Industrial Ecosystems: Developing
Sustainable Industrial Structures

By Nicholas Gertler

 

Chapter 7. The Development of Industrial Ecosystems

[ to table of contents]

Information requirements 

The development of industrial ecosystems requires information above and beyond that necessary for traditional development. Industrial symbiosis requires awareness and coordination among industries, which draws on an expanded base of information, including: 

System boundaries and who is inside 

'The system' in the case of industrial symbiosis is the set of companies or other economic entities who are potential suppliers or recipients of byproduct feedstocks. In some cases, the system is defined by a property boundary, as in the case of proposed eco-industrial parks (EIPs). In an EIP the focus is primarily on the interactions among tenants, although firms outside the park may also be considered as potential symbiosis participants. In lieu of an EIP, a geographic area may delineate 'the system,' as is the case in Kalundborg. Due to the town's relative isolation, the four large industries that are located there form a natural locus of interaction, although firms outside of Kalundborg also participate in the symbiosis. 

To the extent that the size of the system under consideration determines the number of possible inter-firm linkages, a large system is likely to produce more possibilities. At the same time, however, a small and clearly defined system, such as in Kalundborg or as envisioned in an EIP, is more manageable conceptually and makes it easier to foster a sense of community. 

The system concept is of key importance because industrial symbiosis represents an effort to optimize (or at least increase) the systemic efficiency of material and energy use. This is in contrast to the more limited focus of previous approaches to environmental management on individual firms and processes. Industrial symbiosis is an implicit acknowledgment that larger efficiencies can be realized by also considering potential interactions among disparate industrial units. The notion that expanding the scope of concern to a larger set of industries can yield results that treating them separately cannot is at the core of industrial ecology. 

Material and energy flows 

It is clear that materials and energy exchanges among firms require a clear understanding of the inputs and outputs of each participant. It is less clear to what extent firms are aware of their byproduct streams. Eco-audits, which identify the environmental impact, energy use, and waste generation of a firm's activities, are well-suited for providing that information. It is only appropriate that if a firm is to redefine its byproducts from waste to resources, then it take account of those byproduct streams. 

Amount and temporal distribution 

This follows from the above. Byproduct re-use need not be all-or-nothing, meaning that another firm's byproducts need not cover the entire feedstock needs of a firm which uses those byproducts to substitute for virgin feedstocks. Re-users of byproducts can either mix byproducts with virgin feedstocks as needed or receive byproducts from multiple sources. There are several examples of this type of arrangement in Kalundborg. For example, Asnæs Power Plant substitutes Statoil flue gas for its usual coal fuel to the extent that such gas is available, while Gyproc uses Statoil flue gas exclusively when it is available, switching to a butane backup system only when necessary. 

The temporal distribution of byproduct streams is likely to vary by repetitive or irregular cycles. Seasonal fluctuations are to be expected in many cases, including beer brewing. Accordingly, the ZERI demonstration project that is looking to cluster fish farming and beer brewing is proceeding under the assumption that more beer and therefore more beer cake will be available during the summer months than the rest of the year. Symbiotic arrangements therefore need to accommodate cyclical variations and be sufficiently robust to respond to irregular fluctuations in byproduct availability and feedstock demand. 

Quality and reliability 

Byproducts can only compete with virgin materials if they are comparable in quality and reliability. By approaching byproducts as resources, industrial ecology enables companies to expand the range of their products that have economic value. Such an expansion requires a concomitant expansion in management responsibilities. Byproducts have traditionally been the uncontrolled elements of a firm's throughput [ At least until end-of-pipe treatment.] ; while attention to quality has been focused on outputs designated as products, the properties and composition of the byproduct stream have been allowed to vary [ Dan Whitney Principal Research Associate, MIT Center for Technology, Policy, and Industrial Development. Personal communication.] . The sale of byproducts as feedstocks requires quality control for what was formerly the waste stream. Such expanded responsibility can be expected to cause some loss of flexibility and to require the commitment of resources to managing the byproduct stream [ ibid.] . 

Regulatory considerations 

As discussed in the chapter on the Resource Conservation and Recovery Act, the management of solid waste is tightly regulated in cases where that waste qualifies as hazardous. Since industrial symbiosis deviates from common waste management practice, the regulatory response to it is not well established. There is reason to believe that most if not all symbiotic linkages are possible given some regulatory flexibility. Developers of industrial ecosystems are advised to work closely with regulators to ensure public-sector approval and support. The Environmental Technology Initiative's Eco-Industrial Park Project is in the process of preparing policy recommendations to facilitate eco-park development, and these recommendations will hopefully lend credibility to industrial symbiosis in the eyes of regulators. 

