations of environmental problems and the means to address them.
This cognitive learning allows managers to grow awareness of the urgency to pursue radical eco-innovation to address grand-scale environmental challenges, and recognize the associated private benefits and market opportunities. These managers may experience major cognitive reorientations involving changed norms, values, worldviews, and frames of reference, become more positive in their thinking and approach, and develop shared sustainability vision with government partners. This mind-set shift through cognitive learning remains an incentive for GBP partners to discover new products and markets, which may, in turn, facilitate their ability to create transformative environmental changes through the pursuit of radical eco-innovation. As such, I suggest the following:
While exploration and cognitive learning in GBPs may stimulate the emergence of radical eco-innovation, the spread and enforcement of such an innovation path rely on rulemaking.
Rulemaking
The role of power and politics in innovation has been thoroughly researched (see Frost & Egri, 1991; Hardy & Dougherty, 1997; Simon-Lee, 2015). GBPs are likely to be involved with policy sanctions and rulemaking, mainly due to the power and politics resource controlled by the government partner, which include information, know-how, funding, network, rewards, and the ability to impose sanctions. As the government controls the critical resources on which the business sector depends, managers who know the importance of power tools may seek government partners when they aim to initiate a new eco-innovation path. In parallel to the power resource of government, the urgency of grand-scale environmental challenges and the necessity to trigger transformative environmental change also calls for government to play stronger roles in directing and enforcing a radical eco-innovation path through the mechanism of GBPs.
In such a context, profound environmental impacts could be achieved by the deliberate actions of GBP partners to develop new rules or policy that transform corporate operation, the industry landscape, or even institutional environment. For example, the San Diego city alliance is a typical GBP with a policy focus, where corporate partners (GE and CleanTECH San Diego) obtained policy support from the city government to modify existing rules related to the commercialization of electric vehicles. As a result of rulemaking effort of this GBP, the city government evaluated city policies and refined existing disincentives to innovation and urged real-estate businesses to develop infrastructure (e.g., electric vehicle chargers) that help speed up the adoption of electric vehicles. Policy changes, in turn, encourage firms to devote more resources into this GBP for radical innovation (SDC Platinum, 2013).
In moving radical eco-innovation into new markets through rulemaking, businesses need both preferential policies from government and marketing (e.g., branding) support from the civil society. It thus calls for tri-sectors—government, business, and civil society—to collaborate toward such a common cause (Hart, 2005). Such a tri-sector GBP offers various opportunities for businesses to gain access to diverse perspectives (Selsky & Parker, 2005) and knowledge from an extensive network, and enables them to work constructively with representatives from the state and civil society actors. This tri-sector GBP platform helps launch policy dialogue, put issues to the agenda (Elsig & Amalric, 2008), and enable businesses to lobby for policy change that is in favor of radical eco-innovation. Based on the above analysis, rulemaking expedites radical eco-innovation and stimulates the search of transformative change possibilities, thereby lending a GBP an edge in initiating and/or shaping a radical innovation path.
In sum, the above transactional cost and complementary logic explain the rationale of GBPs formation for radical eco-innovation. The follow-up explanation of GBP operation from governance, exploration learning, cognitive learning, and rulemaking aspects further illustrates how GBPs help shape and institutionalize a radical eco-innovation pathway with significant environmental change impact. Based on the above analysis, I posit the following:
Method
Sample
I tested the hypotheses on the subset of environmental alliances listed in Thomson’s Platinum SDC database between 1985 and 2013. The SDC database centers on publicly declared strategic alliances and joint ventures, and includes agreements in which two or more organizations have pooled their resources to create a new, mutually beneficial business arrangement to realize individual and collective objectives. More specifically, this database also incorporates agreements between NGOs, universities, and different levels of government. To search for environmentally related alliances in the Thomson SDC, this study used two search elements: alliance venture economics and industry codes (VEIC) and alliance activity code. Content analysis is conducted to determine whether the alliance had an environmental activity component, which derived a sample of 921 environmental alliances formed between 1985 and 2013.
