“uncovering” industrial symbiosis · special feature on industrial symbiosis “uncovering”...

SPECIAL FEATURE ON INDUSTRIAL SYMBIOSIS

“Uncovering” IndustrialSymbiosisMarian R. Chertow

Summary

Since 1989, efforts to understand the nature of interfirm re-source sharing in the form of industrial symbiosis and to repli-cate in a deliberate way what was largely self-organizing inKalundborg, Denmark have followed many paths, some withmuch success and some with very little. This article providesa historical view of the motivations and means for pursuingindustrial symbiosis—defined to include physical exchanges ofmaterials, energy, water, and by-products among diversifiedclusters of firms. It finds that “uncovering” existing symbioseshas led to more sustainable industrial development than at-tempts to design and build eco-industrial parks incorporatingphysical exchanges.

By examining 15 proposed projects brought to nationaland international attention by the U.S. President’s Council onSustainable Development beginning in the early 1990s, andcontrasting these with another 12 projects observed to sharemore elements of self-organization, recommendations are of-fered to stimulate the identification and uncovering of alreadyexisting “kernels” of symbiosis. In addition, policies and prac-tices are suggested to identify early-stage precursors of po-tentially larger symbioses that can be nurtured and developedfurther. The article concludes that environmentally and eco-nomically desirable symbiotic exchanges are all around us andnow we must shift our gaze to find and foster them.

Keywords

by-product synergyeco-industrial developmenteco-industrial parksindustrial ecologyindustrial ecosystemsindustrial symbiosis

Address correspondence to:Professor Marian R. ChertowSchool of Forestry & Environmental

StudiesYale University205 Prospect StreetNew Haven, CT 06511-2189 USA<[email protected]><http://environment.yale.edu/profile/247/marian chertow>

© 2007 by the Massachusetts Institute ofTechnology and Yale University

Volume 11, Number 1

www.mitpressjournals.org/jie Journal of Industrial Ecology 11

F O RU M

Introduction

Looking back, 1989 was an inspirational yearfor industry and environment. Following theBruntland Commission report in 1987 came twokey events. The seminal article by Frosch andGallopoulos (Frosch and Gallopoulos 1989) inScientific American envisioned “industrial ecosys-tems” in which “the consumption of energy andmaterials is optimized and the effluents of oneprocess . . . serve as the raw material for anotherprocess.” That same year, a cluster of companiesfrom different industries that were intensivelysharing resources was unexpectedly uncovered inDenmark and then described in the internationalpress (Knight 1990; Barnes 1992). What we havecome to call “the industrial symbiosis at Kalund-borg” provided a concrete realization of the in-dustrial ecosystems Frosch and Gallopoulos the-orized.1

Industrial symbiosis has been defined as engag-ing “traditionally separate industries in a collec-tive approach to competitive advantage involv-ing physical exchange of materials, energy, water,and by-products. The keys to industrial symbio-sis are collaboration and the synergistic possi-bilities offered by geographic proximity” (Cher-tow 2000). Since 1989, efforts to understandthe nature of symbiosis and to replicate in adeliberate way what was largely self-organizingin Kalundborg have followed many paths,some with much success and some with verylittle.

This article provides a historical view of themotivations and means for pursuing symbio-sis, with a focus on assessing planned versusspontaneous symbiosis. By examining 15 pro-posed projects brought to national and interna-tional attention by the U.S. President’s Councilon Sustainable Development beginning in theearly 1990s, and contrasting these with another12 projects observed to share more elements ofself-organization, recommendations are offered tostimulate the identification and uncovering of al-ready existing symbioses. In addition, policies andpractices for governments, nongovernmental or-ganizations (NGOs), and businesses are suggestedto identify early stage “precursors” or “kernels” ofsymbiosis that can be nurtured and developedfurther.

Pursuing Industrial Symbiosis

Although industrial symbiosis may appear tohave leapt fully grown onto the sustainabilitystage, notions of trading and resource exchangeare as ancient as primitive peoples sharing animalparts. Scrap dealers, charities organizing clothingdrives, and companies buying and selling residualmaterials on line are engaging in resource ex-change. To distinguish industrial symbiosis fromother types of exchanges, my colleagues and Ihave adopted a “3–2 heuristic” as a minimumcriterion. Thus, at least three different entitiesmust be involved in exchanging at least two dif-ferent resources to be counted as a basic type ofindustrial symbiosis. By involving three entities,none of which is primarily engaged in a recycling-oriented business, the 3–2 heuristic begins to rec-ognize complex relationships rather than linearone-way exchanges. A simple version of this isa wastewater treatment plant providing coolingwater for a power station and the power station,in turn, supplying steam to an industrial user.The words “kernel” and “precursor” have beenchosen to describe instances of bilateral or mul-tilateral exchange of these types that have thepotential to expand, but do not yet meet thefuller 3–2 definition of industrial symbiosis. Seefigure 1.

The symbiotic relationships described aboveare presumed to provide environmental bene-fits, although these benefits have seldom beencarefully measured. They occur in single-industrydominated clusters such as petrochemical com-plexes as well as multi-industry ones such asKalundborg. In general, three primary oppor-tunities for resource exchange are considered(Chertow et al. 2007): (1) By-product reuse—the exchange of firm-specific materials betweentwo or more parties for use as substitutes forcommercial products or raw materials. (2) Util-ity/infrastructure sharing—the pooled use andmanagement of commonly used resources suchas energy, water, and wastewater. (3) Joint provi-sion of services—meeting common needs acrossfirms for ancillary activities such as fire suppres-sion, transportation, and food provision.

Many motivations exist for pursuing industrialsymbiosis, either directly or indirectly as a resultof trying to meet other objectives. The most

12 Journal of Industrial Ecology

S P E C I A L F E AT U R E O N I N D U S T R I A L S Y M B I O S I S

Figure 1 Example of 3–2 symbiosisinvolving a minimum of three differententities exchanging at least two differentresources.

obvious motivations are conventional businessreasons; for example, resource sharing can reducecosts and/or increase revenues. At another level,industrial symbiosis can enhance long-term re-source security by increasing the availability ofcritical resources such as water, energy, or par-ticular raw materials through contracts. In somecases, companies pursue symbiosis in response toregulatory or permitting pressure requiring indus-trial operators to increase efficiency of resourceuse, reduce emissions, or eliminate waste.

Historically, what is often described as “spon-taneous co-location” of businesses in industrialdistricts has been shown to give rise to many pub-lic and private benefits including labor availabil-ity, access to capital, technological innovation,and infrastructure efficiency (Marshall 1890;Krugman 1991; Desrochers 2001; Duranton andPuga 2003). Still, the modern literature on these“agglomeration economies” generally overlooksenvironmental benefits of agglomeration throughresource sharing (Chertow et al. 2006). Althoughself-organizing industrial districts have been ob-served to produce many advantages over the lasthundred years, people involved in planning eco-industrial parks (EIPs) and other concrete mani-festations of industrial symbiosis have anticipatedmany other types of benefits as key rationales foradvancing projects, including economic devel-opment broadly, remediation of pollution asso-ciated with heavy industry, water and land sav-ings, and greenhouse gas reductions, as discussedbelow.

Because industrial development is a form ofeconomic development, there has been interestin using the concept of industrial symbiosis inthe form of eco-industrial parks (EIPs) to (1) re-vitalize urban and rural sites, including brown-field redevelopment, (2) promote job growthand retention, and (3) encourage more sustain-able development. An eco-industrial park project

was proposed as a way to attract jobs through“clean” economic development in a poor ruralU.S. county in Virginia (Hayes 2003). In China,a sugar refining company grew by following thepath of the supply chain for that industry, therebyproviding solutions for two important problems:decreasing the environmental impacts of sugar re-fining through reuse of by-products and increas-ing employment by using these by-products tofuel new enterprises (Zhu and Cote 2004; Zhuet al. 2007). Figure 2 shows the expansion of theGuitang Group beyond sugar refining into relatedindustries that use materials from the two keyby-product streams from sugar cane production:molasses (the sugar refining residue) and bagasse(the fibrous waste product).

Business development strategies involvingindustrial symbiosis are being used in somecountries to alleviate environmental degrada-tion where contamination has already occurredat the “end of the pipe.” Van Berkel (2004),a leading industrial symbiosis researcher fromAustralia, describes the use of resource exchangeas an approach to addressing the high volume ofwastes from mineral, metal, and energy produc-tion in Australia. One well-studied example is theKwinana Industrial Area in Western Australia,operating since the 1950s, composed of numer-ous mineral processing industries. Between 1990and 2001, the number of core process industries atKwinana increased from 13 to 21 and the numberof symbiotic exchanges increased from 27 to 106.Of these, 68 involved core process industries and38 involved services and infrastructure (Althamand Van Berkel 2004, Van Beers et al. 2007).