Three roles to keep in mind 

Three roles are critical to the development of industrial ecosystems. They are: 

• The analyst / planner; 

• The change agent; and 

• The company decision maker. 

The analyst/planner identifies opportunities for symbiotic linkages, evaluates the parameters that need to be known to gauge the viability and desirability of the potential symbiotic arrangement, and carries out the necessary design tasks. 

The change agent introduces the concepts of industrial ecology and industrial symbiosis to decision makers and champions the pursuit and implementation of symbiotic linkages. 

The decision maker is that person or persons within companies who has the authority to commit company resources to pursue symbiotic linkages and to enter into both formal (contractual) and informal relationships with other companies. 

These roles are assumed by different individuals depending on the avenue by which industrial ecosystem development proceeds. Any person may play two or even three of them at a time. The different mechanisms for development can be analyzed in large part by the allocation of these roles, and it is these various avenues to which we now turn our attention. 

The organization of industrial ecosystem development 

In the ideal form of an industrial ecosystem, all material inputs go into products and all energy consumed is used to do work. The process of approaching this ideal may be viewed as one of optimizing the systemic use of materials and energy. As this ideal is approached, the final system boundary is likely to expand toward the inclusion of all human activity; in the meantime it is the hallmark of the industrial ecology paradigm to expand the focus from single firms and processes outward. 

Optimization is a term and endeavor that has traditionally been resident to the disciplines of systems engineering and operations research. In such a context, optimization entails finding a system, establishing system boundaries, parameters, and decision variables, developing a model, and manipulating the inputs to that model to get the best output, given some measure of preference. Implicit in this construct is the existence of an analyst / planner and a decision maker. It is the analyst / planner who develops or selects the appropriate model and uses it to predict the optimal allocation of inputs. It is generally assumed that the decision maker holds the values according to which the optimization is carried out, and accordingly it is he or she who implements the results. Optimization is thus carried out by the analyst / planner to further the interests of the decision maker and his or her constituency. In a broader sense, optimization need not be a formal mathematical exercise; it may be viewed as goal-directed change based on a set of values and measures of performance. 

If the process of approaching the ideal form of an industrial ecosystem entails optimizing the flows of materials and energy within an industrial ecosystem, then someone has to be doing the optimizing according to some set of criteria, which may itself evolve over time. If the model (which encodes the system boundary) or criteria to be used are new and differ from existing practice, then a change agent is required to affect the necessary shift. The decisions to explore and commit to byproduct-as-feedstock linkages are ultimately made by corporate decision makers. The question, then, is, who is optimizing, and how are the decision criteria which are conducive to the development of industrial ecosystems felt by the decision makers?  

Three scenarios emerge from the developments detailed in this thesis: optimization from outside, optimization from within, and third-party intervention. Each is discussed below. They are points on scale that runs from a centralized command economy in which the roles of analyst / planner, decision-maker and change agent are unified to the completely autocatalytic development situation entailed by Kalundborg. 

The centralized command economy is an interesting limiting case, because it is only here that optimization of material and energy flows could be achieved in the formal sense of the word. As alluded to above, the reason for this arrangement's unique position is that the three roles are unified, such that there is but one coherent set of goals [ At least in theory; the history of command economies speaks against this assertion.] which are pursued by a unified decision-maker with control over the industrial system. This scenario sets one endpoint on the spectrum, which is most closely (or rather least distantly) approached by an eco-industrial park. 

Optimization from outside 

The Environmental Technology Initiative's Eco-Industrial Park Project sets the system boundaries for an EIP around an industrial park with a common management authority. This authority is entrusted, among other things, with maintaining 'the right mix' of companies needed to best use each other's byproducts as feedstocks and with encouraging and facilitating such exchanges [ Fieldbook on the Development of Eco-Industrial Parks Indigo Development, Draft April 25, 1995.] . In this case, it is assumed that the park's management authority internalizes the goal of optimizing the utilization of materials and energy by park tenants [ This may or may not be a reasonable assumption. Such an EIP management entity has never existed, (as far as I can tell), and is not assured of being brought forth effectively by fiat. Park developers, like other businesses, are interested in return on investment, so that optimizing the utilization of materials and energy by tenants has to contribute to that end in some fashion for it to be maintained as a stable, long-term goal.] . It is park management that acts as the optimizer, and it is meta to the system. While park tenants see bilateral exchanges that improve both their environmental and economic performance, park management sees an industrial ecosystem with resulting loop-closing benefits. The implied goal of the park's management is to bring about such an ecosystem. 

This arrangement raises a number of questions, like, how is it in the business interest of the park authority to optimize for materials and energy utilization? How is it in the business interest of each of the tenants to be optimized? Would they go along? And, what factors other than direct economic benefit would make symbiotic arrangements attractive to tenants? 