I also built on firm-based financial data from Compustat and firm-based environmental concern data from the KLD Research & Analytics, Inc. (KLD) database. KLD data are commonly used measures of corporate social performance (Sharfman, 1996). Despite ongoing controversy, Chatterji, Levine, and Toffel (2009) found KLD’s environmental “concern” ratings to be fairly good summaries of firms’ past environmental performance. Combining SDC and Compustat databases derives a sample of 411 observations representing 341 U.S. firms’ participation in 348 environmental alliances formed between 1985 and 2013. However, the combination of SDC, Compustat, and KLD databases arrives at another sample of 225 observations, representing 166 U.S. firms’ participation in 192 environmental alliances formed between 1985 and 2013. Every firm’s participation in an environmental alliance is counted as one observation. Taking a business perspective, Government-NGO collaboration is excluded from this study. For robust analysis, I applied both samples for statistical testing of the developed hypotheses and obtained similar results. I report results using the sample of 225 observations, as building on Lin’s (2014) GBP study, it enhances result robustness by including KLD environmental concern data.
Statistical Model
In testing Hypothesis 5, “whether firms choosing GBPs are more likely to be associated with a radical rather than an incremental or a reactive eco-innovation goal,” there is a possibility of a selection bias, as factors that influence the likelihood of a GBP choice may, in turn, be associated with an alliance’s eco-innovation goal choice. I assume the existence of similar selection bias when I predict Hypotheses 1 to 4 on the operational elements of GBPs. One approach to eliminate sample selection bias is through the adoption of the matching theory, which is developed mainly in the medical and biological research fields and has been widely used in economics and finance (Shen & Chang, 2009). Ideally, this theory can simply compare two outcomes for the same unit to examine the average treatment effect (ATE) when the unit is assigned to the treatment and when it is not (Imbens & Wooldridge, 2009). The ATE on the outcome variable in the population of interest can be expressed as ATE = E[Y1 − Y0], where Y1 is the outcome variable with treatment and Y0 is the outcome variable without treatment.
However, a practical problem that arises given my cross-sectional dataset is that I can only observe either Y1 or Y0, because the assignment to the treatment is mutually exclusive. An alternative method to estimate the ATE that has gained a growing recognition in the social science research is to match observations in both the treatment (e.g., GBP participants) and control groups (e.g., non-GBP participants) with similar observable characteristics to enhance comparison validity (Shen & Chang, 2009; Uematsu & Mishra, 2012).
In coming up with these two matching samples, Rosenbaum and Rubin (1983) proposed the propensity score matching (PSM) technique, for which one can use predicted probability of being in the treatment estimated in either logit or probit model (Uematsu & Mishra, 2012) using exogenous characteristic variables as the explanatory variables. Then for each firm in the treatment sample (e.g., GBP participants), firms in the control samples are selected as matched samples according to the closeness of the above-estimated probability, thus coming up with a control group of firms (e.g., non-GBP participants). An important feature of the propensity score model is that it summarizes information contained in the multi-dimensional vector (e.g., varieties of observable characteristics) into a single-index variable (Becker & Ichino, 2002).
I thus employed PSM to create control group for firms with similar attributes and predicted whether firms choosing GBPs are likely to be associated with effective governance (H1), exploration learning (H2), cognitive learning (H3), rulemaking (H4), which enable them to pursue more radical eco-innovation goals (H5), compared with control groups.
Measurements
Eco-innovation goals
Following Lin (2012), I measured the radical scale of eco-innovation by examining the “deal synopsis” element in SDC database, and assigned three intended eco-innovation scales: reactive eco-innovation—pollution control (Scale 1), incremental eco-innovation—pollution prevention (Scale 2), and radical eco-innovation—clean technology (Scale 3). Refer to Table 1 for the detailed measurement schemes and statistical distributions of eco-innovation.
The Measurements of Key Variables and Their Descriptive Distribution.
Note. GBP = government–business partnership; R&D = research and development.
GBP participation
GBP was coded “1” if an alliance involved participants from the public government and business sector, otherwise 0. I used Thomson SDC’s data elements, such as “deal text,” “participant’s public status,” and “participant’s business description,” to determine whether the alliance involves government and business partners.