The differential distribution of rain, surfacewater, and groundwater has led many commu-nities to ask whether there could be more sus-tainable ways to use water. The symbiosis inKalundborg, Denmark began because of the lowavailability of groundwater and the need for a

Chertow, “Uncovering” Industrial Symbiosis 13

F O RU M

Figure 2 The Guitang Group, beyond sugar refining. Source: Zhu and Cote 2004, 1028. Used bypermission from authors.

surface water source which, once identified, be-came a key part of the resource exchange networkthere (Christensen 1998). Several AES powerplants in the United States associated with eco-industrial development in areas concerned withwater availability in Hawaii, Puerto Rico, andNew Hampshire use millions of gallons of sewagewater per day for cooling water.

In many parts of the world, land that was for-merly arable has, over time, degraded with theintrusion of salt water. In Narrogin, Australia,farmers are now planting mallee trees that soakup groundwater with their deep roots, helping toprevent the rise of the saline water table (Barton1999). The opportunity provided by the kernel oftree planting to build industrial symbiosis aroundit has resulted in construction of a pilot plantdesigned to process the harvested shrubs to pro-duce one megawatt (1 MW)2 of renewable en-ergy, 700 tonnes3 per year of eucalyptus oil, and200 tonnes per year of charcoal as a possible sub-stitute for black coal in high-temperature met-allurgical reactors (Department of the Environ-ment and Heritage 2004; Enecon 2006).

Concentrations of industry are often heavygenerators of greenhouse gases associated withglobal climate change. Discussions in Alberta,Canada’s industrial heartland, raised the ideaof establishing tree plantations near the oiland petrochemical industries located there, notonly to compensate for CO2 emissions, but alsoto provide employment and renewable materi-als (Alberta’s Industrial Heartland Association2000; Cote 2003). A heavy industry area ineastern Australia focused on minerals process-ing and chemical production makes a sizeablecontribution to greenhouse gas emissions be-cause of raw material consumption, transport,and waste disposal (Sustainable Gladstone 2005).The Australian government is examining the ex-tent to which by-product reuse and exchange canresult in reduction of greenhouse gases in theGladstone Industrial Area.

So far, we have recounted business benefitsof symbiosis across companies, as well as exper-iments and opportunities for resource conser-vation. Symbiotic activity can also come intoplay in response to a regulatory situation. In



the United States, for example, the Public Utili-ties Regulatory Policy Act (PURPA) bestowedcertain pricing benefits on facilities willing toco-generate steam and electricity, which stim-ulated many industrial plants to become “quali-fying facilities” under the law.4 Rules regardingscrubbing of flue gases in power-generating facil-ities in several countries have led to many sym-bioses involving the use of scrubbed sulfur forproduction of gypsum wallboard. Even when alaw does not directly command reuse, a less tan-gible “license to operate” in the form of regula-tory permits or greater social acceptance can beassociated with projects showing environmentalbenefits (Gunningham et al. 2003; Chertow andLombardi 2005).

Thus, there are many reasons for pursuing in-dustrial symbiosis, beginning with the most basicdesire of businesses to be profitable and com-petitive. Important social, environmental, andregulatory drivers also exist, which play out dif-ferently in different parts of the world, as illus-trated by the examples above. At the same time,although there are numerous measurable bene-fits, the classic expression—“if (the subject un-der discussion) is so advantageous, why aren’t weseeing a lot more of it?”—is applicable to indus-trial symbiosis. In fact, many barriers to industrialsymbiosis are identified and detailed in the liter-ature. In addition to the usual problems of busi-ness development, these barriers are rooted inthe operational, financial, and behavioral issuesraised by the need to work across organizations(Lowe et al. 1995; Chertow 2000; Chertow 2003;Gibbs 2003; Ehrenfeld and Gertler 1997; Mirata2004).5 Neither are the externalities of symbioticarrangements all positive in the minds of citizensconcerned about pollution and health effects ofindustry. Indeed, one town’s economic develop-ment can be another town’s emissions source ortraffic jam.6

Planning and Self-Organizationin Industrial Symbiosis

A surprising aspect of industrial symbiosis nowwell known to researchers and policy makers hasbeen that efforts to plan industrial ecosystemsto achieve the benefits described above have re-sulted in many failures. Gibbs (Gibbs 2003) and

colleagues investigated 63 “eco-industrial” sites:30 in the United States and 33 in Europe. Thework of the Gibbs team found little success inthe United States and somewhat more successin Europe. After carefully examining the data,Gibbs concluded that “initiatives based upon theinterchange of wastes and cascading of energy arefew in number and difficult to organize” (Gibbset al. 2005).

Half of the U.S. sites the Gibbs team re-viewed were associated with the work of the U.S.President’s Council on Sustainable Development(USPCSD), formed during the Clinton admin-istration. In its report, Sustainable America, theUSPCSD (1996, 104) recommended that “Fed-eral and state agencies should assist communitiesthat want to create eco-industrial parks that clus-ter businesses in the same area to create new mod-els of industrial efficiency, cooperation, and en-vironmental responsibility.” These 15 so-called“eco-industrial parks” were at various stages ofplanning or implementation, including four sitesfunded by the U.S. Environmental ProtectionAgency in Chattanooga, Tennessee; Baltimore,Maryland; Brownsville, Texas; and Cape Charles,Virginia. The sites were announced at a PCSDtask force meeting in October 1996. The authorattended this meeting ten years ago and oversawan independent telephone and Internet survey ofthese 15 projects completed in 2005. The resultsare shown in table 1.7

The USPCSD consensus definition of an eco-industrial park was as follows:

A community of businesses that cooperatewith each other and with the local commu-nity to efficiently share resources (informa-tion, materials, water, energy, infrastructureand natural habitat), leading to economicgains, gains in environmental quality, and eq-uitable enhancement of human resources forthe business and local community. (USPCSD1997, 36)

Gibbs and colleagues (Gibbs et al. 2002) de-termined that of the 15 projects (including theBurnside Industrial Park in Nova Scotia, Canada,cited by both Gibbs and colleagues 2002 andthe USPCSD in 1996), five were open, threehad failed, and seven were still identified as“planned.”


F O RU M

Tabl

e1

Stat

usof

eco-

indu

stria

lpar

kpr

ojec

tsite

spr

ofile

dby

the

U.S

.Pre

siden

t’sC

ounc

ilon

Sust

aina

ble

Dev

elop

men

t,19

96

Gib

bset

al.c

ateg

ory

Revi

sed

cate

gory

Nam

e,lo

catio

nlis

ted

in19

968

Sum

mar

yas

of20

05(2

002–

2004

)20

05–2

006

1.Fa

irfiel

dEc

olog

ical

Indu

stria

lPar

k—Ba

ltim

ore,

Mar

ylan

dEx

istin

gin

dust

rialp

ark

whe

nfu

nded

byU

S-PC

SD,n

owno

long

erpu

rsui

ngEI

Pco

ncep

t.A

sof2

002,

chan

ged

nam

eto

Fairfi

eld

Eco-

Busin

essP

ark.

Ope

nO

pen

with

chan

ged

conc

ept

2.Br

owns

ville

Eco-

Indu

stria

lPar

k—Br

owns

ville

,T

exas

Plan

ned

asre

gion

alex

chan

gean

dda

taba

se;B

echt

elC

orp.

perfo

rmed

som

eda

taba

sede

velo

pmen

t.Fa

iled

orst

alle

dR

egio

nale

xcha

nge

conc

eptf

aile

d3.

Riv

ersid

eEc

o-In

dust

rialP

ark—

Burli

ngto

n,V

erm

ont

Now

calle

dIn

terv

ale

Com

mun

ityFo

odEn

terp

rise

Cen

ters

uppo

rtin

gla

ndan

dfa

rm-b

ased

ente

rpris

es.

Plan

ned

Ope

nw

ithch

ange

dco

ncep

t4.

Burn

side

Eco-

Indu

stria

lPar

k—H

alifa

x,N

ova

Scot

ia,C

anad

aO

pene

d“E

co-e

ffici

ency

Cen

tre”

in19

98or

ient

edto

rese

arch

and

deve

lopm

ent;

augm

ents

exist

ing

larg

ein

dust

rialp

ark.