Wielding no coercive power, it is extremely unlikely that park management could compel tenants to organize into an industrial ecosystem if that is not economically beneficial to them. However, the management authority has the systems view and system goals, and can play the role of the analyst / planner. Park management can thus provide joint provision of services, information linkages, and facilitation of byproduct reuse arrangements, as well as providing the institutional context for interaction among park tenants. Decision remains with the tenants, so that the role of the park management as optimizer is to create conditions such that tenants, by acting in their own best interest, further the goals of the system [ As defined by the park authority.] . 

Optimization from within 

More general is the situation without meta-management in the form of a park authority or similar entity. Then each element of the system needs to act individually, or in cooperative arrangements, so that the system as a whole is optimized. That's optimization from within, also referred to here as autocatalytic or self-organizing development. 

Take entropy to be the degree of disorder in a system. Safe to say that industrial symbiosis reduces entropy over no symbiosis because it reduces dissipation. For example, by definition energy cascades increase the proportion of energy consumed that is available to do work. And that ties into one of the classic definitions of entropy. Next, conditioning and reuse of byproducts gives more of everything in economically useful concentrations and mixtures than their conventional fate as waste. 

So industrial symbiosis reduces the production of entropy, raising the prospect of evolutionary pressure toward the development of industrial ecosystems. As Prigogine has written: 

We can consider the evolution of living organisms up to the stationary state as taking place under a certain number of constraints determined by the outside world...Whatever the nature of the [constraints], the stationary state may probably to a good approximation be considered as a state of minimum production of entropy per unit time. This description fits in excellently with some striking characteristics of living organisms. First, the well known stability against external perturbation has its analogue in the stability of stationary states corresponding to a minimum production of entropy. Further, the fact that during growth living organisms actually experience a decrease of entropy production during evolution to a stationary state. Also, the fact that their organization generally increases during this evolution corresponds to the decrease of entropy... [ I. Prigogine, Thermodynamics of Irreversible Processes C.C. Thomas, Springfield IL 1955 p. 91 ] 

Evolution results in decreasing entropy production. It is not much of a rhetorical leap to argue that, as follows from the concept of evolution, those organisms, or rather those interdependencies of organisms, survive in a given niche which operate with the least production of entropy. This is because, by definition, the less entropy, the more efficiently organism/ecosystems function in a closed system. Those that are more efficient crowd out those that are less efficient, to get the lowest overall production of entropy; they have the most survival value. The others die off. And in that way is the system of organisms optimized from within. 

Now to apply this line of reasoning to industrial ecosystems. We claim that the optimization of material and energy flows results in more sustainable industrial activity. (Optimization may be too strong a requirement; greatly increased efficiency should suffice.) This optimization is directly coupled to a state of minimum entropy production. In ecosystems, inefficient organisms and linkages produce more than the minimum possible amount of entropy and are crowded out by more efficient arrangements producing less entropy. This would then be an implicit decision rule that drives ecosystem evolution. Its influence is exerted on ecosystem participants in the form of results; those participants that are more eco-efficient than others survive to continue evolving. Those that are not, die off. In this way, the system evolves. 

For the economic system to be self-optimizing, then, decision rules need to be in place which reward increases in eco-efficiency and penalize the laggards. In the ecosystem case, failure to keep up results in the death of the particular evolutionary line. The direct analogy applied to the industrial system would call for corporate death or shut-down. However, participants in this system are more adaptive. The possible outcomes are not limited to death versus survival and procreation; mid-course adjustments are possible. But if industrial ecosystems are to be self-optimizing in that they entail an evolution toward the minimization of entropy, then the decision rules applied to the system participants, be the rules implicit or explicit, must provide the same signals as they do in natural ecosystems: efficient arrangements thrive at the expense of inefficient ones. 

The decision rules currently applied to the industrial system are in the form of the price system and regulation. If this line of thought is accepted, then these signals need to be tailored by public policy to transmit the correct evolutionary pressures. The concept of economic efficiency needs to be expanded to include eco-efficiency; clearly this entails some form of 'getting the prices right'. In addition, public policy needs to allow for and support the creation and maintenance of industrial 'niches' or arrangements that minimize entropy. 

Kalundborg is the prime example of this sort of evolutionary development. The definition of industrial symbiosis comes from Valdemar Christensen of Asnæs Power Station, and includes reference to 'increased viability' among participating firms, which ties directly into the evolutionary argument developed here. In Kalundborg, the initial wave of links that made economic use of unexploited (by)products was followed by a second wave that was the outgrowth of pollution control measures and pollution control technologies. These latter linkages are more expensive than no pollution control, but represent least-cost compliance strategies. As such, they are adaptations to the changing operating environment in which the Kalundborg firms have been functioning. Regulatory and community requirements have applied evolutionary pressure, in the form of demands for reduced environmental impact. By setting performance standards, instead technology standards, public policy allowed this sort of evolution. 

