Strengthening advanced manufacturing innovation ecosystems: The case of Massachusetts

Technological Forecasting & Social Change xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Technological Forecasting & Social Change journal homepage: www.elsevier.com/locate/techfore

Strengthening advanced manufacturing innovation ecosystems: The case of Massachusetts Elisabeth B. Reynoldsa, Yilmaz Uygunb,c,⁎ a b c

MIT Industrial Performance Center, United States MIT Industrial Performance Center, United States Jacobs University Bremen, Germany

A R T I C L E I N F O

A B S T R A C T

Keywords: Innovation Regional innovation ecosystems Advanced manufacturing Massachusetts

Recent years have brought a renewed focus on the importance of manufacturing to the health and future growth of nations and regions. Several studies have highlighted the need to maintain and build manufacturing capabilities to support economic growth and have linked a nation's as well as region's strength in manufacturing to its ability to innovate. In the U.S., where a manufacturing strategy has largely been absent for the past 25 years, advanced manufacturing capabilities are now seen as essential to the development of new products and processes across a range of industries. Against this backdrop, Massachusetts presents an interesting case since manufacturing in this U.S. state is integral to several of its most important industry clusters, yet it is a high wage, high costs state that must compete globally. This research examines the pathways and opportunities for building and fostering innovation capacity among Massachusetts manufacturers, with a particular focus on small and medium-sized enterprises (SMEs). We employ a systems approach to conduct analytic and empirical analyses that consider how knowledge and sources of innovation flow between key participants within the manufacturing innovation ecosystem. We find that the Massachusetts manufacturing innovation ecosystem is rich in terms of assets but relatively poor in terms of interconnectedness between those assets. In addition, rather than being focused on demand-driven innovation and technological upgrading for SMEs, non-market state-supported manufacturing intermediaries are primarily focused on supply-side, point solutions that work with individual firms rather than at a systems level.

1. Introduction Recent years have brought a renewed focus on the importance of manufacturing to the health and future growth of the U.S. economy. Indeed, several studies and public-private initiatives have highlighted the need to maintain and build manufacturing capabilities to support economic growth, good jobs, and national security. Perhaps most importantly, they have linked the nation's manufacturing capabilities to its ability to innovate. Advanced manufacturing is essential for developing new products and processes across a range of industries, both established and emerging. As others have pointed out, the loss of these capabilities can shift an industry's center of gravity as higher valueadded activities follow manufacturing abroad (Pisano and Shih, 2011). In few states is the link between manufacturing and innovation more evident than in Massachusetts. The state, home to MIT and Harvard, is consistently ranked number one in terms of innovation capacity in the U.S. (Bloomberg Innovation Index, 2016). While manufacturing represents only 9% of employment in the Commonwealth

⁎

(approximately 250,000 jobs), compared to 11% in the U.S. overall, it is integral to several of the state's most important industries, including aerospace/defense, semiconductors and computers, biopharmaceuticals, and medical devices. Massachusetts manufacturers compete globally on their innovation capacity, high skills, product quality, and rapid response. Small and medium-sized enterprises (SMEs) play a critical role in maintaining and growing the manufacturing strengths of the U.S. and Massachusetts economies, and other advanced industrialized economies such as Germany. These companies are the “backbone” of the country's and the region's industrial capabilities and they exist in every community where manufacturing exists often supplying complex as well as commodity parts and components. SMEs supply both the large established firms (known as “original equipment manufacturers” or OEMs that regularly develop sophisticated products and systems as well as the entrepreneurial firms that engage in prototyping or pilot production to advance new products. The former are “firms that […] manufacture […] based on ‘original’ designs” (Sturgeon, 2001). OEMs either make

Corresponding author at: MIT Industrial Performance Center, United States. E-mail address: [email protected] (Y. Uygun).

http://dx.doi.org/10.1016/j.techfore.2017.06.003 Received 29 December 2015; Received in revised form 28 April 2017; Accepted 5 June 2017 0040-1625/ © 2017 Elsevier Inc. All rights reserved.

Please cite this article as: Reynolds, E.B., Technological Forecasting & Social Change (2017), http://dx.doi.org/10.1016/j.techfore.2017.06.003

Technological Forecasting & Social Change xxx (xxxx) xxx–xxx

E.B. Reynolds, Y. Uygun

is frequently used but often poorly specified. Innovation differs from invention in that the latter is the creation of something new and novel while innovation is the process of adding value to an invention such that it becomes useful in the marketplace (Schumpeter, 1934). There are generally considered four different dimensions to innovation; product, service, process, and organizational (Kirner et al., 2009). Product or service innovation is the first-time commercial utilization of a product or service that is new to the market, whereas process innovation is the implementation of methods that are new to the company, but not necessarily new in the market, and that change the way a company manufactures a product. Process improvement measures, like lean manufacturing, Six Sigma, etc., are often included in this category of innovation, though they may be less about true innovation and more about continuous improvement. Organizational innovation is the implementation of new organizational methods within a firm that change the firm's business practices, communication, and/or workplace organization (Uygun and Reynolds, 2016). The primary focus of this paper is on product and process innovation. Innovations are often only realizable if embedded in a fruitful “ecosystem’. The term “innovation ecosystem” has gained popularity in recent years. The “ecosystem” metaphor draws from our understanding of natural and biological ecosystems. An ecosystem comprises all living organisms within a physical environment functioning together as a unit and seeking an equilibrium state with a stable set of conditions to keep a population at desirable levels. Equilibrium is sought through modeling the energy dynamics of the ecosystem operations where energy is a means by which living organisms' energy is transferred to the soil by dying and decomposing which then can be taken up by other organisms. Thus, ecosystems are a “complex set of relationships among the living resources, habitats, and residents of an area whose functional goal is to maintain an equilibrium sustaining state”. Based on this idea, innovation ecosystems refer to the economic relationships between actors (university faculty and students, entrepreneurs, industry leaders, government officials) and entities (market and non-market organizations) whose functional goal is to enable innovation. Innovation ecosystems can be seen as “inter-organizational, political, economic, environmental, and technological systems through which a milieu conducive to business growth is catalyzed, sustained, and supported. A dynamic innovation ecosystem is characterized by a continual realignment of synergistic relationships of people, knowledge, and resources that promote harmonious growth of the system in agile responsiveness to changing internal and external forces” (Jackson, 2011). The concept of an innovation ecosystem is rooted in the literature on systems of innovation that emerged approximately 25 years ago, building upon endogenous growth theory that emerged in the 1980s (Romer, 1986). New growth theory put knowledge creation at the center of economic growth models, though their antecedents can be traced back to Marshallian industrial districts described in the late 19th century. Systems of innovation thus describe the characteristics of environments that support knowledge creation and enhance greater innovation. The systems of innovation literature began at the national level and then were later applied to the regional level. A national innovation system (NIS) is defined most succinctly as “the set of institutions [= organizations] whose interactions determine the innovative performance of national firms” (Nelson and Rosenberg, 1993). The NIS approach assumes homogeneity in a country with respect to national institutions such as legal and regulatory frameworks. There can be, however, significant regional differences within countries that impact the level of innovative activities (Asheim and Gertler, 2005). Regional innovation systems (RIS) stress the regional dimension of production and the exploitation of new knowledge to help explain regional differences in innovation capacity and economic strength where the focus is on the relationship between technology, innovation, and industrial location (D'Allura et al., 2012). While the term innovation ecosystem can be used to refer to

products directly or act as a system integrator before selling directly to the customer. Throughout this paper, the term OEMs typically refers to large enterprises (with over 500 employees). Today, because of the fierce global competition in manufacturing capabilities, most manufacturing that occurs in the U.S. fall into the category of “advanced manufacturing.” In a broad sense the term refers to the use of next-generation technologies in manufacturing processes. More precisely, advanced manufacturing encompasses “a family of activities that depend on the use and coordination of information, automation, computation, software, sensing, and networking, and/or make use of cutting-edge materials and emerging capabilities enabled by the physical and biological sciences” (PCAST — President's Council of Advisors on Science and Technology, 2011). Advanced manufacturing can refer to improving current manufacturing practices of existing products as well as the manufacturing of new products using new advanced technologies. As OEMs progressively look outside their operations toward their supply chains to improve their innovation capacity, key questions arise as to how to build this capacity among the SMEs that operate within global supply chains and support regional manufacturing capabilities. This paper examines the question of how a highly innovative, high wage region like Massachusetts can improve its innovation ecosystem to support SMEs in their efforts to be globally competitive. Manufacturing capabilities are grounded in particular regions, where, historically, they have grown around key industries. Thus, manufacturing lends itself to regional approaches for increasing innovation capacity and upgrading firms' capabilities. Strengthening the regional innovation ecosystem as a whole will improve the “industrial commons” (Pisano and Shih, 2011) and leverage resources by helping all manufacturers in the state, not just a select few. This is particularly important for SMEs. Recent research by MIT's Production in the Innovation Economy (PIE) project (Berger, 2013) concluded that SMEs often find themselves “home alone” when it comes to competing globally and driving innovation in their companies. The large, vertically-integrated corporations of the 1980s have become less vertically integrated over time as they have focused on their core competencies, outsourced much of their production and increasingly relied on smaller suppliers to drive innovation. This process has left “holes” in the industrial ecosystem in the U.S., cutting off many of the important investments and spillovers that used to flow from large corporations to smaller firms, for example, in training, technology adoption, and R & D investments. As a result, many SMEs have been left largely on their own to figure out how to find and train workers, adopt new technologies, and develop and scale new products and services, while shouldering the burden of funding this at the same time. It is within this global and national context that small and mediumsized manufacturers need to innovate in order to remain competitive and participate in technological advances in manufacturing. The ecosystem within which they operate is critical to their ability to do this. The case of Massachusetts presents a dynamic model of how this occurs or could occur given gaps that exist in the manufacturing innovation ecosystem as it is presently organized. This paper focuses on how one region is working to fill these holes as they relate to innovation. For this, an empirically grounded systems approach is used that considers how knowledge and sources of innovation flow between key actors within the manufacturing innovation ecosystem. Strengthening these links and expanding the flow of knowledge between key actors upgrades the system as a whole and enhances the region's competitiveness. As regions and countries around the world increase investment in manufacturing and incentives for manufacturing firms, it is increasingly important to understand what makes for a more effective regional innovation ecosystem. 2. Literature review This literature review begins with defining “innovation,” a term that 2