Ope

nO

pen

asan

indu

stria

lpar

kw

ithR

&D

cent

er

5.Po

rtof

Cap

eC

harle

sSus

tain

able

Tec

hnol

ogie

sIn

dust

rialP

ark—

Cap

eC

harle

s,V

irgin

iaO

pene

das

EIP

in20

00,fi

nals

ale

ofas

sets

occu

rrin

gsp

ring

2005

.O

pen

Clo

sed

6.C

ivan

oIn

dust

rialE

co-P

ark—

Tuc

son,

Ariz

ona

Res

iden

tial“

new

tow

n”de

velo

pmen

tpla

nned

tose

ek“s

usta

inab

lete

chno

logy

”bu

sines

sese

ven

ifno

EIP.

Faile

dor

stal

led

Nev

erem

erge

di

7.T

heV

olun

teer

Site

—C

hatt

anoo

ga,T

enne

ssee

Plan

sto

conv

ertt

heV

olun

teer

Arm

yA

mm

uniti

onPl

ants

iteto

anen

viro

nmen

talt

echn

olog

yce

nter

wer

ene

verc

ompl

eted

.ii

Plan

ned

Nev

erem

erge

d,op

enas

trad

ition

alin

dust

rial

park

8.Ea

stBa

yEc

o-In

dust

rialP

ark—

Ala

med

aC

ount

y,C

alifo

rnia

Plan

ned

asre

sour

cere

cove

rypa

rkfo

rwas

tedi

vers

ion

byA

lam

eda

Cou

nty

Was

teM

anag

emen

tAut

horit

y;ha

sno

tcom

eto

fruiti

onas

EIP.

iii

Plan

ned

Plan

ned,

chan

ged

conc

ept

9.G

reen

Inst

itute

Eco-

Indu

stria

lPa

rk—

Min

neap

olis,

Min

neso

taO

rigin

alpr

ojec

tfoc

usto

crea

tejo

bsin

poor

neig

hbor

hood

;tod

ay“P

hilli

psEc

o-En

terp

rise

Cen

ter”

hasc

reat

edov

er10

0jo

bs.

Ope

nO

pen

assu

stai

nabl

ede

velo

pmen

tpro

ject

10.P

latt

sbur

ghEc

o-In

dust

rialP

ark—

Plat

tsbu

rgh,

New

York

From

orig

inal

emph

asis

onEI

Pw

ithIS

O14

000/

EMS

Prog

ram

;now

calle

dPl

atts

burg

hA

irbas

eR

edev

elop

men

tCor

p.,w

ith60

tena

nts.

Faile

dor

stal

led

Ope

nas

stan

dard

indu

stria

lpar

kiv

11.R

aym

ond

Gre

enEc

o-In

dust

rial

Park

—R

aym

ond,

Was

hing

ton

Orig

inal

plan

tode

velo

pa

park

“with

ina

seco

ndgr

owth

coas

talf

ores

t”di

dno

tocc

ur;d

escr

ibed

bylo

cal

offic

ials

asha

ving

been

“pla

nned

arou

nda

man

ufac

turin

gbu

sines

stha

tdid

n’tm

ater

ializ

e.”v

Plan

ned

Nev

erem

erge

d



12.S

kagi

tCou

nty

Envi

ronm

enta

lInd

ustr

ial

Park

—Sk

agit

Cou

nty,

Was

hing

ton

Plan

tode

velo

pan

indu

stria

lpar

kw

ithso

me

EIP

feat

ures

did

noth

ave

asit

ede

term

ined

and

loca

lgo

vern

men

tsta

ffre

port

edth

ere

was

little

activ

ityfo

llow

ing

a19

96fe

asib

ility

stud

y.v

i

Plan

ned

Nev

erem

erge

d

13.S

hady

Side

Eco-

Busin

essP

ark—

Shad

ySi

de,

Mar

ylan

dEn

visio

ned

reno

vatin

gex

istin

gfa

cilit

yin

disa

dvan

tage

dco

mm

unity

;suc

cess

oror

gani

zatio

ndo

mai

nna

me

for

sale

,no

furt

herm

entio

n.v

ii

Plan

ned

Nev

erem

erge

d

14.S

tony

field

Lond

onde

rry

Eco-

Indu

stria

lPa

rk—

Lond

onde

rry,

New

Ham

pshi

rePr

ivat

ede

velo

pern

amed

in19

98.A

ncho

rte

nant

—72

0-M

Wga

s-fir

edpo

wer

stat

ion—

open

edin

2003

and

faci

lity

sold

tocr

edito

rsin

2004

.vii

i

Ope

nO

pen,

buts

tatu

sun

cert

ain

15.T

rent

onEc

o-In

dust

rialC

ompl

ex—

Tre

nton

,N

ewJe

rsey

Inte

nded

asa

netw

ork

offir

mso

rgan

ized

bym

ultis

take

hold

ergr

oup;

early

feas

ibili

tyst

udy

done

in19

98,c

ityst

affr

epor

tint

eres

tlos

tin

proj

ectb

y19

98.ix

Plan

ned

Nev

erem

erge

d

Sour

ces:

i Brig

gs20

05;ii

Hew

ett2

003;

iii Ba

kke

2005

;ivPA

RC

-USA

2005

;vC

haffe

e20

05;v

i McC

ullo

ugh

2005

;vii

Food

Plan

et.c

om20

05;v

iii A

ESgi

vesG

rani

teR

idge

pow

erpl

antb

ack

toba

nk20

04,ix

Tag

liafe

rri2

005.


F O RU M

Although the 15 proposed projects were alllabeled “eco-industrial parks” a decade ago, noneadheres to the idealistic vision of the USPCSDabove (Chertow 2004). Yet neither do simple la-bels such as “planned” or “failed” give a nuancedsense of what occurred. Gibbs and colleagues cat-egorized ten projects as either “planned” (seven)or “failed or stalled” (three). The Brownsville,Texas, concept did not take hold, but althoughtwo others of these ten failed as EIPs, theydid become conventional industrial parks (Chat-tanooga and Plattsburgh). I record only oneproject in California’s East Bay as still “planned,”although the concept has changed. Now calledthe Alameda County EIP, it was, from the start,sited by a waste transfer station with a plan tobecome a resource recovery park (Liss 2005),although this vision, too, has been deferred bymarket realities (Bakke 2005). A fifth projectin Burlington, Vermont, opened with a changedconcept as the “Intervale Community Food En-terprise Center.” The Web site describes a newpublic market “to be built on the site of the for-mer Eco-Park” (Intervale 2005, emphasis added).The other five in our survey had achieved fewresults: we were told in various ways that, aftera time, “nothing happened.” Sometimes this wasafter a feasibility study or an initial request forproposals, or a series of meetings with stakehold-ers, but we had the impression that most werenever embedded in the planning process but wereideas without a strong basis. These are labeled intable 1 as “never emerged.”

Of the five projects categorized by Gibbs andcolleagues as “open,” the Fairfield Park in Bal-timore reported changing its name in 2002 andleaving behind the EIP concept (Build Fairfield2005). Both the projects in Minneapolis andNova Scotia fulfilled the main objectives estab-lished in the PCSD case studies, but neither is aself-defined EIP (see below). Given the high fail-ure rates of businesses, the fact that 2 of 15 origi-nal PCSD projects, in Londonderry, New Hamp-shire, and Cape Charles, Virginia, did open asself-defined EIPs could be viewed as a success. Un-fortunately, both have run into serious difficultiessince 2004. The Londonderry anchor tenant—anew 720-MW power plant—is operating but isin receivership, given the high price of naturalgas and steep tax liabilities (“AES gives Granite

Ridge power plant back to bank” 2004). CapeCharles still has an office building but has sold offits few remaining assets for other land uses (Slone2005).

Upon further analysis, we can now see thatthe 15 projects were quite a mixed bag. TheBrownsville, Texas, project, according to theUSPCSD (1997) documents, was to be a re-gional materials exchange; it was clearly notedthat “a defined ‘eco-industrial park’ is consid-ered as one possible component of regional indus-trial symbiosis, but not the driving force” (emphasisadded). With respect to the Burnside IndustrialPark in Nova Scotia, the original PCSD (1997)documents name the key feature of Burnside asa “six year multi-disciplinary, multi-institutionalstudy of requirements” using a cooperative part-nership among academia, government, and theprivate sector. The Burnside Eco-Efficiency Cen-tre largely fits this bill and continues to functionas a living laboratory for research and demon-stration, but the Burnside Park itself, with 1400companies, does not meet the definition of anEIP above (Barchard 2005). The Green Instituteproject intended to create employment opportu-nities through a very small EIP and to incorpo-rate environmental education in “the poorest andmost diverse neighborhood in Minnesota.” It hasachieved success as “the Phillips Eco-EnterpriseCenter” with environmental building featuresand job creation organized by a grassroots group.