All documented examples of industrial symbiosis to date have evolved in such an autocatalytic way, but this may simply be due to the fact that until recently no public or private organizations actively encouraged the development of symbiotic linkages among industries. 

Third-party intervention: the role of the broker [ This section is based in part on discussions with Ernie Lowe of Indigo Development.] 

As public and private agencies take up that cause, they should not lose sight of the importance of the institutional context and social fabric within which cooperation takes place. Emilia-Romagna, the origin of the flexible networks and inter-firm collaboration movement, echoes the lessons of Kalundborg in that regard. In both places, participants are part of a web of social interactions that support face-to-face contact and extend beyond the realm of business. 

When normal business practice does not foster interpersonal relationships among various firms, there is a niche for a broker role in bridging the gap. This has been the case in the Danish network experience, as well as in U.S. efforts to foster inter-firm collaboration. In this country, students of flexible networks have identified a lack of institutional linkages among firms. As a result, potential value-adding collaborations are not consummated due to lack of an effective venue and supportive cultural norms. The role of the broker in flexible networks, as it promises to be in industrial symbiosis, is to overcome those barriers to collaboration and to 'complete the circuit.' 

In the EIP model of industrial symbiosis, the eco-industrial park is the staging area for inter-firm linkages, and park management is entrusted with playing an organizing and catalyzing role. The analyst / planner function is hard-wired into the EIP; park management is clearly identified as meta to the industrial subsystem of the park. The autocatalytic or evolutionary model of industrial symbiosis development is at the other end of the spectrum. Here participating businesses individually feel evolutionary pressure toward the formation of an industrial ecosystem and enter into bilateral arrangements. The system is self-organizing, but such self-organization relies on a institutional connections among firms that are largely absent in the U.S. 

The resulting void creates the niche for a broker role. Though not structurally 'above' the system he or she acts on, a broker can have the systems perspective that individual firms lack. The networking experience in both Europe and the U.S. indicates that brokers can effectively catalyze mutually beneficial collaborations that would not come about otherwise. By helping companies perceive their common opportunities in byproduct-to-feedstock arrangements (and in possibly others), and in championing the formation of such linkages, the broker furthers the system goal of increased material and energy utilization while furthering the interests of participating firms. The broker thus plays the role of both analyst / planner and change agent. 

For such a role to be viable, the broker must have an economic self-interest in bringing about symbiotic linkages. Since it can safely be assumed that symbiotic arrangements will come about only when they are financially beneficial or least-cost, the 'value added' by the broker can serve as the basis for compensation. Whether there is enough money to be made in playing such a role is an open question, but experience indicates that at least one firm has successfully achieved that result (see box next page). 

A different allocation of roles is present in the case of the Zero Emissions Research Initiative. Here the focus is on designing and co-locating zero emissions, symbiotic industry clusters. Because this arrangement can be expected to involve a smaller number of participants than an EIP and because the focus of development is cluster-, not site-specific, a tighter coupling can be expected between the analysis and planning roles and those of the decision maker. ZERI represents an entrepreneurial approach whose realization results in environmental benefits, which in turn aid in achieving entrepreneurial results. 

VP Resources: "Finding a home for orphan chemicals" 

More than a waste exchange, VP Resources of Clearlake, Texas, finds a home for orphan chemicals and does everything necessary to ensure their adoption. Proprietor Vance Purcell used to be an executive in the chemical industry before he found a niche in developing secondary markets for chemical byproducts - something which chemical companies have largely failed to do. Byproducts were burned or deepwelled before environmental pressures increased the cost of disposal to where alternative arrangements became attractive. 

A variety of waste exchanges have been attempted in recent years but with only limited success. While such endeavors merely list available byproducts, VP Rescues performs every function required to turn a byproduct into a feedstock, including finding appropriate uses, dealing with regulatory agencies, brokering necessary agreements, and even transporting the materials from the generator to the user. 

What sets VP Resources apart is its full-service, cradle-to-reuse approach to secondary markets for byproducts. Relying on industry knowledge and personal contacts, VP Resources has succeeded where others have failed. 

Source: Vance Purcell, Personal Communication. VP Resources: (713) 488-3496 

The above classifications are regions along a continuum, not distinct and separate arrangements. The table on the following page summarizes some relevant attributes of various approaches to industrial ecosystem development, listed in order of decreasing centralization. 

