Through policy, infrastructural investments, and demand management, government can either nurture or choke off the growth of a cluster . All elements are crucial to a successful cluster however it is often easiest for governments to manage demand, raising taxes or creating preferential treatment policies is faster and easier to do than building an industry or a research institution. In the case of Southern California’s Water industry, providing water to a populous semi-arid urban region establishes sophisticated demand. In meeting these demands, local water technology firms could prosper; however, the organizational structure of the markets might inhibit this advantage. Porter has explicitly argued that environmental conditions—and the legislative controls to fix them—can serve as the impetus that initially spurs innovative industries . It is important to recognize that traditional industrial economic clusters are not intrinsically centers of innovation, static or mature industries that no longer innovate will also often agglomerate doe to the cost saving benefits agglomeration economies. These industrial nodes will still give a region a competitive advantage, but they are often in lower value goods such as industrial products or mature manufactured goods where the opportunity cost of potential economic gains from further innovation is outweighed by the benefits of lower cost inputs—materials and labor.. Under freer trade regimes it tends to be these types of cost saving agglomerations are the most easily outsourced to cheapest global locations .

In rapidly growing technological industries that require constant innovation, economic geographers have documented significant benefits from competition and knowledge spillovers. In fact clustering appears particularly critical to industries that are rapidly evolving . Complex industries those that require industry R&D,ebb and flow table university research and skilled labor as critical inputs . There is a lively academic debate as to what the initial impetus behind innovative clusters formation . Theories such as Porter’s would point to factor endowments for explanations. While others have narrowed this somewhat and looked to individual agents or networks for explanation. For example Feldman and Francis have studied entrepreneurs in the Internet and biotechnology clusters in washington D.C. They argue that individual entrepreneurs are generally the key agents behind innovative agglomerations . Thriving centers of innovation usually have industrial agglomerations as their base, but they also have strong regional institutional assets such as world-class research universities, multiple financing options, or established networks of entrepreneurs . Additionally, these regions have large numbers of highly capitalized firms competing with one another to develop the ‘next big thing.‘ The presence of human capital is particularly important. Ausdrestch and Feldman write that entrepreneurs tend to “originate in locations with strong knowledge assets” and note that these areas also correlate with the highest economic growth. The combination of human capital, research capacity, and inter-firm competition in close proximity strongly promotes innovation. Scholarship has also found a strong correlation with regional R&D expenditures and innovation, with expenditures made by private companies having a larger role then universities . Regions with higher private R&D expenditures tend to have more innovation in larger firms, while regions with higher university expenditures tend to have more innovative small firms .

Zucker and Darby found a strong correlation of start-ups with researchers, but in their case they found that the presence of certain “star scientists,” are the most important element. Regardless of the mechanisms it is clear that Jacobs’ insights into externalities and economic agent diversity seem to be valid. The most popular example of a thriving innovative cluster is modern Silicon Valley, which owes its origin to the presence of a simple cluster of small tech firms loosely centered on Stanford University and eventually growing into the global technology center that it is today . Today the region continues to have a vast diversity of entrepreneurial firms, both large and small. A healthy ecosystem with both universities and large and small firms all investing in R&D seems to yield the strongest innovation systems. Knowledge spillovers have ample opportunities to occur in these systems. In fact these regions are growing even more innovative. For example Sonn & Storper, found that United State’s regional production of patents is growing even more concentrated in certain innovative regions. Michael Storper has argued that regions with strong “relational assets,” or untraded inter dependencies, are the regions where innovation is most likely to occur. Relational assets include local tacit knowledge, face-to-face interchange, social habits and norms, institutions . Storper and Venables have developed the idea of “buzz” in order to try and understand what is going on in these regions. It is clear that something is happening at the regional level and that spillovers are happening, however, the individual mechanisms are unclear. Critics argue that nebulous terms such as ‘relational assets’ are “fuzzy”—in the words of development scholar Ann Markensen . All agree that further scholarship into the individual mechanisms of knowledge transmission is needed . Innovation Systems scholarship attempts to circumvent these difficult to study interactions while seeking to understand the larger context in which innovation happens.