In reviewing the previous decade, then, wecan see in these 15 projects, as well as many otherearly efforts such as those initiated in the pri-vate sector by the group now known as the U.S.Business Council for Sustainable Development(Forward and Mangan 1999; Chertow 2000),what paleontologists might call a series of “evolu-tionary experiments.” Reexamining the originalUSPCSD documents, we see that a variety of ap-proaches to eco-development were tried—fromregional databases, to waste parks, to retrofits ofexisting parks—for a plethora of reasons—innercity development, rural development, and en-hancing the “new town”9 concept. They reliedto a large extent on public funding and althoughsome became active, others died off quickly. Intime, however, it seemed as if all of the earlyprojects were lumped together as a single idealizedmodel and, when the success rate was found to



Liquid Fertilizer Production

StatoilRefinery

Energy E2 AsnaesPower Station

Novo Nordisk A/S Novozymes A/SPharmaceuticals

Farms

LakeTissø

Cement;roads

Fishfarming

BPBGyproc A/SWall-board

Plant

Water

Water

Water

Sludge(treated)

Heat

ScrubberSludge

Ste

am

Boi

ler

wat

er

Coo

ling

wat

er

Ste

amA-S Soilrem

Hot

wat

er

Municipality ofKalundborg

District Heating

Wastewater Treatment Plant

Sulfur

Org

anic

re

sidu

es

Fly ash

Heat

Slu

dge

Gas (back up)

Yeastslurry

Was

tew

ater

Figure 3 Industrial symbiosis of Kalundborg, Denmark.

be very low, the word got out that eco-industrialparks just do not work.

In looking back over this period, I suspect thatsome projects were prematurely put on the PCSDlist in the first place to fill out the cohort fora meeting that would achieve media attentionin October 1996. It was good for the govern-ment agencies to have more projects and goodfor many, especially those within small commu-nities, to be recognized on the national stage andpossibly be made eligible for public funding. Re-viewing the projects codified on the Internet as aseries of case studies by the Smart Growth Net-work (Smart Growth 2005), however, is like en-tering a time warp beginning in 1996 and trail-ing off after 1998. Of the dozens of “Google hits”called up for each of the projects, most are echoesof a handful of Web sites, even if the sites are re-labeled with current dates. As a founding memberof the fledgling U.S.–Canadian NGO called theEco-Industrial Development Council I can saywith certainty that dozens of evolutionary exper-iments continue in North America well beyondthe PCSD projects, but they are seldom neatlypackaged in an EIP box, thus casting the EIPconcept as an artifact of cyberspace.

Clearly, a more robust model of industrial sym-biosis is needed. Alternatively, many of the suc-cessful industrial ecosystems described in this ar-ticle did not arise in the ways pursued by thePCSD. One feature that several of these havein common is the experience of a quiet periodwhere firms engaged in exchanges among them-selves, unconscious of a bigger picture, followedby an act of discovery that revealed the patternof existing symbiotic exchanges and the resultingenvironmental benefits.

The most colorful case returns us to Kalund-borg, Denmark (figure 3). The first exchangeswere in the 1970s, and, by the late 1980s, at leastten additional exchanges had begun across mul-tiple firms (Symbiosis Institute, 2003). Yet, un-til some local high school students prepared ascience project in 1989 in which they made ascale model of all the pipelines and connectionsin their small community, the unique aspects ofthe project went largely unnoticed (Christensen1998). Recognition of Kalundborg’s symbiotic at-tributes was an uncovering of what already ex-isted rather than the exploration of a new fron-tier. Following this high school project, still ondisplay in Kalundborg, came the European media


F O RU M

(Knight 1990; Barnes 1992), and then academics(Engberg 1993; Gertler 1995) to describe the ex-isting network from a broader environmental per-spective.

Two fundamental conclusions have gained ac-ceptance concerning the Kalundborg case. First,we see that rather than resulting from planningor a multistakeholder process such as the onespursued through the President’s Council on Sus-tainable Development, the Kalundborg symbio-sis emerged from self-organization initiated in theprivate sector to achieve certain goals listed ear-lier, such as cost reduction, revenue enhance-ment, business expansion, and securing long-term access to water and energy. This impliesthat the symbiosis was not “seen” by outsidersbecause the exchanges emerged from the invis-ible hand of the market rather than direct gov-ernment policy or involvement. Second, once arevelation was made, a coordinative function wasfound to be helpful in organizing more exchangesand moving them forward. In Kalundborg, forexample, managers belonged to an EnvironmentClub and a coordinative organization, the Sym-biosis Institute, was launched in 1996 as partof Kalundborg’s industrial development agency,specifically working to accelerate the numberand complexity of new exchanges (Jacobsen2005).

If key types of mixed industrial ecosystems areself-organizing and arise in this manner, then asimilar pattern of discovery, in which preexistingsymbioses were later recognized, would be ob-served in other locations. Indeed, we see this pat-tern repeated in several other multiple-industryecosystems wherein significant industrial activ-ity and interfirm sharing were found to be wellestablished in practice, but never mapped anddescribed using ecological metaphors. A com-mon language has even emerged that reflectsthe notion of “uncovering” a pattern of tradesmore extensive or more interconnected than hadbeen previously realized. Several examples havebeen found that follow this pattern of discov-ery in Australia, Austria, Germany, Finland, andthe United States. These are discussed in turnbelow.! In the example of Kwinana, Australia, dis-

cussed above, Van Berkel (Van Berkel

2004) writes that following through on theidea of examining large volume wastes fromminerals processing in the context of theMining, Minerals and Sustainable Devel-opment Project under the auspices of theWorld Business Council for Sustainable De-velopment “revealed that quite significantregional by-product synergies are alreadyhappening in resource processing intensiveregions” [emphasis added]. As noted earlier,this prominent and diversified Australianmineral processing area has over 100 sym-biotic exchanges.! Having become aware of Kalundborg,two Austrian researchers also wonderedwhether that industrial ecosystem would befound to be unique. In developing the con-cept of an “industrial recycling-network,”Schwarz and Steininger (Schwarz andSteininger 1995, 1997) studied the spa-tially broader province of Styria and un-covered a network with a high degree ofdiversity and complexity. The researchersfound exchanges consisting of hundreds ofthousands of tons of materials includingpower plant gypsum, steel slag, sawdust,and recyclable paper and wood. Althoughthe Kalundborg participants became con-scious of the environmental characteristicsof their exchanges over time, the Styriancompanies were not generally aware of theubiquitous networking of regional mate-rial flows, and the authors were concernedthat these companies were missing out onadditional benefits. They stressed the im-portance of a coordinative function as inKalundborg to try to increase exchange andimprove internal and external communi-cation. The notion of recycling networksspread over the last decade so that a central-ized “Regional Recycling Information Sys-tem” (REGRIS) in the Oldenburger Mun-sterland Region of northwest Germany nowsupports the management of intercompanyinformation flows, provides data to localfirms about recycling opportunities, and co-ordinates recycling activities (Milchrahmand Hasler 2002).! In Finland, Korhonen and colleagues(Korhonen et al. 1999) describe the city



of Jyvaskyla where the energy supply is or-ganized around co-production of heat andelectricity and includes industrial wastesused as fuels in a highly efficient system.The system arose for economic/regulatoryreasons, but was not consciously labeled as“industrial ecology” or “industrial symbio-sis” prior to the intervention of professorsthere.! In North Carolina, USA, staff membersof the six-county Triangle J Council ofGovernments came to realize over timethat their planning work to create theIndustrial Ecosystem Development Projectwas actually based on a foundation of ex-change activity. From 1997 to 1999, aninventory of business inputs and outputswas conducted and 182 businesses repre-senting 108 different business segments re-sponded to the inventory survey. Accordingto the final project report, two-thirds of allthose surveyed were characterized as havingsome experience with reuse (Kincaid 1999;Kincaid 2005). Of these, the researchersfound that a surprising 36 percent had al-ready engaged in activities beyond simplerecycling, thus affirming favorable marketconditions and a symbiosis mindset thatwas unseen prior to the survey (Kincaid2005).