The venue: green and brown, new and pre-existing developments 

The main distinction between greenfields and brownfields is that greenfields are undeveloped areas (hence the color) while brownfields are existing developments. From the perspective of industrial symbiosis, brownfields can be further subdivided into sites currently occupied by industry and those that have been abandoned. Abandoned sites of former 
 

Organizational			Role
Form	System bounds	Decisionmaker	Analyst/Plan.	Change agent
Command  Economy	Up to entire  economy	Central planner	Central planner	Central planner
ZERI	A cluster of  industries	Businesses,  entrepreneur	Entrepreneur /  researcher	Entrepreneur /  researcher
EIP	Industrial park	Tenants, park  management	Park management	Park management,  community
Brokered  Network	Diffuse set of  industries	Businesses, broker	Broker	Broker
Autocatalytic  Network	Diffuse set of  industries	Businesses	Self-organizing	Self-organizing

Development of Industrial Ecosystems by Organizational Type  

industrial activity are just as brown, if not more so, as sites of current industrial use, due to likely contamination. From an environmental point of view, there is compelling reason to re-develop brownfields in lieu of converting greenfields. The reasoning is simple: why degrade a pristine (or at least undeveloped) ecosystem when an already disturbed one is available? This is a choice with regard to EIP development in Chattanooga, and the specter of liability for past contamination is the single greatest obstacle to re-development of the brownfield Central Business District. 

From the perspective of industrial symbiosis development, the main distinction is between new and pre-existing developments, because the context in which symbiosis development proceeds depends in part on whether the industrial facilities are designed and built with symbiosis in mind or are retrofitted. For pre-existing developments, the standard of comparison, or baseline situation to contrast with symbiotic arrangements, is business as usual. The necessary condition for establishing symbiotic linkages with other firms is reasonably straight-forward: go ahead with retrofits if the linkages are feasible and financially beneficial. 

New developments provide an additional dimension, in terms of the extent to which this ecopark or ecosystem prospect influences the decision of firms to move to the area and establish operations. If a number of firms are set to locate in somewhat of a cluster, without any consideration of possible symbiotic arrangements, then introducing the concept to them at some early part of the development process (by some agent of action) creates the potential to choose process technologies and develop operations in some sort of cooperation to enable symbiotic linkages. This is probably the most attractive situation for symbiosis development, assuming the companies are there, are compatible, cooperative, and interested. In the limit, new developments provide the opportunity to aim for 'zero emissions' along the lines envisioned by the Zero Emissions Research Initiative. Such development is in the planning stages for the downtown Chattanooga EIP proposal. 

There are of course variations on this situation; for example one or more new companies may be moving into an established industrial area and could explore prospects for symbiotic arrangements. The earlier in the design and process selection process that symbiotic possibilities are considered, the better. But this is not to say that these possibilities will be considered when companies establish operations, only that this would be a good time for them to be. 

So if a greenfield is turning brown of its own accord, or a brownfield is being redeveloped, then the IS concept could and should be introduced (by some agent of action / analyst / planner) since early consideration provides the most flexibility in developing linkages. This is not really in conflict with the evolutionary nature of symbiosis development in Kalundborg, because many of the links were developed in a microcosm of the 'new development' situation, as firms looked for symbiotic arrangements when they had to change processes anyway. Thus the flexibility advantage of greenfields can be captured in smaller parts as brownfield participants change over time. 

The second greenfield/new development alternative, which is implicit in the ecopark concept being developed as the PCSD/ETI/Eco-Industrial Park project, is that companies will locate at a certain park or area specifically because it is an ecopark and offers advantages of symbiotic arrangements. In situations involving pre-existing development, symbiotic links tend to be retrofits or least-cost ways of achieving necessary changes, and therefore they have to be better than 'business as usual' ways of dealing with the change. That is, if at some point my firm has to do X, and doing X by way of a symbiotic link is feasible and cheaper than doing X in some traditional way, then that condition should be enough to give rise to a symbiotic link. 

If the draw of a new development is the ecopark concept, then the necessary condition to be met is likely to be more rigorous. Now the alternative to symbiosis is not business as usual in that given cluster, but rather a range of possibilities including location elsewhere or no new facility. The question in this greenfield situation is whether symbiosis and the other benefits offered by the EIP are enough to attract businesses when the standard of comparison is more broad in scope. In a pre-existing development, any net savings warrants a link, but a more complete evaluation of the economics of symbiosis seems necessary to attract firms to a new ecopark. The ability of EIPs to draw tenants is still very much in question. 