Alliance governance
“Ownership” was measured using a joint venture dummy, checking whether the partners signed a joint venture agreement. A “contractual agreement number” index was created to measure the total number of alliance agreements firms signed in the partnership. “Financial investment” dummy was created to examine whether the partners invested in the environmental partnership. “Reciprocity” was measured using “future ties”—the number of future partnerships with the present partner after the current partnership. According to Heide and Anne (1992, p. 267), the success of reciprocity strategies usually depends on “future returns” (prospects of future interactions), as partners are less likely to act opportunistically in the present when they anticipate possible reciprocal future responses.
Exploration learning
To measure “exploration learning,” I tracked whether firms signed an R&D agreement (dummy variable), as such agreement involves basic research, invention, risk taking, organizational learning, and entering new markets (Koza & Lewin, 1998). To examine the content of the R&D agreement, I create “technology” and “product” dummies by specifically coding whether the partnership focused on technology and product in the partnership. The interaction term of “R&D agreement” and “technology,” as well as “R&D agreement” and “product,” creates two inter-term dummies “RD tech” and “RD product” representing whether firms focused on generating alternative technology and product in their R&D agreements.
Cognitive learning
I measured cognitive learning with a dummy coding measuring whether an alliance’s primary activity focused on awareness raising. Related keywords include education, educate, information, campaign, knowledge, and knowledge sharing.
Rulemaking
I create “rulemaking” dummy to examine whether the alliance focused on developing and/or implementing a new policy or modifying existing policy. Related keywords included “norm(s),” “standard(s),” “regulation,” “mandate,” and “code(s) of conduct,” such as the “forest stewardship council” and “energy star.”
To address potential bias related to personal judgment in the content coding, a colleague coded the above variables independently. The result of this separate coding rate specifically matched 90.5% in eco-innovation, 92.0% in cognitive learning, and 93.5% in rulemaking, indicating a sufficient inter-reliability rate (Cohen, 1960). When coming across variances, raters compared notes and discussed to consolidate the final coding.
Firm-Level Exogenous Attributes
I used PSM technique to test Hypotheses 1 to 5, estimating the treatment effects of GBP on radical eco-innovation goals and predicting GBPs’ four structural elements. To eliminate the potential sample selection effect, it is important to carefully choose the observable characteristics (exogenous characteristic variables) and use probit (or logit) estimation to come up with the matching index. These parameters in the probit model are selected based on empirical findings in the literature.
Specifically, Lin (2014) argued the antecedents of GBP formation from two aspects, that firms with vulnerable strategic positions (e.g., substantial environmental concerns) and strong resource positions (e.g., large firms) are likely to partner with governments. I followed Lin (2014) to select KLD environmental concern ratings as proxy of firms’ vulnerable strategic positions, used firm size and market share as proxy of firms’ social status and influence in the industry, and used firms’ prior profitability (return on equity [ROE]) and risk-taking propensity as proxy of firms’ resource and strategic positing. Organizational risk is important to strategic management, as income variation can have negative consequences (Palmer & Wiseman, 1999) for firm’s strategic decision to select alliance partners in pursuing more radical eco-innovation.
I also followed Lin (2012) to include firms’ environmental alliance experience as a proxy of firms’ capacity to manage complex alliances that involved CSPs. These key indicators of a firm’s exogenous resource attributes are critical in influencing firms’ decisions to select GBPs. In the probit model, I also included alliance formation time and firms’ industry variances in terms of their competitive industry environment and industry sectors.
Firms’ environmental concerns
I measured firms’ environmental concern based on the “environmental” dimension of the KLD index, which include “hazardous waste,” “regulatory problems,” “ozone depleting chemicals,” “substantial emissions,” “agricultural chemicals,” and “other concern.” These six dummies measured specific environmental concern as “0” (no concern) and “1” (concern). Refer to Table 1 for the detailed measurement schemes and statistical distributions of environmental concerns.