A parallel to the self-organizing kernels ofindustrial symbiosis described in the examplesabove can be drawn from an examination ofthe roots and development of business clus-ters as analyzed by Porter. In describing howthese “geographic concentrations of intercon-nected companies and institutions” arise, Porter(Porter 1998b) notes many different motivat-ing forces including the availability of special-ized skills, role of existing suppliers, scarcity con-ditions, and availability of natural resources, aswell as chance, although “what looks like chancemay sometimes be the result of pre-existing lo-cational circumstances” (Porter 1998a). Parallelto the dubious mission of creating eco-industrialparks from scratch, as we have seen, Porter high-lights the difficulty of seeding clusters whereno important preexisting locational advantagesexist.

With respect to the role of government, Porternotes that “the appropriate policy towards clus-ter development is usually to build on existingor emerging areas that have passed a markettest” (Porter 1998a). Many observers of indus-trial symbiosis similarly suggest the desirabilityof working from an established base, a conclu-sion affirmed here (Schwarz and Steininger 1997;Chertow 2000; Korhonen 2002; Baas and Boons2004; Desrochers 2004; Jacobsen and Anderberg2005). An industrial ecology view makes this es-pecially clear, as it depends on the existence ofmaterial and energy flows for its craft.

Understanding EmergingSymbioses

From the analysis presented thus far, contrast-ing a planning approach with self-organizationcan be summarized in two stylized models of sym-biosis as follows:

Planned EIP model. This model includes a con-scious effort to identify companies from differentindustries and locate them together so that theycan share resources across and among them. Typ-ical U.S. planning for these systems has involvedthe formation of a stakeholder group of diverse ac-tors to guide the process and the participation ofat least one governmental or quasi-governmentalagency with some powers to encourage develop-ment, such as land use planning and/or zoning,grant giving, or long-term financing.

Self-organizing symbiosis model. In this model,an industrial ecosystem emerges from decisionsby private actors motivated to exchange resourcesto meet goals such as cost reduction, revenue en-hancement, or business expansion. The individ-ual initiative to begin resource exchange facesa market test and if the exchanges are success-ful, more may follow if there is on-going mutualself-interest. In the early stages there is no con-sciousness by participants of “industrial symbio-sis” or inclusion in an “industrial ecosystem,” butthis can develop over time. The projects can bestrengthened by post facto coordination and en-couragement.

If, as described here, the self-organizing sym-biosis model building from kernels of coopera-tion and exchange tends to be more success-ful, then we must examine it more closely to


F O RU M

Table 2 Projects sharing characteristics of uncovered industrial symbiosis referenced in this article

Project Relevant section reference

Kwinana, Australia See Pursuing Industrial SymbiosisGladstone, Australia See Pursuing Industrial Symbiosis and Bossilkov and colleagues (2005)Triangle J, North Carolina Planned project that uncovered self-organized roots among member

firms—discussed in Planning and Self-Organization in IndustrialSymbiosis

Barceloneta, Puerto Rico See Understanding Emerging SymbiosesGuayama, Puerto Rico See Understanding Emerging SymbiosesKalundborg, Denmark Mentioned throughoutStyria, Austria See Planning and Self-Organization in Industrial SymbiosisJyvaskyla, Finland See Planning and Self-Organization in Industrial SymbiosisNational Industrial Symbiosis See Policy Proposals for Uncovering Symbiosis (U.K. Symbiosis)Guitang Group, China See Pursuing Industrial Symbiosis—some parts of symbiosis planned

but others appeared opportunistically (Zhu et al. 2007)Burnside Industrial Park, Canada See Planning and Self-Organization in Industrial Symbiosis—symbioses

not planned but later coordinated by the Eco-Efficiency CentreAlberta Heartland, Canada See Pursuing Industrial Symbiosis

look for patterns or common characteristics.The USPCSD projects provide a convenient, ifnonrandom, sample of projects proposed througha process. We do not have a comparable setof self-organizing symbiosis projects to exam-ine because, by their very nature, they are notknown until there has been some success andan uncovering event occurs. Acknowledging theproblem of selection bias, projects more closelyresembling self-organizing symbiosis that are re-ferred to throughout this article are listed intable 2.

Boons and Berends (2001) and Baas andBoons (2004) offer an important theoretical per-spective suggesting how the emergence of indus-trial symbioses based on the exploitation of win-win situations among area firms could lead to aform of organization that embraces sustainable in-dustrial development. Their three-part diagram isdepicted in figure 4.

The first stage, regional efficiency, is describedas “autonomous decision-making by firms; co-ordination with local firms to decrease ineffi-ciencies (i.e., ‘utility sharing’).” Stage 2 broadensgoals and membership into regional learning,where the authors see that “based on mutualrecognition and trust, firms and other partnersexchange knowledge, and broaden the definitionof sustainability on which they act.” The thirdstage, a sustainable industrial district, shows fur-

ther evolution toward a strategic vision and col-laborative action rooted in sustainability.

This diagram offers insight on two importantquestions: what do symbiotic relationships looklike before they are broadly known and at whatpoint are they “uncovered” in industrial ecosys-tems? Stage 1 shows that companies can coop-erate for reasons of economic efficiency. Oncea regional efficiency gains momentum based onautonomous decision-making, it may continueto thrive and enter the more advanced stageof learning denoted in Stage 2. With respectto awareness by the companies of the environ-mental benefits being gained, by definition, theycannot be “uncovered” before the exchanges areestablished, which is why attempts at creationoutside of the market are likely to fail. The dia-gram does not offer a temporal explanation—thatis, when the collection of companies involved atthe regional efficiency stage might move to thesubsequent stages illustrated.

In the case of Kalundborg, there were well-developed regional efficiencies (Stage 1) by thetime the symbiosis was brought to broader atten-tion and may even have entered Stage 2 giventhe informal informational networks there. Per-haps one way to understand the dominance ofKalundborg in the literature is to see it as havingmoved directionally further to the right of thesestages than other relevant projects at the time



I Regional efficiency II Regional learning III Sustainable industrial district

Figure 4 From regional efficiency to a sustainable industrial district. Source: Baas and Boons 2004. Used bypermission from authors.

of uncovering. Nascent symbioses found earlierin their possible trajectories are less stable and,therefore, their futures are more uncertain,whereas Kalundborg was fairly well-developed atthe time of its unveiling. Rather than seeingKalundborg as chance or a historical accident,a fuller explanation would be to recognize that itbuilt on existing scarcities (in the case of ground-water), opportunities (in the case of by-productuse by new entrants), regulatory changes (in thecase of reuse of organics and impetus to pursueflue gas desulfurization), and other locational ad-vantages (including the port and the availabilityof industrial land).

With respect to the diagram from Baas andBoons, it is not clear that Stage 3 is coming anytime soon nor that a strongly collective orienta-tion will ever fully fit with the other imperativesof firms. Although these three stages are rootedin existing development, Baas and Boons alsotalk about an earlier “selection” stage for new orgreenfield sites. Inclusion of this selection stageahead of Stage 1 raises the previously unsuccessfulprospect of development from scratch, before thesymbioses take shape, although with two notableexceptions. Clearly, there have been planned de-velopments, especially of single-industry dom-inated systems, that could successfully assem-ble core actors and organize benefits, as seenin many chemical and petrochemical complexes.The other exception occurs most notably in EastAsia, where formal planning is much more insti-tutionalized in countries such as China, Korea,and Singapore.

The symbioses mentioned in table 2, whethermotivated by economic, social, environmental,or regulatory forces, have met a market test andcontinue onward. In these cases, two periods canbe observed: the initiation phase before discov-ery and a second period when the symbiotic ex-changes are developed further with some new en-trants and some die-offs of specific trades. Then,

at some unspecified time, an uncovering eventoccurs, usually catalyzed by members of a thirdparty such as an academic institution or businessassociation, thereby leading to greater awarenessof the exchanges.

Some industrial groupings of companies,as in Australia’s mineral processing industryand Austria’s recycling network in Styria, findproductivity-enhancing efficiencies through self-organization. Stronger regulatory roots combinedwith elements of self-organization are seen inthose uncovered in Puerto Rico. In one case, theBarceloneta pharmaceutical companies teamedup in a private initiative to share wastewa-ter processing in response to regulatory changesand, subsequently, further exchanges developed(Ashton 2003). In the case of Guayama, PuertoRico, regulatory conditions negotiated with gov-ernment drove a new power station developmentto use four million gallons10 per day of municipalwaste water and, in turn, to co-generate steamfor a nearby refinery, with other exchanges fol-lowing in their wake (Chertow and Lombardi2005). Still, the grouping of companies that be-came involved did not call themselves by a namethat expressed recognition of the environmentalbenefits, nor was there a clear coordinative func-tion that might attract new exchanges, let alonean institutionalized sense of organizational learn-ing. In fact, these projects were not attemptingto follow a symbiosis model, nor were they fullyconscious of the collective environmental ben-efits being gained. To generalize, then, at somepoint following self-organization, uncovering oc-curs, which opens the door for, but does notdictate, active coordination and further devel-opment of exchanges, as depicted in figure 5.