The Chattanooga and Baltimore EIP projects reveal differing approaches to symbiosis development based on their new or pre-existing nature. For the pre-developed Baltimore site, Cohen-Rosenthal and colleagues are proposing ambitious goals for continuous improvement rather than zero emissions. The goal of such an approach is to reclaim dirty areas and make current locations survive and work. In contrast, the Downtown Business District EIP in Chattanooga calls for new development of an existing site, with a goal of zero emissions. The Gunter Pauli / ZERI approach entails designing processes for symbiosis (what they call zero emissions). Such an undertaking can only occur before the industrial structures in question are built or are about to be radically altered. As such, zero emissions and continuous improvement are complementary approaches, paired to new and pre-existing development, respectively (see table). 
 

	New Development	Pre-existing Development
Ecopark	Zero-Emissions possible;  Concurrent genesis;  'Meta' analyst	Continuous Improvement;  Brokered or 'Meta' analyst
'Diffuse' Ecosystem	Zero-Emissions(?) or Evolutionary;  Brokered or Autocatalytic	Evolutionary;  Continuous improvement in result but not express goal;  Brokered or Autocatalytic

Relationship of development venue to development organization 

Factors influencing industrial ecosystem development 

Experience and reflection have identified a number of factors which affect prospects for the development of industrial ecosystems. They include: 

Economic viability 

As a general rule, symbiotic linkages are unlikely to be developed and sustained unless they make money or cost less than more traditional arrangements of materials and energy use. Note that regulation and public policy can alter the costs and benefits felt by companies, and thereby influence the attractiveness of symbiotic linkages. 

Public policy as evolutionary pressure 

Regardless of any stories one can tell of closing loops and emulating ecosystems, businesses will only engage in symbiotic linkages if and when such arrangements are economically beneficial. Every single linkage in Kalundborg, the model of industrial symbiosis, is either a sale of a byproduct or a least-cost way of complying with environmental regulations. Therefore, the single best way to foster the development of industrial ecosystems is to make them economically attractive. 

It is implicit in this argument that at any given time there are more potential byproduct-to-feedstock linkages available than are economically viable. The initial wave of links in Kalundborg derived value by selling byproducts which previously had been wasted. Later links developed only in response to stricter environmental regulations. These strategies, such as the use of Novo Nordisk's process sludge as fertilizer, were least cost ways of complying with regulation. In effect, as regulation forced companies in Kalundborg to internalize the environmental cost of their production, symbiotic links became attractive. 

Imposing regulatory requirements and altering the price signals felt by industry are two ways in which public policy can apply pressure to encourage the evolution of industrial ecosystems. This strategy entails a form of 'getting the prices right.' Such external signals are not sufficient, however, since innovative and pioneering cooperation is required among companies for symbiosis to occur. Yet such cooperation is only viable if it makes economic, not just environmental, sense. 

The evolutionary view of industrial ecosystem development would therefore hold that there is a natural progression towards symbiosis if the government response to the environmental impacts of industrial activity sends the right signals and if the regulatory structure is conducive (see below). Of course symbiosis does not necessarily follow from a regulatory push for pollution control; there has to be some propensity. After all, industries in places other than Kalundborg have been forced to decrease pollution, without the effect of inspiring symbiosis. However, it would appear that government involvement is necessary to ‘level the playing field’ in favor of loop-closing arrangements by raising the relative cost of conventional practice (or lowering the cost of symbiosis). It is not, however, sufficient. Symbiosis requires information, cooperation, and creativity, and these requirements are more difficult to supply by public policy intervention. 

Institutional linkages and personal relationships among firms 

Both the Kalundborg case study and experience with inter-firm collaboration and flexible networks indicate very strongly the importance of personal contact among firms as a starting point for collaborative arrangements. Kalundborg is a small society in which managers of the various firms often run into each other, providing opportunities for face-to-face contact and informal discussion. The fact that the four firms are planted in the same interconnected society in which their employees live makes inter-firm cooperation more readily achievable. 

Similarly, students of inter-firm collaboration have concluded that trust and communication are essential building blocks of networking activity. Given the necessary reliance on network partners, owners and managers who do not know and trust each other are usually unwilling to put their businesses at risk by entering into collaborative arrangements. Given the importance of personal relationships, lack of institutional linkages among firms is a major impediment to inter-firm collaboration, a result that carries over to industrial symbiosis. 

Brokers who are credible in the eyes of all participants have been successful in bridging that gap in their bid to establish inter-firm collaboration, and the broker role also holds promise in the development of industrial ecosystems. Brokers are most appropriate in fostering linkages among industries in areas not bound together by existing ties. Other organizational forms may require less assistance from brokers. By focusing on environmental performance and entrusting park management with an expanded role, the eco-industrial park concept may be able to provide the institutional context necessary for fostering symbiosis. The more centralized approach entailed by the Zero Emission Research Initiative effectively bypasses the need for a broker by focusing on designing and co-locating specific symbiotic industry clusters. 