Schumpeter was one of the first innovation scholars to recognize that innovations often clustered around certain industries or time periods . He speculated that these clusters affected business cycles and could generate waves throughout the economy . Seeking to understand how innovations create dynamic interactions between geographical and sectoral clusters and temporal economic waves led scholars to turn to systems theories for explanation. The fact that learning is key to innovation and learning occurs through dynamic relationships and networks only further serves to reinforce the need to seek out dynamic explanations to understand how innovation happens . Systems engineers define a system as “a set of interrelated components working toward a common objective.” They are made up of components, relationships, and attributes . Components are the operating parts of a system, relationships are the linkages between them,flood table and attributes are the properties of each. In mechanical systems components might be the physical parts that make up a machine. In social systems components are generally actors such as firms, individuals, or organizations, but they can also be institutions such as patent laws. To develop theories of systems innovations scholars looked to evolutionary theory for insights into learning and combined these with institutional theories as a basis for systems theories of innovation. Evolutionary theory – with its focus on dynamics and the process of transformation – provides an analytical framework to understand change. Systems approaches emphasize that learning is both an individual and a collective act. This means that learning will occur not only within firms, or individual institutions, but potentially anywhere in the system . Thus an innovation system can be defined as all institutions and economic structures that affect technological change: competing firms, organizations, universities, research centers, government agencies, legal and financial institutions etc. Furthermore each of these will be characterized by “specific learning processes, competencies, beliefs, objectives, organizational structures and behaviors” . It almost goes without saying that the development of an innovation system occurs over decades . There are several different approaches to the study of innovation systems but in general they are defined by either geography, notably National Innovation Systems , Regional Innovation Systems , or by technology, principally the Sectoral Innovation Systems .

Other scholars have refined their approaches through the use of a narrower lens. For example Carlsson and Stankiewicz have pioneered a technological systems approach, a dynamic approach which attempts to follow a technology rather then a sector, or geographical region. Others such as Håkansson have also focused on technologies but linked to their studies on industrial networks. These systems approaches share similar goals but have different boundaries and often employ differing methodologies. Edquist and Johnson , Lundvall and Nelson have all pursued the study of national systems of innovation , but have placed emphasis on different aspects of the systems. For example, Nelson’s fifteen-country analysis identified research and development investment as a critical component of successful NIS . Nelson emphasizes that the economic incentives for innovation have to be present while sources of research and development financing must also exist . Porter’s studies on the competitiveness of industrial agglomerations are also often placed in the NIS literature—although his focus tends to be more on economic performance rather then pure innovation . Porter and Nelson share a focus on market demand as the critical component to shaping innovation investment . In contrast, Edquest and Johnson have often focused their NIS analysis on a nation’s institutional makeup . In their comparative studies of the European industrial regions, most notably a cross comparison of Badden Wurttemburg Germany and Wales, Cooke and Morgan found that Porter’s diamond model of interconnected heterogeneous firms failed to explain why one region succeeded while others failed . Research into identifying the causal factors behind such studies later became the genesis for a more narrow regional systems approach . The RIS approach places a much stronger focus on cultural factors that build trust and social network relationships . This approach closely aligns with the learning regions and relational assets scholarship . It also aligns with the general understanding that urban regions are the critical key nodes in today’s world . An RIS approach recognizes that regions are subservient to the National system which sets research priorities and legal institutions, but believes regions have some sway in those decisions ; the more decentralized a national system is, the more sway powerful regions will have.RIS approaches emphasize five principal concepts: region, innovation, network, learning, and interaction . A region in an RIS is not necessarily politically bounded but is defined by a shared trust and collective order Innovation looks for measurements of innovation, or the successful commercialization of knowledge. Networks seek to identify cooperation-based networks and analyzes knowledge flows, including external knowledge flows. Learning seeks to identify if tacit learning is occurring and if organizations are incorporating that knowledge successfully. Interaction aims to measure if there are opportunities for learning from each other and external groups. A region can be said to become a regional innovation system when it forms a dynamic cluster of firms learning through interaction with one another to innovate. Like Cooke and Morgan’s aforementioned study much of the RIS scholarship has been dedicated to comparisons among ostensibly economically similar regions but seeks to explain differences leading to innovation or economic performance gaps . Some of the mos relevant to this study have looked at why Europe has traditionally lagged behind the United States in innovation. For example Crescenzi et. al, found institutional and cultural barriers that prevent interactions, and thus learning, and are preventing effective market integration. But perhaps most notably Cooke , a key pioneer in the RIS field who has authored numerous comprehensive studies of Europe, identifies Europe’s over-reliance on the state for many of the functions of an innovation system which in the United States are often undertaken by private entities. He believes that incentives for innovative success are better aligned with the profit motive. By studying the functions of innovation systems it is hoped to gain an understanding of how similar results and technological innovation, can come about in vastly different institutional environments. A function can be carried out by a particular set of actors in one innovation system through a uniquely specific form, while the same function might be carried out in a different form by an entirely different actor in a similar system but in a different place or time . Management scholar Carlsson identifies three elements that all systems frameworks should address. First, it is necessary to specify the components of the system and their boundaries. Second, the relationships between the components need to be analyzed. Finally, the characteristics of the components need to be understood to determine the systems performance . The basic components are Malerba’s five building blocks discussed above: actors and networks, knowledge and technologies, and institutions . When it comes to understanding how to analyze these components several comprehensive attempts to identify measurable functions have been made . For example, in a paper titled simply “Functions in Innovation Systems,” Johnson conducted a thorough literature review to derive eight basic functions that all innovation systems share.