Technological Forecasting & Social Change xxx (xxxx) xxx–xxx

E.B. Reynolds, Y. Uygun

innovation systems within firms, districts or at a national level, the most common and perhaps most appropriate geography is the region (or for smaller countries, the country level) in which geographic proximity among the actors is a key attribute of the ecosystem. Thus, RIS literature is the most relevant to this paper. RIS research currently addresses the types and varieties of regional innovation systems, their impact on regional competitive advantage, and how to stimulate RIS growth, whether through the development of clusters and networks, or human capital development. Many have argued that regional performance is ultimately determined by national trends and that regional policy must be linked to a national and international context (Fromhold-Eisebith, 2007). While global integration of RIS is essential because external contracts are crucial to competitiveness and innovation capacity (Asheim and Isaksen, 2002), it is the organizational and institutional thickness at the regional level that often contributes to firm clustering in specific regions and can give rise to RIS. These organizations and institutions help foster the networks and learning process within regions, critical for knowledge transfer and learning as well as developing and attracting high skilled labor. Measuring the effectiveness or efficiency of RIS in contributing to regional competitive advantage is an important area of current research. Oftentimes, public funding is only allocated to economic growth efforts when there is the threat of market failure, but it should also be used as a long-term investment to stimulate growth. Braczyk et al. (1998) argue that policymakers should shift their focus from trying to rapidly monetize the effects of RIS to improving the flow of knowledge from within networks to industry in order to increase overall competitiveness. Critical in this process is the role of intermediaries (Inkinen and Suorsa, 2010) which aid in the technology transfer process, particularly when they can provide funding to support the process. The availability of such intermediaries has been a crucial competitive advantage of the United States over European states (Cooke, 2001), as the latter rely excessively on public funding which is only made available when market failure is already imminent. In the end, the RIS literature presents largely a descriptive analysis of the organizations, institutions, and the links between them; essentially a snapshot of the existing innovation system. However, the picture is relatively static and does not discuss the quality of the links between key nodes (the “thickness” of the link) in the system or how the system could be further enhanced and developed. Our approach in this paper, using the term “innovation ecosystem” rather than RIS, emphasizes the dynamic nature of the system and addresses the issues of both coordination and enhancement of the advanced manufacturing innovation ecosystem as a whole.

manufacturing ecosystem to understand the roles they play in fostering innovation-related knowledge flows. These qualitative interviews provided insights into whether and how advanced manufacturing firms and non-market actors interact as part of a larger system, and the relative strengths and weaknesses of this system. Finally, 15 additional qualitative interviews were conducted in Germany with highly innovative companies and non-market organizations which provided a benchmarking of sorts of how manufacturing innovation ecosystems are organized in one of the best countries globally for manufacturing. 4. Analyzing the manufacturing innovation ecosystem The innovation process is often characterized as non-linear and dynamic, involving different actors with highly interactive relationships (Kline and Rosenberg, 1986; Van de Ven, 1986). While firm innovation might have occurred in isolation in the past, particularly when many firms were vertically integrated, today's firms have high degrees of interaction with a range of other companies and organizations (both market and non-market), such as universities, suppliers, customers, and even competitors, all of which may play a part in building a firm's innovation capacity. External factors or institutions such as laws, regulations, culture, and technical standards also play an important role in setting the stage for innovative activities (Edquist, 2005). For these reasons, the process of innovation cannot be viewed through one single lens (within a single company or institution) but needs to be understood as part of a larger system (Maskell, 2001; Utterback and Suarez, 1991). The following provides an overview of manufacturing in the state of Massachusetts and outlines the key entities within the larger ecosystem. 4.1. The current state of manufacturing in Massachusetts Massachusetts offers an important case study of how small U.S. manufacturers in high wage locations compete in today's global economy and complex supply chains. The Commonwealth has a diverse and sophisticated manufacturing base that includes about 7000 firms in a wide range of industries, including aerospace/defense, semiconductors/electronics, medical devices, and biopharmaceuticals (Bluestone et al., 2012). SMEs with fewer than 100 employees account for about 92% of the manufacturers in the state (US Census Bureau, 2012). The vast majority of these firms participate in global supply chains. However, SMEs account for only approximately 30% of the state's manufacturing employment (OneSource, 2015). Large firms, while representing only approximately 3% of all manufacturing establishments, account for 70% of manufacturing employment in the state. Massachusetts has a long and illustrious history in manufacturing and in product and process innovation beginning with arms manufacturing during the American Revolution and textile manufacturing subsequently (Hounshell, 1984). The advanced manufacturing capabilities it has built over the past 150 years have allowed companies and workers to transition into new or emerging industries as market conditions change. In fact, one of the region's strengths is a diverse manufacturing base that supports cross-fertilization among its key clusters. Manufacturing employment has steadily declined over the past several decades. Since the 1990s manufacturing jobs have declined as a share of the state's overall employment from approximately 19% to about 9% today (this compares with a national-level figure of about 11% in 2013, down from 20% in 1990; BLS — Bureau of Labor Statistics, 2015), where the current data reflect some recovery from the depths of the Great Recession in 2008 (see Fig. 1). The decline in the share of manufacturing jobs at the state level mirrors national trends for the U.S. as a whole, and global trends for other industrialized countries as productivity rates have increased, production has become more fragmented, and global competition has intensified. In terms of manufacturing establishments, Massachusetts

3. Research methodology To gain insights into the advanced manufacturing innovation ecosystem in the Commonwealth of Massachusetts a complementary mix of analytical and empirical methods has been used. We followed a deductive supply chain based approach by analytically identifying the largest companies in the most innovative sub-industries. To identify those advanced manufacturing industries and subindustries in the state, data was collected on firm patent, R & D spending, employment and wage data. Once these industries were identified, analysis was conducted on global industry dynamics as well as the largest firms within each industry in Massachusetts. Approximately 40 qualitative interviews were conducted with these large firms and with their most innovative suppliers (as defined by the large company) to understand how from their perspective innovation occurred within their firms and the sources and flow of knowledge that supported this innovation. In addition, qualitative interviews were conducted with the most important non-market organizations, like intermediaries, universities, research and technology organizations that work within Massachusetts' 3

Technological Forecasting & Social Change xxx (xxxx) xxx–xxx

E.B. Reynolds, Y. Uygun

Fig. 1. Total number of jobs in the manufacturing industry in Massachusetts and in the United States between 2001 and 2013.

Total Number of Manufacturing Jobs: MA vs U.S. 450,000

18,000,000

400,000

16,000,000

350,000

14,000,000

300,000

12,000,000

250,000

10,000,000

200,000

8,000,000

150,000

6,000,000

100,000

4,000,000

50,000

2,000,000

# of employees in the US

# of employees in MA

[BLS Data for NAICS 31-33]

0

Mfg Employees MA

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2001

2002

0

Mfg Employees US

Fig. 2. Total number of establishments in the manufacturing industry in Massachusetts and in the United States between 2001 and 2013.