Personal relationships are also important in that the personalities of those involved with creating symbiotic linkages have a strong influence over the form of the outcomes produced. This effect is illustrated in the difference between the directions of the Chattanooga and Baltimore Eco-Industrial Park demonstration projects. The entrepreneurial approach of Gunter Pauli and his Zero Emissions Research Initiative in Chattanooga is contrasted by the more labor-oriented practices at the Baltimore E.I.P., whose development is headed by Ed Cohen-Rosenthal of Cornell's Work and Environment Initiative. 

A number of approaches avail themselves for overcoming the general lack of institutional linkages among firms in the United States. By and large they are in their nascence and their effectiveness remains to be seen. Initiatives currently underway hold the promise of creating elements of industrial ecosystems. However, widespread creation of symbiotic inter-firm linkages will most likely require a broader shift in business culture, one that is conducive not only to the exchange and reuse of material and energy flows, but to increased collaboration in general. In Kalundborg, the symbiotic linkages among local firms have expanded to other areas of collaboration, such as worker training and safety. In turn, experience with flexible networks and inter-firm collaboration indicates that collaboration in one area fosters collaboration in others. Thus, one way to promote industrial symbiosis is to strengthen inter-firm linkages in general. 

Awareness and Interest 

Symbiotic linkages will not come about unless decisionmakers within companies begin to see byproducts as resources, not wastes. While all businesses market their products, few are versed at marketing their byproducts. Finding possibilities for symbiotic linkages with other firms is new to the realm of business practice. Initiatives such as the Eco-Industrial Park project are therefore valuable not just for the tangible results they may produce, but also as vehicles for bringing industrial ecology into the mainstream. To the extent that industrial ecology is a lens through which to see material and energy flows, broader awareness of IE can only help in bringing about industrial ecosystems. 

Regulatory reform and clarification 

Two regulatory issues are salient with respect to the development of industrial ecosystems: clarification and increased flexibility of regulation under the Resource Conservation and Recovery Act (RCRA), and a preference for performance standards over technology standards for pollution control. 

As discussed in a prior chapter, regulation of the management of waste under RCRA has created unintended barriers to the reuse of byproducts as feedstocks. While only about one-third of the industrial byproducts generated in the United States fall under the 'hazardous' label, uncertainty over the classification of particular waste streams and the significant requirements of hazardous waste management are a daunting force against byproduct reuse. EPA policy and regulations can be adjusted to reduce the burden on symbiotic arrangements while maintaining the required level of safety. Given the history of the RCRA program, any reform should begin with EPA headquarters issuing a policy statement in support of the legitimate reuse of industrial byproducts as feedstocks. EPA regions and states should then be allowed greater discretion in approving reuse/recycling activities, on a case-by-case basis and according to clear guidelines. 

A more subtle point is the preference for performance standards over technology standards in pollution control regulation. As noted before, the second wave of symbiotic linkages in Kalundborg came about in response to regulatory pressure to decrease pollution. The stricter environmental regulations which have in large measure been the driving force for the more recent linkages have been performance standards, not technology standards. 

This distinction is significant because it has allowed the local firms to choose pollution control technologies which rendered their waste streams usable as feedstocks elsewhere, yielding an additional benefit beyond pollution reduction. For example, from among several alternatives, Asnæs Power Station chose a scrubber technology which resulted in the production of gypsum as its byproduct. This gypsum is sold to Gyproc, reducing Asnæs’ cost of pollution control and replacing a virgin feedstock. Flexibility in meeting environmental standards is an important requirement for the development of industrial ecosystems, and this issue is on the agenda of the Eco-Industrial Park project. That undertaking will result in recommendations to greater flexibility, which, given the current political climate, will likely be acted upon. 

The systems view and place-based development 

From command-and-control regulation to pollution prevention, approaches to environmental problems that preceded industrial ecology have focused on individual plants and processes as the units of interest. Industrial ecology calls for a broadening of focus to encompass the system in which production takes place. Closing loops and increasing the efficiency of material and energy use requires a holistic view of firms as part of their surroundings, both natural and human-built. Place-based development is being advanced through the Eco-Industrial Park project and similar undertakings as a way to make use of local features and conditions and to better integrate industrial activity into both the natural and the man-made environment. This approach requires a heightened awareness and sensitivity to the interactions of a given process with its surroundings. From such awareness can spring novel synergies, such as finding valuable uses for byproducts. 