Firms’ other characteristics
I also extracted other firm-level characteristics from the Compustat database: (a) Size was measured as the logarithm of firms’ asset prior to firms’ participation in the current alliances. Market share can be expressed as a company’s sales revenue (from that market) divided by the total sales revenue available in that market. (b) Prior profitability was measured as return on assets (ROA). (c) Organizational risk is measured using variance in ROA (firms’ income stream uncertainties; Palmer & Wiseman, 1999). (d) I also included firms’ environmental alliance experience, which is measured by counting the total number of environmental alliances firm participated prior to the current one.
In analyzing a firm’s industrial competitive environment, I used the total number of competitors in an industry as a proxy of the complexity of industry environment. I also controlled for a firm’s industry sector variance using four dummies: mining and construction standard industrial classification (SIC) 10 ~ 27; heavy manufacturing SIC 28 ~ 39; transportation and public utilities SIC 40 ~ 49. For each firm, I collected the above firm-level data 1 year prior to a firm’s participation in the focal environmental alliance. For alliance-level attributes, “formation time” was measured using five ordinal scales representing five time periods between 1985 and 2013.
Empirical Results
Table 1 summarizes the key constructs and their operationalization. Table 2 presents the descriptive statistics and a correlation matrix. As shown in Table 1, I noted that there are 32 GBPs among the 225 observations (14.22%). Among the 225 observations, 113 (50.22%) pursue radical eco-innovation in alliances, 80 (35.56%) pursue incremental eco-innovation, and 32 pursue reactive eco-innovation. This sample thus offers a good setting to compare GBPs and other alliances in the pursuit of radical eco-innovation. In predicting the operational design of GBPs, I noted that the sample also provides sufficient observations in equity ownership governance (26.67%), reciprocity (8.44%), exploration learning (24%), cognitive learning (12%), and rulemaking (8.44%).
Correlation Table.
Note. GBP = government–business partnership.
p ≤ .05.
In terms of exogenous firm attributes, on average firms have higher environmental concern ratings in hazardous waste (37.78%), regulatory problems (39.56%), and substantial emissions (27.11%). Most firms are from the heavy manufacturing sector (30.67%), followed by light manufacturing sector (21.78%) and transportation, communications, and utility sector (18.67%). As shown in Table 2, firms in the sample are relatively large in size, with an average of CAD$44.47 million in asset. Their average ROE is 12.7%. On average, these firms have 57 competitors in their respective competitive markets, and their average market share is 26.5%. On average, these firms have 1.44 prior environmental alliance experiences. Regarding the alliance formation time, there are two peak periods of environmental alliance formation; the first period is 1990-1995, 43.56% of observations are formed during this period. The second peak period is between 2006 and 2009, during which 32.89% of the alliances are formed.
I ensured that collinearity was not a problem in the probit model prior to running the PSM technique. I checked the correlational matrix and calculated the variance inflation factor (VIF) through running ordinary-least-squares (OLS) regression to specifically predict the likelihood for firms to be selected in the treatment sample (GBP participants). The largest VIF value in these models was 3.09. Thus, following the indicator of collinearity when its value is greater than 10.0 (Meyers, 2006), I concluded that collinearity was not an issue in my model. Note that in my study sample, not all observations are independent, as some alliances may involve multiple firms in the sample representing multiple observations participation in the same alliances. Following Sampson (2007), I corrected for this lack of independence between some observations using a cluster command when running probit (logit) regression models to adjust standard errors of the regression models. This correction of standard errors ensures that inclusion of multiple observations per alliance is not driving significant findings.
Table 3 reports parameter estimates for the probit model. Consistent with prior studies (Lin, 2012, 2014), results indicate that GBP participants (treatment sample) tend to have regulatory problems (β = 1.03, p < .01), have concern in ozone-depleting chemicals (β = 2.00, p < .01), and other environmental concerns (β = 0.61, p < .10). These firms tend to be larger in size (β = 0.38, p < .05), have more environmental alliance experiences (β = 0.11, p < .05), are from the heavy manufacturing sector (β = 0.77, p < .05), and operate in a competitive industry environment (β = 0.36, p < .01).
Probit Model Parameter Estimates.
Note. GBP = government–business partnership.