Given the phenomenon of “uncovering” re-lationships in industrial ecosystems, how can weestablish more foresight in addition to hindsight?A role for government and policy ideas are offeredin the following section.


F O RU M

1. Period of initial self-organization for reasons of economic efficiency and/or to meet regulatory conditions

2. Period of sustaining interfirm exchange including some new entrants and some die-offs

Time during which an uncovering event may occur typically based on third party review

Choices concerning efforts to coordinate and attract additional symbioses

Figure 5 Empirical findings of industrial symbiosis progression.

Policy Proposals for UncoveringSymbiosis

The more spontaneous view of industrial sym-biosis raises the question of whether a laissez-faire approach might be desirable, what role therecould be for government, and, more fundamen-tally, whether there is any stage at which gov-ernment intervention might be effective. Giventhe importance of coordination, it should also beasked who might best deliver this function, in-cluding various levels of government, nongovern-mental organizations such as trade associationsand universities, or on-site entities at industrialparks or clusters.

A basic rationale for public involvement hasbeen indicated by John Ehrenfeld (Ehrenfeld2003). He points out that industrial ecosystemsprovide a greater level of public benefit than stan-dard industrial networks because they offer in-creased environmental benefits. Consequently,they are likely to need some type of public as-sistance to continue to deliver public goods atthis greater level because, left to their own de-vices, private firms will typically underdeliverthem. Firms also face risks of association, suchas an increased level of dependence on othersand the extent of the transaction costs involved

in participation across firms, including searchand coordination costs, some of which couldbe borne externally (Ehrenfeld and Chertow2002).

Three policy ideas, then, are useful for gov-ernment and business to move industrial symbio-sis forward during different stages of discovery.These are to:

1. bring to light kernels of cooperative activ-ity that are still hidden;

2. assist the kernels that are taking shape; and3. provide incentives to catalyze new kernels

by identifying “precursors to symbiosis.”

These three potential programs also imply notsupporting projects, through public or private in-vestment, that have much wishful thinking butno tangible kernels to roast.

Bring existing kernels to light: Academic re-searchers, having revealed many patterns of ex-change, have a good sense of where others mightlurk, especially based on industry type. Businessmanagers know what large by-product exchangesthey are engaged in but often lack access toinformation about their neighbors in differentindustries. As a start, reconnaissance teams ofresearchers and managers who can map the flowsfrom heavy industry areas such as mining, steel,



cement, and chemicals would be likely to un-cover many kernels of exchange. As the researchon recycling networks suggests, these exchangerelationships are quite broadly dispersed and caneven include smaller and less industrial compa-nies. Systematic means of bringing these existingkernels to light through mapping of flows andidentification of relevant institutions will informhow to proceed.

Assist kernels beginning to take shape. Where akernel is known to exist and is emerging, it isan excellent candidate for assistance because itimplies that at least two or three firms are al-ready on a trajectory toward industrial symbio-sis. As innovation research has shown repeat-edly, it is much easier to continue on a giventrajectory than to shift to another, perhaps ex-plaining why many firms find it difficult to con-tinuously innovate, thus missing out on key op-portunities (Christensen 1997; Dosi et al. 1988).With respect to companies that adopt an indus-trial symbiosis mindset, it is likely that some oftheir managers will be able to spot new oppor-tunities, as they will be thinking “what can wetrade?” even knowing that additional transactioncosts may be incurred by the establishment of fur-ther exchanges. Technical or financial assistanceto facilitate the exchanges envisioned by thesemanagers could accelerate the evolutionary pro-cess. In Guayama, Puerto Rico, for example, sincethe power plant opened in late 2002, symbioticrelationships have been proposed by two phar-maceutical companies for additional wastewaterreuse, as well as by two chemical companies forspecific by-product sharing (Chertow and Lom-bardi 2005).

A key question with existing kernels, how-ever, is whether it is the company itself or an in-dividual project champion (or champions) thatdrives the nascent symbioses. Here, attempts tohelp a kernel to “pop” must consider what willhappen if key personnel are removed or retire.Thus, policy must strive to be sustainable even inthe face of generational change, business mergers,and economic trends. One useful model here maybe the creation or extension of NGOs that arebusiness associations to assist with kernel devel-opment. An offshoot of the U.K. Business Coun-cil for Sustainable Development has now cre-

ated the world’s largest coordinating entity forby-product reuse within regional business clus-ters, called “The National Industrial SymbiosisProgramme.” NISP describes as its mission thatit “facilitates links between industries from dif-ferent sectors to create sustainable commercialopportunities and improve resource efficiency”(NISP 2005).

Identify and assess “precursors to symbiosis” ascatalysts for new kernels. Many common envi-ronmentally related activities can be seen asprecursors to symbiosis—defined as a resourceexchange with a public goods component but in-volving only one or two firms or other organiza-tions. Examples of these precursors to symbiosisare resource sharing projects involving (1) co-generation, (2) landfill gas, and (3) wastewaterreuse. These projects can be driven by the pub-lic or private sectors, but where they have begunfor whatever array of reasons, they can be usedas bridges to more extensive exchange. Anotherprecursor to symbiosis is the existence of one suc-cessful material exchange, sometimes referred toas green twinning, which is shown to be finan-cially and environmentally beneficial, such as airpollution control waste to gypsum board, or steelslag to cement (Forward and Mangan 1999). Asa means of screening and targeting public invest-ment in symbiosis, projects involving these andother precursors could be identified as meritingfurther investigation as likely kernels of indus-trial symbiosis.

As noted in figure 5, actors in “discovered”kernels have the choice to opt collectivelyfor an increased level of coordination and/orconsciously pursue more symbiosis. The act ofuncovering is a critical inflection point at whichpublic incentives to continue symbiosis andmaintain the spillover environmental benefitscould be offered, if desired by the key actors.The three roles outlined above of identifying ker-nels, assisting them to grow, and catalyzing newones are all promising areas of policy develop-ment that can be done in ways that still recog-nize similarities and differences among projects.The development and implementation of policyin this arena needs to be informed by a sophisti-cated understanding of market function and firmbehavior.


F O RU M

Conclusion

In general, the empirical research in indus-trial symbiosis discussed here corroborates andexpands earlier threads in the literature in find-ing that attempts to plan “eco-industrial parks,”particularly from scratch, that involve significantmaterial and energy exchanges have rarely cometo fruition in a sustainable way (Ehrenfeld andGertler 1997; Chertow 1999; Baas and Boons2004; Gibbs and Deutz 2004; Korhonen andSnaikin 2005). We do not yet see new indus-trial ecosystems emerging from highly structuredplanning processes in Asia, despite the poten-tial to do so, although many symbiotic relation-ships have been identified (Liu et al 2004).11

Rather, an emergent characteristic of geographi-cally proximate firms successfully exploiting syn-ergistic exchanges is their evolution from op-portunistic business decisions. These kernels ofsymbiosis across firms, such as sharing ground-water or a specific material, are observed to benecessary preconditions for what sometimes be-come more extensive exchange networks. Cer-tain identifiable precursors of symbiosis, such asco-generation or waste water reuse, also emergefrom business decisions often rooted in regula-tory situations and can lead to more extensivesymbiotic cooperation as well.

Even when several exchanges have been im-plemented based on self-organization, the partic-ipants and neighbors generally have not recog-nized or described these phenomena in environ-mental terms, until some sort of uncovering eventhas occurred, either during the initial period orlater on when exchanges are more deeply rooted.Policies prescribed to encourage the uncoveringof symbioses include (1) forming reconnaissanceteams to identify industrial areas likely to havea baseline of exchanges and mapping their flowsaccordingly, (2) offering technical or financialassistance to increase the number of interactionsonce some kernels are found to be in place, in-spired by managers with a symbiotic mindset, and(3) pursuing locations where common symbioticprecursors already exist, such as co-generation,landfill gas mining, and waste water reuse, oftenas one-off activities, to determine whether theymay be likely candidates for technical or financialassistance as bridges to more extensive symbiosis.