Total Number of Manufacturing Establishments: MA vs U.S. 12,000

450,000 400,000

10,000

350,000

8,000

300,000 250,000

6,000

200,000

4,000

150,000 100,000

2,000

50,000

0

Mfg Establishments MA

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

0

# of mfg. establishment in the US

# of mfg. establishments in MA

[BLS Data for NAICS 31-33]

Mfg Establishments US

• Suppliers that are able to quickly provide difficult-to-manufacture parts of very high quality and reliability, • World-class universities, and • Innovative startups and a dynamic entrepreneurial ecosystem.

experienced a steady decline from about 10,000 establishments in 2001 to about 7000 in 2013, for a total contraction of about 30%, i.e. twice the national rate. The total number of U.S. manufacturing establishments fell from about 400,000 in 2001 to about 335,000 in 2013 (see Fig. 2). Despite the fact that a significant number of Massachusetts SMEs are engaged in contract manufacturing of what are often termed “commodity products”, OEMs interviewed consistently referenced the following attributes as key characteristics of the state's manufacturing production system:

For all these reasons, the Massachusetts manufacturing base has stabilized since the 2008 crisis and remains strong today. Indeed, the state's manufacturers are well positioned to take advantage of some of the national and global trends that suggest the U.S. may be more globally competitive in manufacturing in the future. In particular, declining energy costs, rising labor costs in traditionally low-wage countries, and concerns about the protection of intellectual property are making the U.S. environment more competitive for certain types of manufacturing, including those in which Massachusetts excels. In addition, the development of new “game-changing” advanced manufacturing technologies such as additive manufacturing, cyber-physical systems, and integrated circuit photonics, is providing additional opportunities for U.S. firms to innovate and increase efficiency. This topic has been the focus of two reports to the U.S. President's Council of Advisors on Science and Technology (PCAST — President's Council of Advisors on Science and Technology, 2011, 2014). In the reports the Advanced Manufacturing Partnership (AMP), an industry-academiagovernment partnership, put forward several recommendations for boosting innovation in advanced manufacturing in the U.S. through the creation of new R & D infrastructure and technology road maps.

• Small-batch niche production, rather than large-volume mass production, • Extremely high quality and performance requirements (0% failure), • High knowledge and innovation content, • New or early-stage products and prototyping, • Products with high proprietary content, • Products where proximity to market is desirable, • Products where regulatory factors encourage siting in the U.S., and • Customized products with quick turnaround time if needed. To sustain these characteristics, OEMs draw on four primary assets:

• A well-educated and highly skilled labor force, particularly in engineering,

4

Technological Forecasting & Social Change xxx (xxxx) xxx–xxx

E.B. Reynolds, Y. Uygun

Fig. 3. Procedure for defining focus industries and their relationship to key sectors in Massachusetts.

In terms of industry breakdown, we used three filters to help determine which manufacturing sub-industries could be considered especially advanced or innovative. Starting with NAICS (North American Industry Classification Systems) codes at the four-digit level, we considered patent data as a proxy for innovation (albeit one that is not particularly well suited for manufacturing), R & D spending per worker and share of STEM (science, technology, engineering, and math) occupations, and employment data. The identified sub-industries this way can be seen in Fig. 3. These seven industries also pay the highest annual average wages per employee reflecting the higher value-added and advanced nature of the jobs within these industries. Ultimately, we delved into these seven industries down to the sixdigit level NAICS codes to identify the nine advanced manufacturing sub-industries (out of 345) with the highest employment. Fig. 4 also shows the overall ranking of these sub-industries by employment. Perhaps counter-intuitively, we also included machine shops as a focus sub-industry, despite the fact that this type of enterprise was not identified in our filtering process. The companies in this category are

primarily process specialists with no proprietary products. They are overwhelmingly SMEs with fewer than 100 employees. Machine shops are a valuable part of the ecosystem and support all of the key manufacturing-related sectors of the economy. As a sub-industry they not only have one of the highest employment levels in the state, they are also important enablers of product innovation by OEMs, delivering high-precision, small-batch products with short lead times. Machine shops are also the sub-industry with the highest number of ISO-certified companies reflecting their commitment to high precision and quality. Overall, the 10 manufacturing sub-industries we identified are primarily concentrated in the greater Boston area, although machine shops are located throughout most of the state.

4.2. The key elements of the innovation ecosystem Based on our research, there are four key nodes comprised of market-oriented entities (Large OEMs, Supplier SMEs, Startups) and non-market entities (Universities/Research Institutes), that play a major role in the state's advanced manufacturing innovation ecosystem

Fig. 4. Sub-industries with the highest employment within the defined focus industries.

5

Technological Forecasting & Social Change xxx (xxxx) xxx–xxx

E.B. Reynolds, Y. Uygun

Fig. 5. A schematic of the manufacturing innovation ecosystem in Massachusetts.

university research drives “disruptive” innovation and is often focused 10 to 15 years out in terms of new technological developments, which is of little use to SMEs. Finally, a vibrant startup community can be an important source of innovation for OEMs. The strength of the link between startups and OEMs depends in part on the industry and to the extent to which OEMs are receptive to, and actively engaged with, the startup community. The following provides more in-depth analysis of the role of each of the nodes within the ecosystem as well as opportunities and challenges for SMEs.

(Fig. 5). The state plays an important role facilitating interactions between these nodes as well as supporting the system as a whole but does not constitute in and of itself a node in the system. Fig. 5 presents a stylized representation of the key actors in the ecosystem. Obviously, the innovation system relies not only on flows between the four nodes depicted in this figure but also on knowledge that comes into the region from outside sources such as R & D networks, trade associations, and global partnerships/networks. The lines connecting each of the four nodes represent the general strength and direction of the knowledge flows between them. Since we are mainly interested in learning how companies come up with innovations (that may lead to patents), where they get the knowledge/ ideas from, how innovation-related idea generation and collaboration with partner companies changed over time, we want to make visible the directions (arrows) of the individual contribution to the innovation capacity of firms. The thickness of those arrows is based on the answers we got. Here, we distinguish between thick, thin, and no arrows based on the number of interviewees' answers to the question, where do you get the ideas for new innovations or with whom you develop those innovations, respectively. The threshold value was 50%, i.e. if more than half of the interviewees pointed to one particular node, the corresponding arrow turns thick. Again, the state and other non-market intermediaries are not depicted as key nodes in the system because of the supporting role they play in this ecosystem. Their work focuses on assisting in various ways with each of the nodes (i.e., lean training for SMEs) or facilitating interactions between the nodes (i.e., prototype development between startups and SMEs; incentives for research collaboration among OEMs and universities). To summarize, OEMs have the strongest links within the innovation ecosystem because they are the primary drivers of innovation within it. Knowledge flows between OEMs and research universities are strong in both directions, while knowledge flows with SMEs are relatively unidirectional flowing from OEM to SME. With respect to innovation, startups typically bring new ideas to the OEMs. In contrast to OEMs, SMEs generally have the weakest links within the ecosystem. Historically, they have most often been on the receiving end of knowledge flows from their large customers. Their ability to drive knowledge and ideas in the other direction, toward the OEMs, has been limited, though this is highly dependent on the OEM. SMEs also generally have weak links to universities and to the startup community. Universities have relatively strong links with large OEMs and with the startup community, but limited engagement with SMEs. Much

4.3. OEMs within the manufacturing innovation ecosystem OEMs are the most important drivers of innovation in manufacturing in Massachusetts with connections to all other actors in the innovation ecosystem. Interviews with OEMs in the top 10 manufacturing sub-industries suggested that OEMs draw on the region's capabilities in different ways depending on their industry structure, their development time horizons, and their regulatory environment. In all cases, OEMs consider the region a place for new product development and new product introduction as evidenced by the number of OEM advanced manufacturing R & D facilities located in the state. Companies with an R & D presence include Gillette, Medtronic, ThermoFisher, Raytheon, Phillips Healthcare, and more recently Nihon Kohden as well as GE that moved its headquarters to Boston in 2016. The global production systems for each industry differs and affects what production might be done in a high-wage location like Massachusetts. For example:

• Semiconductors and electronics are largely manufactured in Asia • •

and Mexico and then integrated into other products in the U.S.; there is some specialized production in the U.S. as well. The aerospace and defense industries require largely domestic production, but there is increasing pressure on OEMs to manufacture in the countries of their foreign customers. Manufacturers of measuring devices and medical devices are more likely to keep high-end production in the U.S., particularly if they require regulatory approval; they also benefit from proximity to suppliers for rapid response and small-batch production.

As stated above, OEMs manufacture in Massachusetts for reasons largely linked to innovation and talent. Access to innovation and talent helps the OEMs respond to increasing pressure to cut lead times and 6

Technological Forecasting & Social Change xxx (xxxx) xxx–xxx

E.B. Reynolds, Y. Uygun

• Shorter lead times overall and highly responsive supply chains to meet customer demands that cannot be anticipated ahead of time. • Increasing globalization of the supply chains such that supplies can

meet high quality standards. Interviews highlighted the following attributes of the Massachusetts innovation ecosystem:

• The presence of world-class research universities with high-impact • • •

research groups gives OEMs the opportunity to support unique cutting-edge applied research that can be integrated or translated into competitive products to gain market share Graduates from the state's research universities constitute an important talent pool for large OEMs Besides universities, the state's vibrant startup community is a source of new ideas for products and services; in addition, collaboration with or acquisition of startups can open new market opportunities To rapidly introduce new products, OEMs in Massachusetts can rely on flexible, quick, and reliable suppliers, especially machine shops that can manufacture special parts and components on a small scale.

