Integrative Leadership and the Creation and Maintenance of Cross-Sector

spongereasonInternet και Εφαρμογές Web

12 Νοε 2013 (πριν από 3 χρόνια και 8 μήνες)

93 εμφανίσεις



1

Integrative Leadership and the Creation
and Maintenance
of
Cross
-
Sector
Collaboration
s


By


Barbara C. Crosby

and

John M. Bryson


Abstract



This article
present
s

a theoretical framework for understanding integrative leadership and
the creation and mainte
nance of cross
-
sector collaborations

that create public value
.
We
define integrative leadership as bringing diverse groups and organizations together in
semi
-
permanent
ways


and
typically across
sector

boundaries



to
remedy complex
public problems and ac
hieve the common good.

Our framework
highlights

in particular

the leadership roles and activities of collaboration sponsors and champions. The
framework

is illustrated with examples from
the development of
MetroGIS
, a geographic
information system that pro
motes better public problem solving in the Minneapolis
-
St.
Paul region of the U.S.

A set of propositions is offered to guide further research and to
prompt reflective practice.






2


Integrative Leadership and the Creation
and Maintenance
of
Cross
-
sector
Co
llaborations


By


Barbara C. Crosby and John M. Bryson


M
any major public
problems or
challenges


such as global warming,

HIV/AIDS,
economic development, poverty, homelessness


can
be addressed effectively only if
many organizations

collaborate. Collabo
rators would include
governments certainly, but
often
m
ust

include
businesses, nonprofit organizations,
foundations, higher education
institutions,
and community groups as well
.
Leaders and managers in government
organizations thus face the need to inspire
, mobilize, and sustain their own agencies, but
also
to engage numerous other partners in their problem
-
solving efforts.


As we see it, this is the basic challenge of
integrative
public leadership



defined
as bringing diverse groups and organizations toge
ther in semi
-
permanent ways
,

and
typically across
sector

boundaries,
to remedy complex public problems and achieve the
common good. We have argued elsewhere that such problems are
often

due to the
characteristic failings of government, business, and civil
society and that sustainable
remedies must draw on the characteristic strengths of each sector

while overcoming or
minimizing their weaknesses

(Bryson and Crosby, 2008)
. In other words, the power to
adopt and
actually deliver

effective solutions is shared
among sectors and organizations
within the sectors. Integrative public leaders will have to lead across
sector

boundaries to
foster
the
requisite relationships and resource flows

needed to produce desirable
outcomes

Several analysts (
e.g.,
Cleveland,
2002
;

Crosby and Bryson, 2005) have provided
insights abo
ut leadership in this “
shared
-
power
,

no
-
one
-
wholly
-
in
-
charge world,”

an
increasingly apt descriptor in the early years of the 21
st

century.

Scholars also have made
headway in considering the implications
for government power
, authority,

and
responsibility in such a world.

What does it mean, they have asked, when so
-
called


3

“public” problems spill beyond government’s power and authority, yet citizens still look
to
democratic

governments to
help
solve them?

H
arlan Cleveland
(1977, 1993, 2002)
was
among those who a few decades ago
first
began popularizing the term “governance” to
describe arrangements (regimes) in which government bodies share power with other
types of organizations to create
significant achiev
ements of lasting public value (
Kettl,
2002
, 2009
;

Light, 2002
;

Osborne, 20
10
).

A substantial body of scholarship now describes how public administrators create
and manage collaborations among governments, businesses, and non
-
profits.

Indeed,
collaborative

public management has become a hot topic

(e.g.,
Goldsmith and Eggers
,
2004
;

Agranoff
, 2007;

B
ingham and O’Leary
,
2008
;

O’Leary and Bingham
,
2008
;

and
Kettl
,
2009). Much of this work
builds on a long
-
standing tradition of research into
public
-
private partn
erships and other cross
-
sector policy “tools” (Salamon, 2002
;
Osborne, 20
10
).
At the same time,
leadership language and scholarship
have
been

remarkably scarce
in the academic literature on collaboration
, although
the literature
typically
does highlight
im
plicitly

the
role
s

of
what we call sponsors and
champions.
Huxham and Vangen

(2005)

are an exception
in their
focus on leadership as enacted
through the

media


of people, processes and structures. We agree with them in part
,

but
see leaders as agents as w
ell as media.

In addition, a

National Academy of Public
Administration
-
sponsored

book (Morse, Buss, and Kinghorn, 2007) brings together a
number of scholars who discuss public leadership in collaborative settings. Chrislip and
Larson (1994)

and

Chrislip (2
002) offer guidance f
or collaborative civic leaders;
Linden
(2002) also describes qualities of government and nonprofit leaders engaged in cross
-
agency collaboration.