Additional research is needed to illuminateconditions under which kernels and precursorshave survived and thrived. At the same time,although this article has focused primarily onpositive externalities of firms, there are manynegative ones, and some form of assessment ofwhen a reasonable carrying capacity is reachedin an area is also warranted. Evolutionary ex-periments should continue to provide better in-formation on which industrial sectors tend toproduce good participants, what the capacity lim-itations might be, and which incentives and be-haviors are most effective in fostering on-goingsymbiosis.

Economically and environmentally desirablesymbiotic exchanges are all around us. Identi-fying and fostering emerging industrial ecosys-tems offers the promise of many environmen-tal and other benefits. Grounded in what hasbeen learned empirically over the last 15 yearsabout the phenomenon of business co-locationknown as industrial symbiosis, this article helpssteer public and private actors to higher probabil-ity approaches to resource sharing in geographi-cally related industrial areas by choosing projectswith demonstrable kernels of self-organizationthat can emerge more fully as viable industrialecosystems.

Notes

1. With respect to the vocabulary of industrial sym-biosis, many related terms are used, such as indus-trial ecosystems, eco-industrial park, and industrialrecycling network. Definitions of the various termscan be found in the Encyclopedia of Energy in thearticle “Industrial Symbiosis” (Chertow 2004).

2. One megawatt (MW) = 1 ! 106 joule/sec (J, SI)" 3.412 ! 106 British Thermal Units/hr (BTU).

3. One tonne (t) = 103 kilograms (kg, SI) " 1.102short tons.

4. The U.S. Energy Policy Act of 2005 makes signif-icant changes to the Qualifying Facility (QF) pro-visions of the Public Utility Regulatory PoliciesAct (PURPA) by, among other things, condition-ally terminating the mandatory purchase and saleobligations for new QFs and tightening thermaloutput requirements for new qualifying cogenera-tion facilities (Hunton and Williams 2005).

5. Gibbs and colleagues (2002) summarize many ofthe descriptions of barriers as follows:



First, there are technical barriers, includ-ing the possibility that local industries donot have the potential to “fit together.”Second, informational barriers may makeit difficult to find new uses for wasteproducts, relating to poor informationregarding the potential market and po-tential supply. Third, economic barriersmay inhibit the incentive to use wastestreams as a resource if there is no reli-able market for them. Fourth, regulatorybarriers may prevent industries or indus-trial processes being linked together. Fi-nally, there may be motivational barrierswherein firms, public sector agencies andother relevant local actors must be will-ing to co-operate and commit themselvesto the process.

6. In addition, there has been concern for manyyears about the use of industrial by-products, es-pecially in symbioses involving agriculture. Thisis an important environmental and health issue tobe carefully examined in every case of by-productreuse, especially with increasing concerns aboutthe spread of diseases (Lifset 2001; Chertow 2004).In the examples cited in this article I have found nowidely reported evidence of environmental healthproblems resulting from by-product exchanges.

7. With thanks to K. Drakonakis, the survey in-volved contacting the last parties identified fromthe PCSD list and continuing until a further con-tact or related professional was reached to addressthe fate of the project. Internet searches were usedto continue to track and document outcomes be-cause project land does not, per se, disappear, buteventually goes to another use.

8. The names are drawn from the 1996 agendaavailable at <http://clinton2.nara.gov/PCSD/Publications/Eco Workshop.html#v>.

9. The “new town concept” refers to an urbanplanning idea to design communities for self-sufficiency by mixing uses such as housing, indus-try, cultural resources, and shopping.

10. 1 gallon " 3.8 liters.11. Currently, Chinese and Korean policies are focus-

ing on retrofit of existing parks to increase sym-biotic exchanges as a means of conserving waterand other resources and simultaneously increasingcompetitiveness.

References

AES gives Granite Ridge power plant back to bank.2004. Platts Magazines July 1:1.

Alberta’s Industrial Heartland Association. 2000. Astrategy for development in Alberta’s industrial heart-land: Eco-industrial networking. Summary for theMarch 2000 McCann Report. Alberta Indus-trial Heartland Inventory. Edmonton, Alberta,Canada.

Altham, J. and R. van Berkel. 2004. Industrial sym-biosis for regional sustainability: An update onAustralian initiatives. Paper presented at 11thInternational Sustainable Development ResearchConference, March, Manchester, UK.

Ashton, W. 2003. Why participate in eco-industrialnetworks? The case of Puerto Rico’s pharmaceu-tical manufacturing cluster. Paper presented at In-ternational Society for Industrial Ecology Confer-ence, July 1, University of Michigan, Ann Arbor,MI.

Baas, L. W. and F. A. Boons. 2004. An industrial ecol-ogy project in practice: Exploring the boundariesof decision-making levels in regional industrialsystems. Journal of Cleaner Production 12(8–10):1073–1085.

Bakke, R. 2005. Personal communication with R.Bakke, Senior Program Manager, StopWaste.org,15 April 2005.

Barchard, W. 2005. Personal communication with W.Barchard, Senior Pollution Prevention Advisor,Environment Canada, 4 March 2005.

Barnes, H. 1992. Fertile project exploits recycledwastes. Survey on Denmark. The Financial Times,8 October, 5.

Barton, A. 1999. The oil mallee project: A multifacetedindustrial ecology case study. Journal of IndustrialEcology 3(2–3): 161–176.

Boons, F. and M. Berends. 2001. Stretching the bound-ary: The possibilities of flexibility as an organisa-tional capability in industrial ecology. BusinessStrategy and the Environment 10(2): 115–124.

Briggs, T. Case Enterprise. 2005. Personal correspon-dence, March 7.

Build Fairfield. 2005. <www.buildfairfield.com>. Ac-cessed 8 April 2005.

Chaffee, R. Port of Willapa Harbor. 2005. Personalcorrespondence. March 7.

Chertow, M. R. 1999. The eco-industrial park modelreconsidered. Journal of Industrial Ecology 2(3): 8–10.

Chertow, M. R. 2000. Industrial symbiosis: Literatureand taxonomy. Annual Review of Energy and En-vironment 25:313–337.

Chertow, M. R. 2003. Evaluating the success of eco-industrial development. In Eco-Industrial Strate-gies: Unleashing Synergy Between Economic Devel-opment and the Environment, edited by J. M. E.


F O RU M

Cohen-Rosenthal. Sheffield, UK: Greenleaf Pub-lishing.

Chertow, M. R. 2004. Industrial symbiosis. In Ency-clopedia of Energy, edited by C. J. Cleveland. SanDiego: Elsevier.

Chertow, M. R. and D. R. Lombardi. 2005. Quan-tifying economic and environmental benefits ofco-located firms. Environmental Science and Tech-nology 39(17): 6535 –6541.

Chertow, M. R., W. Ashton, and J. C. Espinosa. 2007.Industrial symbiosis in Puerto Rico: Environmentally-related agglomeration economies. Forthcoming.

Christensen, C. 1997. The innovator’s dilemma: Whennew technologies cause great firms to fall. Boston,MA: Harvard Business School Press.

Christensen, J. 1998. Personal communication with J.Christensen, Kalundborg Industrial DevelopmentCouncil, October 1998.

Cote, R. P. 2003. A primer on industrial ecosystems:A strategy for sustainable industrial development.<http://sres.management.dal.ca/Files/PrimerApril2004.pdf>.

Department of the Environment and Heritage,Australia. 2004. Western Power’s integrated woodprocessing demonstration plant: Combining re-newable energy generation with salinity abate-ment. <www.deh.gov.au/industry/corporate/eecp/case-studies/western-power.html>. Accessed 20June 2004.

Desrochers, P. 2001. Cities and industrial symbiosis:Some historical perspectives and policy impli-cations. Journal of Industrial Ecology 5(4): 29–44.

Desrochers, P. 2004. Industrial symbiosis: The case formarket coordination. Journal of Cleaner Production12(8–10): 1099–1110.

Dosi, G., C. Freeman, R. Nelson, G. Silverberg, and L.Soete. 1988. Technical change and economic theory.London: Pinter Publisher.

Duranton, G. and D. Puga. 2003. Micro-foundationsof urban agglomeration economies. National Bu-reau of Economic Research Working Paper 9931.<http://dsl.nber.org/papers/w9931.pdf>.

Ehrenfeld, J. 2003. Nudging the market: Kinder, gen-tler public policies. Paper presented at USC Sym-posium on Sustainable Industrial Development,7 November, Los Angeles, CA.