•

These changes directly impact SMEs within the supply chain. Expectations from OEMs about what it means to be “world class” are evolving and raising the bar for SMEs. The following criteria are seen as standard requirements for top suppliers today:

• Standard certifications (e.g., ISO, AS) • Technical skills (IT, CAD/CAM) • Zero defects in shipped product • 100% on-time delivery • Truly “lean” practices (as discussed in Uygun and Straub, 2011; Uygun and Wagner, 2011) • “Nimbleness and curiosity”, i.e., a mindset of openness to new challenges • Regular (usually yearly) price reductions • Commitment by the SME, at the level of the CEO, to communicate directly with the OEM • Transparency as to cost drivers

At the same time, OEMs in Massachusetts face several key challenges. First, while a well-educated, highly skilled labor force is one of the Commonwealth's major strengths, OEMs are emphatic that access to labor remains a serious problem. This is an area where Massachusetts is under strong pressure from other regions. In some cases the supply of (skilled) labor is better in southern parts of the U.S., and in some cases better abroad (particularly in Mexico). Part of the challenge is the younger generation's perception of manufacturing jobs, which does not reflect the clean, technologically advanced nature of the industry. The Massachusetts manufacturing community is acutely aware of the problem of skilled labor shortages and has taken a number of actions in response, including strengthening its outreach to community colleges and local organizations to promote manufacturing as a viable career, and revising and standardizing training programs to facilitate skills acquisition and credentializing (e.g., the Manufacturing Advancement Center Workforce Innovation Collaborative (MACWIC) program). Second, the importance of government's role in attracting or retaining manufacturing investments cannot be underestimated. Some U.S. states have taken a very aggressive approach to attract manufacturing investment and jobs, actively recruiting manufacturing firms and offering significant incentives to locate facilities in their state. Further, governments of many developing or emerging economies require manufacturers to set up operations in the country if they would like to do business there. U.S.-based OEMs have often responded to such requests without moving essential manufacturing but these kinds of quid pro quo or offset pressures are increasing. Finally, despite the attractiveness historically of low-wage locations like China for production, these places may be less attractive in a new production paradigm where intellectual property is more important and rapid responsiveness and shorter lead times are valued. Several executives in OEMs interviewed stated that China is progressively losing its attractiveness as a low-cost manufacturing location because of rapid wage escalation, poor workforce stability, and the total costs of addressing intellectual property protection. OEM executives outlined several ways in which their relationship with SMEs is changing given increasing global competition. Several important changes in the supply chain in recent years for OEMs in Massachusetts include:

As is apparent from this list, the bar is high and continues to rise for SME suppliers to OEMs. The global pressures on OEMs translate into demands for higher performance from suppliers while at the same time creating some opportunities for regions like Massachusetts to respond to greater demands for customization and flexibility in the supply chain. Clearly, the relationship between OEM and supplier is changing and for those OEMs on the “high road,” investing in and collaborating with a smaller number of suppliers leads to greater innovation and positive outcomes. At the same time, SMEs are required to meet an ever rising standard of excellence. 4.4. SMEs within the manufacturing innovation ecosystem As noted above, 97% of all manufacturing establishments in Massachusetts are small or medium sized. Precision engineering or “machine shops” account for a significant number of these manufacturing SMEs which typically perform contract manufacturing. We divide the SME landscape in Massachusetts into four different types of businesses based on what we learned from our interviews with SMEs. These are classified according to company life-cycle and type of product architecture, as depicted in Fig. 6. Along the horizontal axis, the figure distinguishes between newly founded and incumbent SMEs; along the vertical axis, the distinction is between SMEs that produce parts and components and SMEs that make end products. Startup or spin-off suppliers (bottom left) produce less complex parts and components and seek to engage with large OEMs to sell their products. In terms of life cycle, high-performing startup or spin-off suppliers are on a path to grow to mature small suppliers (the fourth quadrant in the figure) and ultimately to become strategic suppliers. Small suppliers (bottom right) normally start as emerging startup or spin-off suppliers and grow to become part of OEM supply chains. The large number of machine shops in Massachusetts fit into this category. Machine shops are not positioned to become strategic partners because they are engaged in made-to-order manufacturing of less complex parts and do not have proprietary products. For these suppliers, programs to improve process efficiency with initiatives like lean manufacturing or Six Sigma is essential. Startup or spin-off OEMs (upper left) produce more complex, proprietary products that can be marketed by the OEM or as part of a larger

• Integration of supply chain management with engineering to bring • • •

be sourced from firms in any corner of the world as long as they are cost competitive and deliver quality products on time. Instances of firms moving production back to the U.S. where it is becoming more competitive to manufacture, particularly given the emphasis on shorter lead times.

design and technological innovation into the supply chain procurement process earlier. Centralization of supply chain operations across business units or particular products rather than within each business unit. Consolidation of the supply chain to reduce the overall number of suppliers and attendant complexity. Greater emphasis on collaborative partnerships with a select number of strategic suppliers, and a more solutions-oriented approach to suppliers in general. 7

Technological Forecasting & Social Change xxx (xxxx) xxx–xxx

E.B. Reynolds, Y. Uygun

Fig. 6. Classification of SMEs.

incorporate new technological developments (as embodied for example in Industry 4.0 concepts, such as in Karakaya et al., 2016) will SMEs be able to adapt to the significant pressures and changes taking place in global production systems.

system. The pathway for these kinds of SMEs is to grow through new customers and markets into a mature small OEM and ultimately to become a large OEM. Small OEMs (upper right), which have their own product portfolio, seek to enter new markets and connect with other OEMs. These companies are also often better positioned to partner with universities. They often enable new product introductions and undertake prototyping activities for OEMs based in the region. They also play important roles as providers of key equipment and strategic parts for OEMs' physical production systems within the state. Obviously, SMEs in the top left and right quadrants represent the most innovative SMEs and can help drive innovation in the ecosystem. Moving SMEs into these quadrants is one of the challenges for increasing their innovation capacity. Despite growing cost pressures and increasing consolidation within OEM supply chains, Massachusetts SMEs are not only in a good position to take advantage of heightened interest in innovation and shorter lead times, they may also be buoyed by trends that are making manufacturing in the U.S. more attractive generally. OEM interest in greater collaboration also creates new opportunities to build long-term relationships. In addition, the relatively diverse manufacturing-related key sectors of the Massachusetts economy that rely on the state's “manufacturing backbone”, i.e. precision engineering/machine shops, provide a diverse customer base for those SMEs. The ability to supply across sectors helps SMEs in terms of business cycles, cross-selling, and also cross-fertilization with respect to learning and best practices. However, despite these positive trends for SMEs, they still are not easily integrated into the innovation ecosystem in order to learn about and participate in the development of new products and processes. SMEs have little knowledge or access to frontier technologies, whether through OEMs or universities. In particular, despite some pilot efforts within the state, SME relationships with universities are weak. In interviews, SME managers repeatedly stated that many universities are not “user-friendly” places that are hard to navigate. Weak linkages with startups are a further challenge for SMEs. Improving these linkages could open new market opportunities, especially since the vibrant startup community in the greater Boston area needs manufacturing services that could be delivered by small suppliers. Overall SMEs must be constantly engaged in continuous improvement and technological upgrading. Yet networks that help particularly with the latter are difficult to create for SMEs. Only with an ability to

4.5. Universities in the manufacturing innovation ecosystem Much has been written about the important role universities play in fostering innovation and generating economic development benefits for the regional economies in which they operate. One of the obvious advantages of a university to the ecosystem is that, “unlike so many participants in the local economy, they are immobile” (Lester, 2005). Massachusetts universities in particular have an enormous impact on the region's economy: throughout the state, about 500,000 students are enrolled in > 100 institutions of higher education and billions of dollars go to support world-class basic and applied research at these institutions. Entrepreneurial activities on university and college campuses have led to the founding of many innovative startup firms. In addition to all the tangible outcomes they generate, universities also create many positive externalities for surrounding communities (Almeida and Kogut, 1999) and play at least two important roles that can help foster regional economic development. First, universities create a “space for open-ended conversations about industry development pathways and new technological and market opportunities.” They also “increase the local capacity for scientific and technological problem-solving” through the flow of ideas from startups, joint research with companies, consulting, and the hiring of students (Lester, 2005). Advanced manufacturing in Massachusetts has benefited from all of these innovation externalities associated with local universities and colleges. In particular, the state's universities boast top research labs and centers (often supported in part by state and federal funding) that are developing the next generation of advanced manufacturing technologies. Examples include the recently launched Raytheon-UMass Lowell Research Institute, which is focused on flexible and printed electronics and the Novartis-MIT Center for Continuous Manufacturing, which focuses on pharmaceutical manufacturing. Both of these centers are sponsored by large OEMs and support basic and applied R & D. Other centers build on regional strengths in areas such as robotics (e.g., the Wood Hole Oceanographic Institution Center for Marine Robotics and the Umass Lowell New England Robotics Validation and Experimentation Center or NERVE), advanced materials (e.g., the MassNanoTech Institute at Umass Amherst, the Northeastern Nanoscale 8

Technological Forecasting & Social Change xxx (xxxx) xxx–xxx

E.B. Reynolds, Y. Uygun

innovation capacity among SMEs, something German collaborative research models do quite well (Kirkels and Duysters, 2010). Universities play a critical role in the advanced manufacturing innovation ecosystem in the region. However, this role is changing as funding agencies emphasize collaboration and broaden their relationships beyond unilateral agreements with OEMs to include consortiadriven models that in theory engage SMEs. These models require new governance structures and incentive systems to fully engage all the parts of the ecosystem.