...and some concluding thoughts 

In some ways, this time, Spring of 1995, is the end of the dawn of a new field of inquiry, one that seeks to harmonize the workings of human economic activity with the natural systems that sustain it. Interest in industrial ecology has been steadily increasing since the notion was introduced in 1989-1990. The Clinton Administration's National Environmental Technology Strategy, Bridge to a Sustainable Future, identifies IE as the dominant paradigm for sustainable industrial development as we move into the twenty-first century [ National Science and Technology Council, Bridge to a Sustainable Future National Environmental Technology Strategy . April 1995.] . Other evidence of the ascendancy of industrial ecology confirms that IE, at least as a phrase, has moved into the mainstream and is unlikely to go away any time soon. 

Interest in what has been referred to here as industrial symbiosis has taken longer to develop. The first article about Kalundborg that concerted effort could turn up was "A rebirth of the pioneering spirit," published in November of 1990 in Britain's Financial Times [ Peter Knight "A rebirth of the pioneering spirit" Financial Times Wednesday, November 14, 1990. p. 15.] . A two-page description of the Kalundborg industrial ecosystem in Hardin Tibbs' 1993 novella Industrial Ecology An Environmental Agenda for Industry [ Hardin Tibbs, Industrial Ecology An Environmental Agenda for Industry Global Business Network 1993. An earlier version of this paper was published in 1991 by Arthur D. Little.] firmly entrenched the town in the mythology of this developing field. Holger Engberg, a Danish Professor of Finance and International Business at NYU's Stern School of Business followed in October of 1993 with an extensive descriptive case study entitled "Industrial Symbiosis in Denmark." Though it provides an excellent background on Kalundborg, this paper was never published (as far as I can tell) and was not widely circulated. In the realm of industrial ecology, Kalundborg was an often-referred-to, mythical place which inspired the imagination of many but about which little was known. Such was the backdrop to my own visit in July of 1994. 

The Danes' hospitality was moving; Asnæs Power Plant housed me in their little straw-roofed guest house 100 yards from one of the massive boiler buildings. The kitchen was stocked with food and coffee and I was provided with sustainable transportation: a bicycle. Five days of riding that bicycle to interviews and factory tours resulted in a case study detailing the development of the industrial ecosystem that comprises Chapter 2. of this document. Circulated since August of 1994 as a working paper of the MIT Technology, Business, and Environment Program, which has sponsored this research, the case study was well-received, indicating that interest in Kalundborg had far outstripped available information. 

The notion of industrial symbiosis is now being advanced by, among others, the Zero Emissions Research Initiative (though not in so many words) and by the Eco-Industrial Park project. Spontaneous efforts to create that sort of development have been registered here at MIT and by others in various parts of the country. While far from being well-established, interest in industrial ecosystems is slowly on the increase. 

In many ways, however, current interest in industrial ecology and industrial ecosystems is neither new nor novel. A journal called, curiously, Industrial Ecology was first published in the winter of 1970. It did not last long, but its discovery recently led industrial ecologist Brad Allenby to remark that there is, in fact, nothing new under the sun [ Brad Allenby, Personal communication with John Ehrenfeld, Spring 1995.] . Similarly, a study entitled "Systems-Integration Requirements for the Synergistic Co-Siting of Industrial Activities" was concluded in 1980 by researchers at the Georgia Institute of Technology [ The full cite is Spurlock, J. and Ward H. 1980 "Systems-Integration Requirements for the Synergistic Co-Siting of Industrial Activities" U.S. Department of Commerce, National Technical Information Service publication PB81-15029. It is mentioned in Fieldbook on the Development of Eco-Industrial Parks by Indigo Development.] . Those previous efforts have faded, but the current period is one of great promise. As one artifact of this renaissance, The Journal of Industrial Ecology, unrelated to its namesake and predecessor, is about to be launched. 

The notion of industrial ecosystems offers a very powerful vision to guide the development of sustainable industrial structures. This research has sought to identify the characteristics of such structures as well as the factors and conditions that effect their development. While public-sector interventions have a significant role to play in establishing the appropriate evolutionary signals, bringing industrial ecosystems into being will ultimately depend on pairs or groups of businesses that find novel, collaborative ways to make use of byproducts and meet environmental demands. For industrial ecosystems to make a significant mark on the business and environment landscape, groups of firms must independently adopt symbiotic linkages, like the "thousand points of light" in that bygone analogy. This result requires that individuals in businesses become aware of the latent possibilities provided by the structures around them. It also requires a regulatory climate and business culture that encourages such development. 

Successful applications of industrial symbiosis should be publicized to encourage others to try. While Kalundborg is the most often-cited example and the one receiving the most attention here, industrial symbiosis can be found on a smaller scale in many parts of the U.S. Many of these examples are being collected in the Fieldbook on the Development of Eco-Industrial Parks, to which the reader is referred for details. Perhaps these accounts foretell of a shift to a new form of development, one in which human activity is harmonized with the workings of natural ecosystems. 

[ to table of contents]