I also checked the quality of balancing in creating the treatment and control samples by using covariate imbalance testing (pstest) Stata command, using t tests for equality of means in the treated and non-treated groups before and after matching. Table 4 illustrates the descriptive characteristics of treatment and control groups before and after matching. To be considered as good balancing, these t values should be insignificant after matching (Largoza, Favorada, Reinante, Tan, & Thai, 2015). As shown in Table 4, none of the chosen covariates (exogenous firm characteristics) demonstrate significant t value. The estimated probit model satisfied the balancing property using the algorithm detailed in Becker and Ichino (2002). Thus, I concluded that I have obtained the balance matching sample for testing the treatment effect of GBP participation.
Pstest Checking the Balancing in Creating Treatment and Control Samples.
As the sample and control groups have similar characteristics, the resulting difference between two matched observations is theoretically the ATE—whether GBP participants are fundamentally different from non-GBP participants in the operationalization of effective governance (H1), exploration learning (H2), cognitive learning (H3), rulemaking (H4), and in the pursuit of more radical eco-innovation goals (H5). In Stata software, the “teffects psmatch” command (for PSM method) completes all these procedures related to ATE testing simultaneously. This command allows sensitive analysis by requesting the number of matches in the control groups in comparing with treatment group. There is a trade-off in this request. For instance,
When m = 1, each treated observation is matched with an observation in the control group with the closest distance, however, any unmatched observations in the treatment are not utilized in estimating the ATE. When m is larger, on the other hand, more observations can be utilized, but the quality of match may be compromised. (Uematsu & Mishra, 2012, p. 58)
For the robust test, I requested one to four matches for each hypothesis testing, and most results are significant regardless of the number of matches requested. I reported results using one or two requested matches to ensure match quality. Table 5 reports the testing results of these PSM models.
Average Treatment Effect for GBP Operational Elements and Innovation Goal.
Note. GBP = government–business partnership; ATE = average treatment effect; R&D = research and development.
The PSM models in Table 5 specifically demonstrate testing results of Hypotheses 1 to 5. All PSM models are significantly supported at .01, or .05, or .1 levels. As predicted, results in Model 2 support Hypothesis 2 that GBP participants are significantly more likely to conduct exploration learning, predicted by firms’ likelihood to sign R&D agreement (β = 0.44, p = .08), and their likelihood to create new technology (β = 0.52, p = .04) and new product (β = 0.46, p = .05) through these R&D agreements than non-GBP participants. Moreover, as shown in Model 3, GBP participants are significantly more likely to conduct cognitive learning (β = 0.13, p = .002) than non-GBP participants (Hypothesis 3 is thus supported). Similarly, my results in Model 4 support Hypothesis 4, which posit that firms selecting GBPs are more likely to be involved in rulemaking (β = 0.20, p = .005) than non-GBP participants. Furthermore, results in Model 5 show that the group of firms selecting GBPs is more likely to pursue radical eco-innovation goals than the non-GBP participants (β = 0.48, p = .05). Hypothesis 5 is also supported.
Related to effective governance (Hypothesis 1) in Model 1, as expected, GBP participants are significantly less likely to use ownership governance, predicted by firms’ likelihood to establish a non-equity engagement (β = −0.28, p = .000) than non-GBP participants, and they are less likely to sign significant number of formal contractual agreements (β = −0.16, p = .003) than non-GBP participants. These results indicate GBP participants’ likelihood to employ self-control governance mechanism.
An unexpected governance result I obtained relates to financial investment that GBP participants are less likely to financially invest in the partnership activities relative to non-GBP participants (β = −0.04, p = .01). In contrast, using “future ties” as a proxy, the association between reciprocity and GBP is significantly and positively supported. Results indicate that GBP participants have a stronger likelihood to continue to collaborate with the current partner after the present alliance (β = 0.59, p = .05) than non-GBP participants.