Technology and Manufacturing Research Center, and MIT.Nano), life sciences (e.g., the MIT Medical Electronic Device Realization Center or MEDRC and the Umass Lowell Biomanufacturing Center), defense-related research (e.g., Draper Labs and the U.S. Army Soldier Research, Development and Engineering Center), and advanced manufacturing technologies more generally (e.g., the Advanced Technology and Manufacturing Center at Umass Dartmouth, the Lab for Manufacturing and Productivity at MIT, and the Fraunhofer Center for Manufacturing Innovation at Boston University). Some centers, like the Umass Dartmouth Massachusetts Accelerator for Biomanufacturing, are designed specifically to work with startups that can benefit from the use of shared facilities. Clearly, universities are already a critical part of the state's manufacturing innovation ecosystem. Moreover, they are positioned to play an even greater role going forward given the current focus on emerging technologies and industries that are important to the Massachusetts economy. Looking ahead, there are at least two areas of opportunity for universities in the state's advanced manufacturing innovation ecosystem. First, investments in developing advanced manufacturing technologies are critical. Despite flat or declining public funding for basic research in recent years (NSB — National Science Board, 2012), advanced manufacturing has attracted significant national attention and investment. The creation of a National Network for Manufacturing Innovation (NNMI) (AMNPO — Advanced Manufacturing National Program Office, 2015), which proposes to create at least 15 Institutes for Manufacturing Innovation (IMI) around the country, is arguably one of the most important science and technology initiatives put forth by the federal government in recent years. It includes several centers that are supported and led by public-private consortia and that focus on the development of pre-competitive technologies while also building regional capabilities in their focus areas. This effort recognizes the importance of manufacturing to the country's innovation capacity and is based in part on the German Fraunhofer Model and the applied research model of public/private, university and large/small company collaborations in Germany. Massachusetts universities have been successful with three of the NNMI calls for proposals (thus far, they are participating in three IMIs: optoelectronics, flexible hybrid electronics, and advanced fibers and textiles) and will no doubt be included in future bids, given the range of expertise that exists in the state. The NNMI process should help the state develop strategies for strengthening the ecosystem as a whole (particularly around workforce training) and could also be helpful in terms of developing advanced manufacturing technology roadmaps for the region. Another area of opportunity for strengthening the manufacturing innovation ecosystem involves increasing the engagement between universities and SMEs. While there have been several successful examples and pilots, some fundamental obstacles exist that make such collaborations challenging. First, SMEs face organizational challenges when working with universities. As already noted above, SMEs report that they find universities hard to navigate and not user-friendly. Second, universities and SMEs have different objectives and agendas. Academics see innovation as “something that is radically new deriving from newly created knowledge” while SMEs see innovation as creating a product or process that will increase the firm's profits (Massa and Testa, 2008). Third, SMEs are usually working under short- or mediumterm time constraints. Universities work with longer timeframes. Finally and perhaps most importantly, the costs of collaboration can be prohibitive unless funding is provided by the SME or a third party. Finding ways to engage SMEs in research and discussions about new technologies is crucial to increasing their innovation capacity. One way to engage SMEs in university collaborations is through competitive grants, like those offered by the Small Business Technology Transfer (STTR) program. Facilitating and broadening SME-centered industryuniversity collaborations offers another promising path for increasing

4.6. Startups in the manufacturing innovation ecosystem Massachusetts is widely regarded as one of the most innovative and entrepreneurial states in the country. The Milken Institute's State Technology and Science Index 2014 as well as the ITIF's 2014 State New Economy Index rank Massachusetts as number one. The former analyzes technology and science capabilities of each U.S. state alongside their success at transforming those capabilities into companies (BLS — Bureau of Labor Statistics, 2005). The latter evaluates states' fundamental capacities in the “new economy “in terms of knowledge jobs, globalization, economic dynamism, digital economy, and innovation capacity” (Foss, 1996). Innovative startups, which may grow out of universities or out of larger established firms, are at the heart of the state's innovation ecosystem. What is less well known is the extent to which these startups are engaged in advanced manufacturing processes. Research on startups based on technology developed at MIT and licensed through the MIT Technology Licensing Office (TLO) found that approximately 80% of all TLO startups founded between 1997 and 2008 required some kind of production-related capabilities (Reynolds et al., 2014). Generally speaking, firms that license technology through the TLO are less likely to be software-related. In addition, a study of Massachusetts firms that are receiving federal Small Business Innovation Research (SBIR) grants found that at least 15% (or 500 firms) that received grants between 2009 and 2013 were engaged in advanced manufacturing processes. These grants accounted for approximately $200 million of the $1.2 billion total that Massachusetts firms received in SBIR grants over this time period. In terms of total SBIR and STTR grants, Massachusetts is the second most successful state in the country behind California and is the leading state in the country in terms of SBIR/STTR grants per capita. Given the region's strong and growing engineering capabilities and the trend toward combining hardware and software to form “hybrid” technologies (in consumer and medical devices, for example), startups have become an increasingly important source of manufacturing innovation. The emergence of startup incubators and venture capital funds, such as Greentown Labs and Bolt, in the Boston area that focus on hardware companies reinforces the support system for such startups. Private companies such as Autodesk and Flextronics are also investing in new spaces for manufacturing-related startups. While there are increasingly more resources for these types of startups, these companies often face challenges in the scale-up phase. Growing innovative companies is a subject that is increasingly drawing attention, both in the United States and globally, as regions and countries focus on reaping some of the downstream benefits of their startup ecosystems. The scale-up process is particularly challenging for startups engaged in the production of complex production-oriented technologies (as opposed to software). Such technologies often require larger amounts of capital and longer time horizons (often over ten years) to demonstrate their viability at commercial scale (Reynolds et al., 2014). There are several points in the early stages of this process where actively engaging with the manufacturing innovation ecosystem could help startups achieve scale and, importantly, facilitate scale-up in the Commonwealth. First, startup technology companies often have a promising idea for 9

Technological Forecasting & Social Change xxx (xxxx) xxx–xxx

E.B. Reynolds, Y. Uygun

6. marketing.

a new product but lack the skills to manufacture it. Early-stage prototyping, which requires multiple iterations that can take several months or several years, often requires close proximity between the startup and its suppliers so that the latter can respond to changes quickly while still providing high quality. Massachusetts, with its extensive network of high-precision machine shops and experience in new product introductions, provides competitive advantages to startups at this stage of development. However, connections between the innovative startup community and the state's high-precision machine shops are weak, with few formal or systematic forms of interaction. Thus, it will be important to underscore the region's capabilities in prototyping and early-stage piloting and open better channels of communication between these communities. A recent pilot with Greentown Labs, an incubator for clean energy companies, exemplifies a first step in this process. Whether the state can also position itself to support scale-up beyond pilots remains to be seen. Recently, companies have been more likely to go abroad to lower-cost locations for commercial scale-up. A second area of opportunity for supporting the scale-up process in the region is with potential customers. Early adopters are among the most important factors that can help a startup “cross the chasm” in the early stages of scale-up (Moore, 1991). Customers or potential customers who are willing to partner during beta testing of a new product are critical. Increasingly, strategic partners have been playing this role in the United States. Such partners, which are usually large companies (including OEMs), are becoming more engaged in startups through equity investments and other arrangements (Lerner, 2012) in which they provide not only capital but capabilities and know-how in exchange for the exposure and experience they gain from the startup. This is particularly important for startups that, because of their longer development horizons and higher capital needs, do not necessarily fit well with a venture capital funding model. Given the diversity and sophistication of OEMs in Massachusetts, a more systematic effort could be made to connect startups and OEMs. Introducing large potential customers to startups is the goal of several initiatives that are already in place (e.g., the Northeast Clean Energy Council Strategic Partners program and Fintech Sandbox's efforts to provide “scrubbed” financial data from large financial services firms to financial services startups for beta testing). More could be done in this area, particularly with respect to advanced manufacturing companies, where the scale-up process can be more challenging due to capital requirements and longer time horizons.