This
article

adds to the growing
attention to leadership

among scholars of public
admini
stration and governance
.
It also builds on the work of leadership scholars who are
increasingly describing the existence of
,

and increased need for
,

shared, collective, and
distributed leadership within organizations and networks (see Pearce and Conger, 20
03;
Uhl
-
Bien
, Marion, and McKelvey, 2007
;

Ospina and Foldy, this issue
).

The
article
present
s

a framework for understanding integrative leadership in
cross
-
sector
collaborative settings

in which government is
typically
an important actor
, but not the
only
actor
. The
starting point is
a widely cited cross
-
sector collaboration framework
and


4

set of propositions
developed by
Bryson, Crosby and Stone

(
2006)
. Th
at

framework
consists of five main elements: initial conditions, process, structure and governance,
con
tingencies and constraints, and outcomes and accountabilities. The
framework
addresses

factors affecting cross
-
sector collaboration in general, and
is not focused
specifically on leaders and leadership.

The revised framework presented here
acknowledges tha
t leaders and leadership are crucial in integrating all aspects of the
framework. Said differently, we argue that leadership work is central to the creation and
maintenance of cross
-
sector collaborations that advance the common good.


The

revised
framewor
k
builds on an extensive literature review, as well as
on

our
subsequent research into cross
-
sector collaborations
(Bryson, Crosby and Stone,
2007
),
particularly
in the areas of
employment

(Stone,
2007
), urban transportation
systems
(Bryson, Crosby and Sto
ne, 2008
; Bryson, et al., 2009
), and regional geographic
information system development (Bryson, Crosby and Bryson, 2009)
.
The revised
framework draws attention to crucial leadership work
related to
bridging processes and
structures, including: bridging ro
les
and
boundary spanning
activities
(Maguire, Hardy
and Lawrence, 2004),
the creation of
boundary experiences and boundary
groups and
organization
s

(Feldman, et al., 2006),
boundary object

creation and use

(
Carlile
, 2002,
2004; Kellogg, Orlikowski and Yat
es, 2006), and
the development of
nascent or proto
-
institutions (Lawrence, Hardy and Phillips, 2002)
.

In the revised framework leadership
work
clearly
is central.


Throughout this
article

we offer illustrations from the creation and
institutionalization of

MetroGIS, an award
-
winning geographic
information
system

(GIS)

initiative in the Twin Cities (Minneapolis


Saint Paul) metropolitan area of Minnesota,
USA.

M
etroGIS

is a completely voluntary
collaborative network of
over
300
governmental
units
, business
e
s
, and nonprofit organizations that has created a
mostly
virtual geographic information systems organization under the auspices of the
M
etropolitan
C
ouncil

(MC)
, the regional

government (
www.metrogis.org
).

N
ote that

the
case is used for illustrative purposes only; it
is not

a test of the framework. Indeed, the
flow was the other way


our study of MetroGIS
(Bryson, Crosby, and Bryson, 2009)
helped crystallize the revision of the
earlier

framework.
In the

article
’s
ne
xt

section we
briefly describe
geographic information systems

and their importance
, and

also
present a


5

thumbnail history of
the
MC

and
MetroGIS. In the
following

section we present the
revised framework. In doing so, we briefly recap the propositions drawn

from the
literature and presented previously.
M
ost space
, however,

is devoted
to presenting the
propositions related to the highlighted bridging process
es

and structures.
In
the

final
concluding section we emphasize theoretical, methodological, and practi
cal implicat
i
ons
of the revised framework.


Introduction

to
Geog
r
aphic Information Systems (
GIS
)
, the Metropolitan Council,

and
MetroGIS



Technological inno
vations in recent decades have produced powerful
web
-
based
geospatial
mapping tools that can help
a variety of groups solve problems and achieve
ambitious goals.
Users of the tools might be, for example, nonprofits or governments
seeking to combat public health problems or a business entrepreneur wanting to corner
new markets. Yet putting together a ma
pping system that draws on expertise and
databases of multiple organizations (at multiple levels of government and across sectors)
remains a challenging endeavor for leaders like those who were central to the creation
and continuation of MetroGIS under the

sponsorship of the Twin Cities Metropolitan
Council.