Discussion
This study applied both transactional cost and complementary logics to explain the rationale of GBP formation for radical eco-innovation. Integrating instrumental and normative perspective, this study explains how GBP is operationalized from governance, explorative learning, cognitive learning, and rulemaking aspects that facilitate the pursuit of more radical eco-innovation goals. This study empirically tested these theoretical associations using 225 observations representing 166 U.S. firms’ participation in 192 environmental alliances between 1985 and 2013. It used PSM technique to match GBP participants with non-GBP participants to assess the treatment effect of GBP participation on firms’ likelihood to choose their alliance operational elements and pursue more radical eco-innovation goals. As predicted, the study results suggest that GBPs play a significant and positive role in channeling public and private efforts in pursuing transformative environmental change via the adoption of radical eco-innovation goals. Results further highlight four critical elements of GBP operation that enable participants to embrace these radical environmental solutions. These insights make the following contributions to the academic literature:
First and foremost, environmental strategy scholars (Aragon-Correa & Sharma, 2003; Hart, 1995) delineated a “roadmap” for sustainability and call for firms to move beyond their reactive environmental stances and undertake more radical environmental innovations. However, firms do not want to embrace these radical eco-innovations alone due to resource constraints, and that the targeted innovations generate mixed goods—public goods with unpredictable private benefits. Neither the state nor the firms make satisfactory investments in the mixed goods, thus leading to the “underinvestment” dilemma, especially evident in the energy sector (Kolk, van Tudler, & Kostwinder, 2008, p. 263). Complex environmental challenges call for collective actions from government, business, and civil society to transform common practices, technologies, rules, and beliefs deeply entrenched in the institutional environment (Hart, 2005). This study enriches business sustainability, CSP, and PPP literature by highlighting GBP’s potential to combine cross-sector effort in promoting radical eco-innovation that brings decisive social and environmental change.
These insights confirm the government’s potential in designing and directing GBPs as an alternative policy scheme to channel government and business efforts toward the promotion of radical eco-innovation. They resonate with the propositions of Majumdar and Marcus (2001) and Starik and Heuer (2002), who argue that environmental policies should move away from prescribing technological solutions and allow for more flexible strategies that promote innovation in products, processes, and technologies. They suggest that governments may consider going beyond their traditional regulatory role, and explore alternative roles as collaborators and enablers that bear greater environment change potential.
Second, this study further enriches business sustainability, CSP and PPP scholarship (Bryson et al., 2006; Stadtler, 2015) by explaining how GBPs can be operated to facilitate participants’ pursuit of a radical eco-innovation path. Consistent with the dual structure of GBP, its operational elements contain a blend of instrumental and normative aims. Specifically, the study results seem to highlight the normative operational elements that explain how GBPs facilitate participants to embrace more transformational rather than incremental or reactive environmental innovation. These identified operational elements make the following contribution to alliance governance and learning literature:
The study contributes to alliance governance literature by illustrating “What is an effective governance pertaining to radical eco-innovation.” Examining governance from ownership, investment, and reciprocity dimensions, my results suggest that GBP tends to lend itself to a reciprocity-based, non-equity governance mechanism compared with non-GBP participants. Contrary to my prediction, these participants are less likely to be governed by financial investment as an economic hostage for participants. This unexpected result could be due to data availability. As SDC database does not have funding agreement data, I contently coded “deal text” of every environmental alliance to identify whether partners financially contribute to the partnership. However, in some scenarios such financial data were deemed confidential and not manifested in the deal text. And partners’ other commitment such as technical expertise or patent is not included in the “financial investment” dummy. The data availability limits my result interpretation.
The second explanation of this counterintuitive result relates to the limitation of instrumental means. Hartwich and Tola (2007) highlighted that GBP may be originated from financial reasons, such as “competitive grant schemes that provide funding conditional on a certain level of collaboration and co-financing”; however, partnerships that originate in these contexts “do not always make the best of their potential because they are biased toward the interests of one partner or they originate solely from the search for funding without regard to partner interests” (p. 12). Therefore, while financial arrangement is an important element for GBP formation for infrastructure projects, it seems that GBPs formed for radical eco-innovation goal tend to go beyond instrumental consideration. The main rationale for such GBP is “to bring together a pool of innovative talents, with complementary skills to foster a mutual learning and the development of creative ideas” (p. 12). For these kinds of partnerships, reciprocal commitment is a more important governance mechanism that helps develop mutual understanding and commitment, trust, and long-term vision toward more radical eco-innovation goals. My results confirm this “reciprocal commitment” aspect of GBP. These insights enrich alliance governance literature pertaining to radical eco-innovation.