These activities are executed through a range of intermediaries, i.e. organizations funded by the state that may also have some industry support through cash or in-kind contributions. On the whole, state supported programs have assisted SMEs on a one-on-one basis whether through workforce training grants, lean practices, executive education, or certification programs. Several of these programs are of high caliber and recognized nationally, including the aforementioned MACWIC program which has developed a stacked credential system for workers to climb a certification program in manufacturing skills. In addition, the state piloted a program for one year that provided matching grants for SMEs to consult with “Innovation Centers” on how to develop new products or upgrade their technology or engineering capabilities. Technology and cluster development work have had little focus until recently on manufacturing-related industries. Many of the efforts to support SMEs are critical to their survival but on the whole, they represent the bare minimum for what is needed for SMEs to remain competitive and to drive greater innovation within the regional ecosystem. The programs have very little leverage in improving the system as a whole since, rather than support the connectivity of the ecosystem on a “systems-wide” basis, they tend to focus on “point solutions”; one company/one worker at a time. They are also supply-side focused (workers, suppliers) and do not take into sufficient account the demand-side of the equation, namely OEMs and global markets. Not surprisingly, this approach has led to a short-term outlook for the state's advanced manufacturing and the lack of an overall strategic vision that looks out five to ten years in terms of supply chain developments, technology roadmaps, and talent and training needs. This review of the key nodes in the manufacturing innovation ecosystem highlights the multiple ways these actors coexist, in many cases working closely together, but more often than not, not as well aligned as they could be. To fully maximize the system as such, there are several ways discussed above that these nodes could be better linked, leveraged and aligned to enhance the systems as a whole. 4.8. Case study: increasing innovation capacity in SMEs in Germany Germany provides an interesting case study for the United States with respect to strengthening SMEs in manufacturing. Despite Germany's relatively high labor costs, 19% of all employees in the country work in manufacturing (Statista, 2015). German “Mittelstand” companies in particular (medium-sized family-owned private companies with a long tradition and solid finances) are highly successful in global manufacturing markets. In terms of overall manufacturing output Germany ranks 4th in the world (UNCTAD — UN Conference on Trade and Development, 2012). German approaches to innovation, upgrading and training/apprenticeships have often served as models (see e.g. Berger, 2013). We chose Germany as a useful case study for Massachusetts because of Germany's success with building a strong SME manufacturing base. The presence of the Fraunhofer Institutes which act as bridges between research universities and industry is a prominent and highly cited factor for Germany's successful SMEs. The Fraunhofer Society, headquartered in Munich, is Europe's largest application-oriented research organization and comprises over 60 institutes across Germany each of which focuses on a particular technology. Fraunhofer's mandate is to develop applied technologies for industrial companies, in particular working with SMEs to bring cutting-edge technologies to market (Fraunhofer, 2015). Numerous branches of the Fraunhofer Institutes have been inaugurated around the globe, including in Massachusetts. In the context of our research, probably the most important way that innovation is fostered among manufacturers in Germany, particularly among SMEs, is through the support of industry-university applied research. Germany has a long history of investing in applied research in which industry plays an important role. The German Federal Ministry

4.7. State government & manufacturing intermediaries The state plays an important supporting role in fostering innovation and directly or indirectly working with the key nodes within the ecosystem. The role of government in supporting innovation according to Vertova (2014) are two-fold. They can act as institutions through innovation-friendly laws, regulations, legal and patent systems on the one hand, and as organizations that foster or enable innovation through the creation of state-owned research labs, centers, or sponsors of other third-party organizations on the other. Overall, government, through its institutions, organizations and associated programs and policies can contribute in a variety of important ways including encouraging an innovation culture, enhancing technology diffusion, promoting networking, and leveraging R & D while creating a predictable and stable environment for investment. In the case of Massachusetts, the state's policies with respect to the manufacturing ecosystem have historically focused primarily on SME workforce training. Below we list the six primary activities that are supported by the state: 1. 2. 3. 4. 5.

process improvements such as lean manufacturing workforce training strategic technology and cluster development technical and engineering process support managerial and professional education 10

Technological Forecasting & Social Change xxx (xxxx) xxx–xxx

E.B. Reynolds, Y. Uygun

Table 1 Overview of different applied research funding models in Germany. Type of program

Multilateral consortium-based research projects

Bilateral research projects

SME network projects

Industry-oriented research projects

Aim

Consortium-based joint development of pre-competitive product and process innovations Universities Research institutes (FhG, “AnInstitutes”, etc.) Large companies SMEs Consultancies Intermediaries 100% for Univ./RI < 50% for SMEs Individual rate for large comp. (usually 20%) BMBF standard programs BMBF special programs (Excellence Clusters, Research Campus, etc.) BMWi special programs (“Autonomik”, etc.)

University-SME-based development of marketable prototypes of product innovations Universities or research institutes SMEs

Configuration of SME networks to jointly develop marketable product innovations Universities or research institutes SMEs (2 +)

Making research findings accessible to SMEs to facilitate all kinds of innovations Universities or research institutes SME advisory board

Target group

Funding scheme

Funding body

• • • • • • • • • • • •

• •

• •

for Univ./RI • 100% • < 50% for SMEs

for Univ./RI • 100% • < 50% for SMEs

standard program “KMU• BMBF innovativ” ZIM SOLO • BMWi–AiF • Ziel2 (2 + SMEs possible)

• BMWi–AiF ZIM KOOP

a

• •

for Univ./RI with • 100% mandatory SME participation

b

• BMWi–AiF-IGF

BMBF: German Federal Ministry for Education and Research. BMWi: German Federal Ministry for Economic Affairs and Energy. AiF: German Federation of Industrial Research Associations. ZIM: Central Innovation Program SME. Ziel2: regional development program of the EU for economically less-developed regions in Europe. a To date: EUR 750M for 1100 Projects. b In 2013: EUR 424M for 3000 Projects.

universities and large OEMs. Significant inter-firm communication with OEMs allows for first-hand knowledge exchange and mutual understanding of OEMs' future developments, technology roadmaps, and market opportunities. The model enables faster development and wider diffusion of innovations that would otherwise have occurred much more slowly and on the basis of bilateral cooperation, reducing diffusion into the wider manufacturing ecosystem. Based on fieldwork done in Germany, several success factors were identified that determine the effectiveness of such projects in bringing innovations to market that need to be taken into consideration before adopting such a model to the U.S.:

for Education and Research (“BMBF”) and the German Federal Ministry of Economic Affairs and Energy (“BMWi”) have created several programs that focus on building innovation capacity among SMEs. Table 1 gives an overview of the most important initiatives and programs of funding applied research. Funding by BMWi is mainly through special programs and the German Federation of Industrial Research Associations (German abbreviation: “AiF”). BMBF has regular funding programs that are announced at random intervals with a clear technological focus. In addition, BMBF also starts special programs, like the Leading-Edge Cluster Initiative, to support bigger projects in conjunction with the National High-Tech Strategy. One of the standard funding models of the German Federal Ministry for Education and Research is the research consortium comprised of SMEs, universities, research institutes, large companies, consultancies, and intermediaries that focus on the development of pre-competitive product and process innovations. This model is also promising for the U.S. in general and Massachusetts in particular. Our research in Germany revealed two different kinds of consortium-based relationships which foster different types of innovation:

• Size of the consortium with the ideal size being between three and • •

• The value-chain based approach where industrial partners represent

•

adjacent tiers in the value chain with potential supplier-buyer relationships. This type of relationship is in general more likely to result in marketable products since a potential value chain with well-known supply chain structures is already in place. On the other hand, this approach seems to support relatively incremental innovations. The complementary-competency based approach that brings together partners from different industries and with different technologies to promote experimentation in order to develop truly novel products and processes. In this model, there is a higher degree of uncertainty as to project results and more radical innovations are likely.

five partners, ideally at least one participant from each category, SME, University or Research Institute, and large OEMs. Close alignment between the consortium R & D objectives and participants' existing strategies: companies focusing on a specific future technology (e.g. carbon fiber), are more likely to be committed and contribute to the project since it feeds directly into firm interests. Pre-competitive research is a mandatory requirement not only in Germany but throughout Europe. The pre-competitive nature of these projects prohibits any kind of favoritism in terms of funding companies' direct R & D activities. The research focuses on demonstrating the viability of technologies at the pre-prototyping stage before any one company has taken on greater risk with a particular product or technology. This is of particular concern in the Medical Device industry where the subsequent clinical trials and the approval process by the authorities (FDA, etc.) can take many more years.