Ehrenfeld, J. R. and M. R. Chertow. 2002. Industrialsymbiosis: The legacy of Kalundborg. In A Hand-book of Industrial Ecology, edited by R. U. Ayresand L. W. Ayres. Cheltenham, UK.: EdwardElgar.

Ehrenfeld, J. and N. Gertler. 1997. Industrial ecologyin practice: The evolution of interdependence at

Kalundborg. Journal of Industrial Ecology 1(1): 67–79.

Enecon, P. L. 2006. <www.enecon.com.au/about.html>. Accessed February 17 2006.

Engberg, H. 1993. Industrial symbiosis in Denmark. NewYork: New York University, Stern School of Busi-ness.

Food Planet.com 2005. <www.seeq.com/popupwrapper.jsp?referrer=http%3A%2F%2Fwww.sustainable.org%2Finformation%2Fpartnership.html&domain=naturaledge.org>

Forward, G. and A. Mangan. 1999. By-product synergy.The Bridge 29(1): 13–16.

Frosch, R. A. and N. E. Gallopoulos. 1989. Strate-gies for manufacturing. Scientific American 261(3):144–152.

Gertler, N. 1995. Industrial ecosystems: Developingsustainable industrial structures. Master’s the-sis, Massachusetts Institute of Technology, Cam-bridge, MA.

Gibbs, D. and P. Deutz. 2004. Implementing industrialecology? Planning for eco-industrial parks in theUSA. Geoforum 36(4): 429–439.

Gibbs, D., P. Deutz, and A. Procter. 2002. Sus-tainability and the local economy: The role ofeco-industrial parks. Paper presented at Ecositesand Eco-Centres in Europe, 19 June, Brussels,Belgium.

Gibbs, D., P. Deutz, and A. Procter. 2005. Industrialecology and eco-industrial development: A newparadigm for local and regional development? Re-gional Studies 39(2): 171–183.

Gibbs, D. C. 2003. Trust and networking in interfirmrelations: The case of eco-industrial development.Local Economy 18(3): 222–236.

Gunningham, N., R. Kagan, and D. Thornton. 2003.Shades of green, business, regulation and environ-ment. Palo Alto, CA: Stanford University Press.

Hayes, T. 2003. Cape Charles Sustainable TechnologyPark: The eco-indsutrial development strategy ofNorthampton County, Virginia. In Eco-industrialstrategies: Unleashing synergy between economicdevelopment and the environment, edited by E.Cohen-Rosenthal and J. Musnikov. Sheffield,UK: Greenleaf Publishing.

Hewett, K. S. 2003. Eco-construction trend may cometo Trousdale or Smith. The Tennessean, Septem-ber 22.

Hunton and Williams Client Alert. 2005. Energy Pol-icy Act of 2005 significantly changes the rules forqualifying facilities. August.

Intervale 2005. www.intervale.org/Overview.htm<http://www.intervale.org/Overview.htm> Ac-cessed April 10.



Jacobsen, N. B. 2005. Personal communication with N.B. Jacobsen, Oresund Environmental Academy, 4June 2005.

Jacobsen, N. B. and S. Anderberg. 2005. Under-standing the evolution of industrial symbioticnetworks—The case of Kalundborg. In Economicsof Industrial Ecology: Materials, Structural Change,and Spatial Scales, edited by M. J. J. C. J. M. vanden Bergh. Cambridge, MA: MIT Press.

Kincaid, J. 1999. Industrial ecosystem development projectreport. Research Triangle, NC: Triangle J Councilof Governments.

Kincaid, J. P. 2005. Personal communication with J. P.Kincaid, Sage Collaboration, 25 March 2005.

Knight, P. 1990. A rebirth of the pioneering spirit. TheFinancial Times, sec.1, p. 15, 14 November.

Korhonen, J. 2002. Two paths to industrial ecology:Applying the product-based and geographical ap-proaches. Journal of Environmental Planning andManagement 45(1): 39–57.

Korhonen, J. and J.-P. Snaikin. 2005. Analyzing theevolution of industrial ecosystems: Concepts andapplication. Ecological Economics 52(2): 169–186.

Korhonen, J., M. Wihersaari, and I. Savolainen. 1999.Industrial ecology of a regional energy supply sys-tem: The case of Jyvaskyla Region. Journal ofGreener Management International 26(1999): 57–67.

Krugman, P. 1991. Increasing returns and economic ge-ography. Journal of Political Economy 99(3): 483–489.

Lifset, R. 2001. Closing the loop and honing our tools.Journal of Industrial Ecology 5(4): 1–2.

Liss, G. 2005. Personal communication with G. Liss,Gary Liss and Associates, 11 April 2005.

Liu, Y., Z. Zhang, F. Wu, and N. Deng. 2004. Devel-opment of ecological industrial parks in China.Fresenius Environmental Bulletin 13(7):600–607.

Lowe, E. A., S. R. Moran, and D. B. Holmes. 1995. Afieldbook for the development of eco-industrial parks.Report for the U.S. Environmental ProtectionAgency. Oakland, CA: Indigo Development In-ternational.

Marshall, A. 1890. Principles of economics. London:Macmillan.

McCullough, D. 2005. Personal correspondence.March.

Milchrahm, E. and A. Hasler. 2002. Knowledge trans-fer in recycling networks: Fostering sustainabledevelopment. Journal of Universal Computer Sci-ences 8(5): 546–556.

Mirata, M. 2004. Experiences from early stages of a na-tional industrial symbiosis programme in the UK:

Determinants and and coordination challenges.Journal of Cleaner Production 12(8–10): 967–983.

NISP (National Industrial Symbiosis Programme).2005. <www.nisp.org.uk/>. Accessed April–May2005.

PARC-USA 2005. <www.parc-usa.com/TenantsFrameset/Tenants Frameset.htm>, May 23.

Porter, M. 1998b. Clusters and the new eco-nomics of competition. Harvard Business Review,November–December, 77–90.

Porter, M. E. 1998a. On competition. Boston: HarvardBusiness School Press.

Schwarz, E. J. and K. W. Steininger. 1995. The industrialrecycling-network enhancing regional development.Research Memorandum #9501. Graz, Austria:Department of Economics, University of Graz.

Schwarz, E. J. and K. W. Steininger. 1997. Implement-ing nature’s lesson: The industrial recycling net-work enhancing regional development. Journal ofCleaner Production 5(1–2): 47–56.

Slone, D. 2005. Personal communication with D.Slone, Partner, McGuireWoods, 2 June 2005.

Smart Growth 2005. <www.smartgrowth.org/casestudies/ecoin brownsville.html>.

Sustainable Gladstone. 2005. <www.sustainablegladstone.com/by products.htm>. Accessed March2005.

Symbiosis Institute. 2003. <www.symbiosis.dk>. Ac-cessed May 2003.

Tagliaferri, P. 2005. Economic Development Office,Trenton, NJ. Personal correspondence, March.

USPCSD (U.S. President’s Council on SustainableDevelopment). 1996. Sustainable America: A newconsensus for prosperity, opportunity, and a healthyenvironment for the future. Washington, D.C: U.S.Government Printing Office. February.

USPCSD. 1997. Eco-industrial park workshop pro-ceedings: October 17–18, 1996, Cape Charles,Virginia. <http://clinton2.nara.gov/PCSD/Publications/Eco Workshop.html#v>.

Van Beers, D., G. Corder, A. Bossilkov, and R. vanBerkel. 2007. Industrial symbiosis in the Aus-tralian minerals industry: The cases of Kwinanaand Gladstone. Journal of Industrial Ecology11(1): 55–72.

Van Berkel, R. 2004. Industrial symbiosis in Australia:An update on some developments and researchinitiatives. The Industrial Symbiosis Research Sym-posium at Yale: Advancing the Study of Industry andEnvironment. Yale School of Forestry & Environ-mental Studies, New Haven, CT.

Zhu, Q. and R. P. Cote. 2004. Integrating greensupply chain management into an embryonic


F O RU M

eco-industrial development: A case study of theGuitang Group. Journal of Cleaner Production12(8–10): 1025–1035.

Zhu, Q., E. Lowe, Y. Wei, and D. Barnes. 2007.Industrial symbiosis in China: A case study ofthe Guitang Group. Journal of Industrial Ecology11(1): 31–42.

About the Author

Marian R. Chertow is associate professor of indus-trial environmental management and the director ofthe Industrial Environmental Management Program atthe Yale University School of Forestry & Environmen-tal Studies in New Haven, Connecticut, USA.


“uncovering” industrial symbiosis · special feature on industrial symbiosis “uncovering”...

Documents