Furthermore, this study contributes to alliance learning (March, 1991) and technological path constitution literature (Garud & Karnoe, 2001) by suggesting GBPs’ potential in convening government, business, and civil society (university) in exploration learning, cognitive learning, and rulemaking, with an aim to initiate and/or shape a chosen innovation path. My results support Robin and Schubert (2013), indicating that although knowledge transfer and learning can occur through a variety of channels (see Schartinger, Rammer, Fischer, & Frohlich, 2002, for an overview, Robin & Schubert, 2013), the interaction between the government (especially public research labs) and the business sector remains one of the most important institutional interfaces for knowledge diffusion and organizational learning (Robin & Schubert, 2013).
In this regard, this study advances sustainability and alliance learning literature by delineating two types of learning (explorative and cognitive) that help participants build transformational change capacity. Specifically, cognitive learning moves beyond the instrumental “effectiveness thinking” to encourage a triple-loop learning (Argyris & Schön, 1978, for an overview, see Tosey, 2011). Based on Tosey’s (2011) review, such a cognitive learning is driven by normative considerations (Roper & Pettit, 2002). Given the unsustainability of the present system (which therefore requires radical and systemic change), cognitive learning enables individuals in the organizations to understand the urgency of changes needed. It questions “the role or the mission of the organization” (Lassey, 1998, p. 11), and “provides feedback and a change mechanism for the individual” (Argyris, 1991; Dishman & Pearson, 2003, p. 616; Tosey, 2011). As such, such a cognitive learning encourages a reflective process regarding the “issues of ethics,” underlying purposes, principles, or “paradigms” (Argyris & Schön, 1978; Isaacs, 1993, p. 30). Such a cognitive learning thus has a potential to lead to “change in the embedded tradition system within which the governing values of a behaviour can be nested” (Nielsen, 1993, p. 118). From this end, this study advances organizational learning, especially the triple-loop learning literature, by elaborating and empirically testing the purpose of learning as there is lack of empirical study examining these learning aspects in literature (Tosey, 2011). This learning aspect explains how organizations develop greater incentive to go for radical innovation.
In summary, the four hypothesized operational elements of GBP (namely effective governance, exploration learning, cognitive learning, and rulemaking), advance business sustainability, CSP, and PPP literature by explaining “how” GBPs facilitate radical eco-innovation. These four elements are complementary and interlinked. For instance, effective governance ensures an engaged and flexible support for the operation of exploration learning, cognitive learning, and rulemaking. Reciprocity governance, with the prospects of future ties, helps facilitate cognitive learning by building long-term commitment and shared vision among government, business, and civil society partners in the pursuit of radical eco-innovation. According to Hart (1995), such a dedication to a compelling long-range vision “was the key to generating the internal pressure and enthusiasm needed for innovation and change” (p. 1002).
Similarly, there is a strong complementarity between explorative and cognitive learning as they involve transformational change in both technical and mind-set aspects that constitute a holistic and more effective approach to develop new, feasible innovation that addresses tenacious social problems (Robin & Schubert, 2013). Furthermore, exploration learning and rulemaking are complementary in that, while exploration learning stimulates the emergence of radical eco-innovation, the spread and enforcement of such an innovation path rely on rulemaking and policy development. Likewise, cognitive learning and rulemaking are also complementary and reinforcing. Working together, they have a potential to drive paradigm shifts that lead to system change in addressing environmental challenges. Therefore, examining these operational elements in a GBP setting explains how GBPs facilitate partners to move beyond incremental modifications in existing practice to introducing systems change.