Currently, under the banner “Industry 4.0,” Germany is very keen to push SMEs toward making more intelligent manufacturing systems and products by using the above mentioned funding models. The German framework for Industry 4.0 is the technical integration of cyber-physical systems in production and logistics as well as the application of the industrial internet of things (IIoT) to create smart factories. In such a decentralized manufacturing model, intelligent products are developed

In both cases, the model is an innovation enabler that brings together partners who otherwise would not meet, particularly in the case of SMEs, who are central to the conversation that occurs with 11

Technological Forecasting & Social Change xxx (xxxx) xxx–xxx

E.B. Reynolds, Y. Uygun

collaboration across OEMs in the same or different industries when it comes to upgrading the supplier base in the state, even when OEMs share similar suppliers. Initiatives to upgrade supplier capabilities based on collaboration across OEMs from different industries could provide a robust mechanism for leveraging state resources, sharing best practices, and expanding support to SMEs. Such initiatives could focus not only on process and quality improvements but also on technical problem solving and new product development. The abovementioned consortium-based applied research projects can also create incentives to support this work. Grant funds should be used to encourage regional consortium-based projects including universities, OEMs, and SMEs that focus on pre-competitive product and process innovations, similar to the German model. Those consortiumbased projects are a unique and important initiative to support the joint development of pre-competitive innovative products or processes between large OEMs, SMEs, universities, and applied research institutes (especially Fraunhofer) in Germany. The participation of SMEs is mandatory. For universities and applied research institutes, funding covers all project-related expenses (personnel, hardware, etc.), while participating companies are reimbursed with up to 50% of their costs depending on company size; in general SMEs receive up to 50%, large companies around 20%. For Massachusetts, experience in consortiabuilding in the process of applying for the federal Institutes for Manufacturing Innovation (IMIs) could be instructive in developing such kind of regional, project-based consortia. Third, the role of the state should be to support and enhance those linkages within the innovation ecosystem and work at a systems-level where possible. As outlined earlier, most of the state-level programs and policies are supply-side focused (firms, workers) and assist firms individually through cost-sharing programs that provide one-on-one point solutions in process improvement, training and certification. These are important building blocks that firms need in order to meet the basic requirements for suppliers today. However, to better leverage resources and work at a systems-level scale, state programs and policies should be more focused on the demand-side and how to connect suppliers and workers to the demands of a rapidly changing environment from both a technological and customer perspective. An outwardlooking perspective on markets, technologies, skills and competitors will position the state as a forward-looking partner with the private sector. In 2015 and 2016 the state of Massachusetts began to shift its emphasis in this direction with greater investment in innovation-driven activities (consortium-based R & D efforts) and the development of a state-wide advanced manufacturing strategy and agenda.

in horizontally and vertically integrated production systems that lead to consistent engineering throughout the value chain (CPG — Communication Promoters Group of the Industry-Science Research Alliance and National Academy of Science and Engineering, 2013). This is the strategic advanced manufacturing agenda for the coming years and even decades to funnel investments into R & D with tailored funding programs and initiatives. Participants from industry, academia, society, and politics came together to develop a strategic plan around challenges and R & D needs as well as how best to manage the resulting societal change. This overall agenda and process for engagement of key stakeholders is a valuable model for Massachusetts as well. 5. Strengthening the manufacturing innovation ecosystem The previous section outlined the role of each of the key nodes in the advanced manufacturing innovation ecosystem and the challenges and opportunities they face in an increasingly competitive global system of production. There are clearly steps to be taken to strengthen each of the nodes while also strengthening the links between them. Beyond these steps, there are three broader goals that should be pursued to strengthen the innovation ecosystem as a whole. First, the region needs an advanced manufacturing strategy and agenda. In contrast to the state's other cluster-focused strategies (e.g., in biotech), advanced manufacturing requires the development of crosscutting capabilities that work across industries. This makes it more challenging to develop strategies around particular capabilities because they exist across numerous industry clusters and don't have one constituency. To support advanced manufacturing in the state, a deep understanding of manufacturing capabilities, their importance within key clusters, and trends in technology as well as in global production systems is required. A robust analysis of the state's advanced manufacturing capabilities combined with engaging key manufacturing leaders in the state is necessary to develop a solid advanced manufacturing strategy and agenda that can set priorities for the next five to 10 years, similar to the German Industry 4.0 initiative. In doing so, participants from industry, academia, society, and politics need to come together to develop a strategic plan with R & D needs as well as how best to manage the resulting societal change. Massachusetts, in particular, would benefit from the establishment of some kind of Manufacturing Innovation Advisory Board with representatives from several high-performing OEMs, SMEs, universities, and others. Such a private-sector led board could promote long-term strategic thinking with an understanding of the competitive landscape nationally and globally, support collective action (and impact), as well as help position SMEs for success. Second, OEMs need to collaborate with each other as well as with SMEs. As Rodrik (2004) states, modern industrial policy should seek to create linkages between firms and economic actors and provide scope for experimentation so that firms can engage in a “self-discovery process” to find out where their true comparative advantages lie, something that is often not obvious ex ante. To this end, the key to strengthening the advanced manufacturing innovation ecosystem are the linkages that can be encouraged among OEMs and key innovation partners, and between OEMs and SMEs. As outlined in this paper, OEMs are a driving force for innovation in Massachusetts, yet their collective perspective on how to increase and accelerate innovation is not well understood. They also work in isolation one from the other, though they share common concerns about regional capabilities (i.e., labor force, supply chain, infrastructure). With a window into global trends, R & D opportunities, supply chain demands, and training needs five to ten years out, OEMs need to be engaged to help set the state's manufacturing innovation strategy going forward and ensure their suppliers are globally competitive. Most OEMs have their own individual supplier development programs to help suppliers produce efficiently and meet the OEMs' delivery, cost, and quality requirements. However, there is little

6. Conclusion As outlined, Massachusetts has significant assets and expertise in advanced manufacturing that have developed over decades, creating deep capabilities that help to drive innovation in some of the state's leading industry clusters. Despite its relatively high wages, the region attracts significant investments in advanced manufacturing, particularly related to new product introductions. But important changes are taking place, within companies and how they are organized for production, in terms of new “game-changing” technologies, and in the global economy as regions and countries work aggressively to increase manufacturing investments and build capabilities. This changing landscape requires more and asks more of the innovation ecosystem within which firms are operating. Regions like Massachusetts, with strong nodes within the ecosystem, particularly in the area of universities and startups, must find ways to strengthen the connections between the key actors and foster greater collaboration to support building innovation capacity. Existing work in this field tends to describe innovation systems in static terms with little attention to the potential for dynamic change. For this reason, the term “ecosystem” is more appropriate than regional 12

Technological Forecasting & Social Change xxx (xxxx) xxx–xxx

E.B. Reynolds, Y. Uygun

Jackson, D.J., 2011. What is an Innovation Ecosystem? National Science Foundation, Arlington, VA. Karakaya, E., Guller, M., Uygun, Y., Kuhn, A., 2016. Development of an agent-based simulation for the cellular transport system and scenario-based performance analysis. Int. J. Electr. Electron. Data Commun. 4 (1) (January). Kirkels, Y., Duysters, G., 2010. Brokerage in SME networks. Res. Policy 39 (3), 375–385. Kirner, A., Kinkel, S., Jaeger, A., 2009. Innovation paths and the innovation performance of low-technology firms — an emprical analysis of German industry. Res. Policy 447–458. Kline, S., Rosenberg, N., 1986. An overview of innovation. In: National Systems of Innovation. Oxford University Press. Lerner, J., 2012. The Architecture of Innovation: The Economics of Creative Organizations. Lester, R., 2005. Innovation and the competitiveness of local economies. In: Working Paper 05-010. MIT Industrial Performance Center, Cambridge, MA. Maskell, M., 2001. Towards a knowledge based theory of the geographical cluster. Ind. Corp. Chang. 10 (4), 921–943. Massa, S., Testa, S., 2008. Innovation and SMEs: misaligned perspectives and goals among entrepreneurs, academics, and policy makers. Technovation 28, 393–407. Moore, G., 1991. Crossing the Chasm: Marketing and Selling High-tech Products to Mainstream Customers. Nelson, R.R., Rosenberg, N., 1993. Technical innovation and national systems. In: National Innovation Systems: A Comparative Analysis. Oxford, Oxford University Press. NSB — National Science Board, 2012. Diminishing Funding and Rising Expectations: Trends and Challenges for Public Research Universities. (URL: https://www.nsf.gov/ nsb/publications/2012/nsb1245.pdf, Arlington, VA). OneSource, 2015. OneSource Global Business Browser. [Online]. Available: http:// globalbb.onesource.com/web/company/companyfinder.aspx. PCAST — President's Council of Advisors on Science and Technology, 2011. Report to the President on Ensuring American Leadership in Advanced Manufacturing. (URL: https://www.whitehouse.gov/sites/default/files/microsites/ostp/pca, Washington D.C.). PCAST — President's Council of Advisors on Science and Technology, 2014. Report to the President: Accelerating U.S. Advanced Manufacturing. (URL: https://www. whitehouse.gov/sites/default/files/microsites/ostp/PCAST/amp20_report_final.pdf, Washington D.C.). Pisano, G., Shih, W., 2011. Producing Prosperity. Harvard University Press, Cambridge, MA. Reynolds, E., Samel, H., Lawrence, J., 2014. Learning by building: complementary assets and the migration of capabilities in U.S. innovative firms. In: Production in the Innovation Economy. MIT Press, Cambridge, MA, pp. 81–107. Rodrik, D., 2004. Industrial Policy for the Twenty-first Century. John F. Kennedy School of Government, Harvard University. Romer, P.M., 1986. Increasing returns and long run growth. J. Polit. Econ. 94, 1002–1038. Schumpeter, J.A., 1934. Theory of Economic Development. A Study About Company Profit, Capital, Credit, Interest, and Economic Cycle. Harvard Economic Studies, Cambridge, MA. Statista, 2015. Anzahl der Erwerbstätigen im produzierenden Gewerbe (ohne Baugewerbe) in Deutschland von 1991 bis 2013. [Online]. Available: http://de. statista.com/statistik/daten/studie/2190/umfrage/anzahl-der-erwerbstaetigen-improduzierenden-gewerbe/. Sturgeon, T., 2001. How do we define value chains and production networks? IDS Bull. 3. UNCTAD — UN Conference on Trade and Development, 2012. UNCTADstat Data Center — 2012 Data. [Online]. Available: http://unctadstat.unctad.org/wds/ReportFolders/ reportFolders.aspx. US Census Bureau, 2012. Manufacturing: geographic area series: detailed statistics for the state: 2012. [Online]. Available: http://factfinder.census.gov/faces/tableservices/ jsf/pages/productview.xhtml?pid=ECN_2012_US_31A2&prodType=table. Utterback, J., Suarez, F., 1991. Innovation, competition, and industry structure. Res. Policy 22 (1), 1–21. Uygun, Y., Reynolds, E.B., 2016. Advanced manufacturing ecosystems. In: Jeschke, S., Brecher, C., Song, H., Rawat, D.B. (Eds.), Industrial Internet of Things — Manufacturing Cyber-physical Systems. Springer, pp. 691–715. Uygun, Y., Straub, N., 2011. Human-centred model for application of lean production in networks. In: ElMaraghy, H. (Ed.), Enabling Manufacturing Competitiveness and Economic Sustainability. Springer, New York, pp. 660–665. Uygun, Y., Wagner, S.U., 2011. Guidelines for human-based implementation of lean production. In: Duffie, N. (Ed.), Proceedings of 44th CIRP International Conference on Manufacturing Systems — New Worlds of Manufacturing. Omnipress, Madison, Wisconsin. Van de Ven, A.H., 1986. Central problems in the management of innovation. Manag. Sci. 32 (5), 590–607. Vertova, G., 2014. The State and National Systems of Innovation: A Sympathetic Critique. Levy Economics Institute of Bard College. Annadale-on-Hudson, NY.