Geo
graphic Information Systems


Maps
, of course,

have been

used
through
out

human history to visually represent
geographic space
, the elements making it up, and the relationships among the

elements
.
Th
e creation of an analogical space representing a larger geography is one of the great
accomplishments of human history, on par with
the development of

language and
numeracy (
Robinson, 1982
). Maps are crucial to knowing where anything is and to
navigating b
etween points;
to assertions of sovereignty and the rights and duties of those
under the sovereign power; to understanding amounts, capacities and/or flows of various
things
(land, water, weather,
traffic
);
to

e
stablishing ownership and the legitimacy of r
eal
property exchanges; and
to
a host of other purposes.
Maps typ
ically are two
-
dimensional,


6

but
represent three
-
dimensional spaces
. T
hey also can be three
-
dimensional, as in globes
;
or four
-
dimensional via time
-
lapsed presentations.
1

Since the 1960s, it h
as become possible to produce digitized geospatial
information in order to create computerized maps (models) and to format, reformat, and
analyze them using various analytic tools. Geographic information systems (GIS) are
computerized models containing dig
itized, manipulable,
and
geospatially referenced data.
In principle, with a GIS

you can study not just this map or that map, but every possible map. With t
he
right data, you can see what
ever you want


land, elevation, climate zones,
forests
,

political bo
undaries, population density, per capita income, land use,
energy
consumption
, mineral resources, and a thousand other things


in whatever
part of the world interests you (Ormsby, et al., 2004, p. 2).


In a GIS, the maps are made up of layers (think of th
e zoom feature on Google
Earth). Each layer consists of features (cities, jurisdictions
, tracts of land
) and/or surfaces
(lake
s, land uses,
snow
cover).
Each geographic object in a layer is called a feature, but
not all layers contain features (e.g., the o
cean layer may just be a single expanse which
change
s

from place to place in terms of depth).
The features have shapes

and sizes
, while
the surfaces have values (
elevation, slope,

temperature, depth). Features have specific
locations identified by coordina
te systems and can also be displayed at different sizes
(scales)

(Ormsby, et al., 2004, pp. 2


10).
Google Maps (
www.maps.google.com
) is the
best
-
known GIS. Each year it includes more and increasingly accurate d
ata, including
geospatially referenced video feeds. Automobiles increasingly feature onboard GIS
systems as standard equipment to assist with navigation; most include voice directions.


The M
etropolitan Council

(MC)



The Metropolitan Council

(MC)

was crea
ted in 1967 to be

the
regional
planning
and
coordinating agency
for the Minneapolis
-
St. Paul region of Minnesota
.

It

formally
sponsors MetroGIS and
has assumed primary responsibility

for the
system.

T
he Minneapolis
-
St.Paul region since the 1960s has experi
enced many of the
same problems as other metropolitan centers in the U.S. and other “developed” nations.

Integrative leaders in the region responded by creating over many years regional
government structures

especially the
MC



that increased the capacity

of local, state,


7

and federal governments to tackle regional public problems (Bryson and Crosby, 1992;
Metropolitan Council, 2007).

The c
ouncil works with local communities to
provide the
following services (
http://www.metrocouncil.org/about/about.htm
):



o
perating the region's
largest bus system



c
ollecting and treating wastewater



engaging communities and the
public

in planning for future growth



providing
forecasts

of the region's population and household growth



providing affordable
housing

opportunities for low
-

and moderate
-
income
individuals and families



providing planning, acquisitions and funding for a regional system of
parks and
trails



providing a
framework

for decisions and implementation for
regional systems

including aviation, transportation, parks and ope
n space, water quality and water
management.

The
MC
’s governing board consists of 17 members, 16 of whom represent a
geographic

district

and one

chair who serves at large. They are all appointed by and serve
at the pleasure of the governor
.
At present, t
he
MC

has staff of 3,700 and an annual
operating budget of about $700 million, 90 percent of which is funded by state
appropriat
ions and user fees such as wastewater treatment charges and transit fares.
Ten

percent comes from local property taxes. The bulk of the
MC
’s employees operate the
region’s transit and regional wastewater treatment systems.

While the council had accomplish
ed many things since its establishment, b
y
the
1990s

regional officials and planners were still struggling to have timely,
accurate,
reliable and comparable geospatial information about local conditions so they could
:

understand the contours of transportat
ion, housing, open space, and waste treatment
challenges
;

generate solutions that were more finely tuned to local and regional realities
;

and build
the
coalition
s
needed
for necessary
policy changes

and resource allocation
choices
.
Said differently, i
n any

democratic society based on the rule of law, accurate,
timely, geospatially referenced information is absolutely necessary for effective
governance, planning, and coordination
; the MC had for years produced information, but
it was often based on imprecise

estimates and projections
.