Conclusion, Policy Implication, and Future Study
GBPs work at the intersection of markets and regulations to bring about social and environmental change that none of the partners could achieve alone. The unique GBP platform has the potential to greatly reduce uncertainty and mobilize critical resource sharing through effective governance. They encourage innovative and knowledge-creation activities through exploration learning, as various government, business, and civil society partners are better able to share resources and risks associated with eco-innovation. GBPs also engage partners for cognitive learning and environmental rulemaking in a way that may facilitate the scaling up and expedition of the environmental change process. Therefore, within the promotion of GBPs, there are potential opportunities for government, business, and civil society partners to collectively respond to today’s environmental challenges, and jointly promote a radical eco-innovation path with far-reaching impact to society and the natural environment.
This study has important managerial implications. As firms’ unilateral efforts to adopt more radical eco-innovation generally encounter risk and resource constraints, they may embrace GBPs to hedge the business sector’s investment risks and gain access to critical government resources related to the exploration of radical environmental solutions. This study also has important policy implications, which entails an increase in governments’ deliberation regarding complementing environmental command-and-control policies with voluntary strategies, such as GBPs, which incorporate greater stakeholder involvement and environmental improvements (Starik & Heuer, 2002). GBP’s involvement in eco-innovation through exploration learning, cognitive learning, and rulemaking may act as alternative policy mechanisms that help transform extant practices, rules, and beliefs that are deeply entrenched in the extant institutional environment.
Nevertheless, for governments to play a stronger role in this regard there is a necessity to revisit extant policy schemes and ensure a consistent support of radical eco-innovation. A radical eco-innovation path commitment requires sustained access to critical resources, such as policy support and incentives, as well as the associated market intelligence, technological expertise, selling and distribution systems, and capital funding. However, if governments’ resource power instead supports routine business practices, this can counter radical innovation. Moreover, the promotion of GBPs for radical eco-innovation may not be straightforward. There may exist potential conflict and trade-off between the government and business partners regarding the alignment of public and private benefits, and that firms’ economic interests linked to GBPs may be materialized in a staggered way (Stadtler, 2015). In this regard, both sectors still face a considerable learning curve regarding the design of effective GBPs for more radical eco-innovations. It is therefore important for future studies to explore the potential conflicts and reconciliation among the government and business partners in their pursuit of radical environmental improvements.
The study outlines four operational elements of GBPs. Future study can further explore and test the complementarity and interaction of these elements that may help enforce the GBPs’ effectiveness in the pursuit of the radical eco-innovation goals. It would be also interesting for future study to explore a series of research questions related to business partner selection for the pursuit of radical innovation through GBPs. For instance, what factors or incentives are desirable in facilitating a legacy company to actively seek GBPs for radical innovation that would risk making its former products/services (and profits) obsolete? Should the promotion of GBPs focus on the transformation of incumbents or on supporting new innovative entries? Future studies can also extend the research to examine other inter-firm alliances and CSPs (such as firm–NGOs partnerships) in the pursuit of radical innovation goals. They can apply transaction cost, complementary logic, or a learning perspective to compare the effectiveness of different environmental partnerships in engaging business to tackle different environmental issues, contingent on whether the collaboration target public, mixed, or private goods. Furthermore, my analysis is based on a U.S. sample’s environmental efforts, which limits the generalizability of the findings to other contexts. In the future, scholars should explore this study’s research questions in other geographical settings and crosscheck my research findings.
Finally, the financial crisis and its aftermath have revealed the limits of economic value creation, which has turned academics’, public funders’, and regulators’ attentions to CSPs, such as GBPs, to encourage the government and business sectors to cross sectoral boundaries and achieve collective prosperity, rather than merely yielding to individual benefits. Future studies are encouraged to continue this research theme and assess GBPs in a broader social and environmental setting.
Footnotes
Acknowledgements
I thank Dr. Jennifer Oetzel, associate editor of Business & Society, and the two anonymous reviewers for their insightful comments and feedback that help improve the quality of the article. I also thank Azin Ranstan and Xinya Yan for their research assistances.
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: I also gratefully acknowledge the financial support from Social Science and Humanities Research Council (SSHRC) Insight Grant (SSHRC # 435-2012-1557) titled “Cross-sector solutions to complex environmental issues”.