innovation system (RIS) because it conveys the dynamic process in play as opposed to a permanent static system. This paper has attempted to convey the dynamic nature of the ecosystem by analyzing the key actors and institutions within the manufacturing innovation ecosystem, the roles they play, and the strengths and weaknesses within the ecosystem. It also provides a window into current trends and dynamics within global manufacturing supply chains and how increasing competition is putting pressure on regional innovation ecosystems and the firms, particularly SMEs, operating within them. Our central empirical claim about Massachusetts is that it is “assetrich” but “linkage-poor,” i.e., there is significant innovation capacity within each node of the system but communication between them is limited (universities, SMEs, OEMs, startups), with SMEs above all cut out of crucial networks. Future research should focus on a more micro-level on the mechanisms by which linkages within the innovation ecosystem are fostered and strengthened. Additional case studies could shed light on specific examples of how the nodes within an innovation ecosystem interact (or not) and the specific roles of public, private and non-profitsector actors. Acknowledgement The research for this paper was supported by a grant from MassDevelopment. The authors are grateful to MassDevelopment for its commitment to and support of this work. We also wish to thank the many people we interviewed in the course of the research who were generous with their time and insights. References Almeida, P., Kogut, B., 1999. Localisation of knowledge and the mobility of engineers in regional networks. Manag. Sci. 45 (7), 905–917. AMNPO — Advanced Manufacturing National Program Office, 2015. NNMI: snapshot. [Online]. Available: http://manufacturing.gov/nnmi.html. Asheim, B., Gertler, M., 2005. The geography of innovation. In: Fagerberg, J., Mowery, D.C., Nelson, R.R. (Eds.), The Oxford Handbook of Innovation. Oxford University Press, Oxford. Asheim, B.T., Isaksen, A., 2002. Regional innovation systems: the integration of local ‘Sticky’ and global ‘Ubiquitous’ knowledge. J. Technol. Transf. 27, 77. http://dx.doi. org/10.1023/A:1013100704794. Berger, S., 2013. Making in America. MIT Press, Cambridge, MA. Bloomberg Innovation Index, 2016. URL: https://www.bloomberg.com/news/articles/ 2016-12-22/here-are-the-most-innovative-states-in-america-in-2016. BLS — Bureau of Labor Statistics, 2005. Location Quotient Calculator. [Online]. Available: http://www.bls.gov/cew/cewlq.htm. BLS — Bureau of Labor Statistics, 2015. International Comparisons of Annual Labor Force Statistics, 1970–2012. [Online]. Available: http://www.bls.gov/fls/flscomparelf. htm. Bluestone, B., Gartsman, A., Walsh, D., Eckel, R., Huessy, J., 2012. Staying Power II: A Report Card on Manufacturing in Massachusetts 2012. The Kitty and Michael Dukakis Center for Urban and Regional Policy — School of Public Policy and Urban Affairs — Northeastern University, Boston. Braczyk, H.J., Cooke, P., Heidenreich, M., 1998. Regional Innovation Systems: The Role of Governances in a Globalized World. Psychology Press. Cooke, P., 2001. Regional innovation systems, clusters, and the knowledge economy. Ind. Corp. Chang. 10 (4), 945–974. CPG — Communication Promoters Group of the Industry-Science Research Alliance & National Academy of Science and Engineering, 2013. Recommendations for Implementing the Strategic Initiative Industrie 4.0 — Final Report of the Industrie 4.0 Working Group. acatech, Frankfurt a.M. D'Allura, G., Galvagno, M., Mocciaro Li Destri, A., 2012. Regional innovation systems: a literature review. Bus. Syst. Rev. 1 (1), 139–156. Edquist, C., 2005. Systems of innovation approaches — their emergence and characteristics. In: Systems of Innovation: Technologies, Institutions, and Organizations, Abingdon, Oxon, Routledge, pp. 1–35. Foss, N., 1996. Higher-order industrial capabilities and competitive advantage. J. Ind. Stud. 3 (1), 1–20. Fraunhofer, 2015. About Fraunhofer. [Online]. Available: http://www.fraunhofer.de/en/ about-fraunhofer.html. Fromhold-Eisebith, M., 2007. Bridging scales in innovation policies: how to link regional, national and international innovation systems. Eur. Plan. Stud. 15 (2) (February). Hounshell, D., 1984. From the American System to Mass Production: 1800 to 1932. Johns Hopkins Press. Inkinen, T., Suorsa, K., 2010. Intermediaries in regional innovation systems: high-technology enterprise survey from northern Finland. Eur. Plan. Stud. 18 (2) (February).

Elisabeth B. Reynolds, Ph.D., Executive Director, MIT Industrial Performance Center and Lecturer, MIT Department of Urban Studies and Planning.

Elisabeth works on issues related to systems of innovation, regional economic development and industrial competitiveness. She has focused in particular on the theory and practice of cluster development and regional innovation systems and advises several organizations in this area. Her current research focuses on the pathways that U.S. entrepreneurial firms take in scaling production-related technologies, as well as advanced

13

Technological Forecasting & Social Change xxx (xxxx) xxx–xxx

E.B. Reynolds, Y. Uygun manufacturing, including the globalization of the biomanufacturing industry. She is a member of the Massachusetts Advanced Manufacturing Collaborative Executive Committee.

Yilmaz' research focuses on developing dynamic simulation models to understand innovation “ecosystems” for advanced manufacturing companies. With his appointment in the Department of Mathematics and Logistics at Jacobs University in Bremen (Germany) he also conducts comparative studies with Germany in the realm of advanced manufacturing innovation ecosystems.

Before coming to MIT for her Ph.D., Liz was the Director of the City Advisory Practice at the Initiative for a Competitive Inner City (ICIC), a non-profit founded by Professor Michael Porter focused on job and business growth in urban areas.

Prior to his appointment at the IPC, he worked as research associate with the Fraunhofer Institute of Material Flow and Logistics as well as with Dortmund University of Technology.

Liz has an A.B. from Harvard in Government and was the Fiske Scholar at Trinity College, Cambridge. She holds a MSc. from the University of Montreal in Economics and a Ph.D. from MIT in Urban and Regional Studies.

Yilmaz holds a MSc. from the University of Duisburg-Essen and two Ph.D.s from Dortmund University of Technology and University Duisburg-Essen, respectively.

Yilmaz Uygun, Research Affiliate, MIT Industrial Performance Center and Professor for Logistics Engineering at Jacobs University Bremen.

14