8

MetroGIS grew out of the efforts of a
group of public officials and managers,
along with partners in other sectors
,

to
remedy this shortcoming. They sought to
create a
shared GIS for the region that linked
and made easily acces
sible
business
,
government

and nonprofit databases
of accurate, timely, standardized, and needed information;
and
acquired or developed the software applications to make use of the data to solve public
problems
. These leaders

practiced integrative leadersh
ip as

the
y

strove to improve
multiple governments’ capacity for public problem
-
solving
around a host of issues
affecting the Twin Cities metropolitan region, including
urban traffic congestion,
economic development, affordable housing,
threats to water
ava
ilability and
quality
,
provision of parks and other recreational opportunities, waste management, and
crime
.


Regional capacity building may

be seen as more a management than
a
leadership
challenge. Yet, government structures and tools often are simply ina
dequate to allow
government agencies to carry out responsibilities and partner effectively with other
organizations

(Kettl, 2009; Osborne, 20
10
)
. Developing these structures and tools

can be
a major integrative leadership challenge



and certainly was in t
he MetroGIS case.


M
etroGIS

T
his

article
’s
illustrations
trace efforts of
MC

administrators and appointed
officials, along with several county commissioners

and others
, to develop a sustainable
cross
-
governmental, cross
-
sector system for sharing detailed
geographic information (for
example, exact location of land parcels, streets, sewer and utility lines)

across numerous
jurisdictional boundaries
.
MetroGIS

is
now
14
year
s
old

and involves
300
governmental
units,
businesses
,

and nonprofit organizations

(www
.metrogis.org)
.
The organization’s
small coordinating staff is housed in the
MC
. Its policy board consists exclusive
ly

of
government representatives, but its management
-
level coordinating committee and
technical advisory

team
consist of members representin
g a variety of units of
government, businesses, and nonprofits.

The data on which the
s
e

illustration
s

are

based come from several sources:
archival research, including a review of materials on the MetroGIS website;
an

unpublished written history of MetroG
IS; individual interviews with ten leaders involved
in MetroGIS’s founding and subsequent development; one group interview with five


9

knowledgeable MetroGIS leaders; and participant observation by the second author in the
build
-
up to, facilitation of, and f
ollow
-
up to both major MetroGIS strategic planning
efforts.
2

MetroGIS

is
now nationall
y

and
internationally recognized as one of the best GIS

organizations in the world
. Its accomplishments include
, among other

thing
s

(
http://www.metrogis.org/about/accomplishments/index.shtml
)
:



Implementing, or making substantial progress on implementing,
regional solutions

for
nine of the Met
roGIS community's thirteen priority information needs: jurisdictional
boundaries; street addresses/where people live; parcels/parcel identifiers; highway
and road networks; census boundaries; lakes, wetlands, water courses; land cover
;

and planned land use
.



Implementing
MetroGIS DataFinder

as a registered node of the Federal National
Spatial Data Infrastructure, fully integrated into the State of Minnesota's
GIS,
with a
state
-
of
-
the
-
art do
wnloading capability (DataFinderCafé). Over 200 datasets are
currently accessible via DataFinder. Over 800 data downloads per month
occur

and

the trend
is

a steady increase.



Implementing
(
in conjunction with the State of Minnesota’s Land Management
Inform
ation
Center
[
LMIC
])

GeoService Finder
, a resource for finding geospatial
applications and web services.



Execut
ing

agreements

that provide access by all government interests serving the
seven
-
county
metropolitan
area, without fee and subject to identical access
requirements, to parcel and other geospatial data produced by all s
even metro area
counties and the
MC
.



Rec
eiving
several state, national and international awards for innovation.



Maintain
ing
active involvement of key stakeholder representatives at the policy,
management, and technical levels since MetroGIS's inception in
1995.



The

road to these achievements has been a long and not necessarily easy one.
The
journey began when
Rick Gelbman
n
, an MC manager,

concluded
in the mid
-
1990s
that
the time had come to improve the system the
MC

used to develop projections about the
f
uture development of Twin Cities communities. Local planners like Randall Johnson of


10

Shoreview (a St. Paul suburb) were complaining that
the
projection
system’s
methods
were not sufficiently based on local data and as a result the council’s projections
wer
e
unreliable
. Meanwhile, new technology for more precise local data gathering was
becoming available.
Gelbmann

persuaded
Richard Johnson

(no relation to Randall
Johnson)
, deputy
administrator

of the
MC
,
to seek approval from

the council

members
for creatio
n of a small internal unit to
explore the possibility of
creating a multi
-
county
GIS
.
The unit was to undertake the necessary planning to define what a metro GIS would
include, how it would operate, and how it would be financed.

Randall
Johnson was hired
to head the unit and in
1995
he and his staff organized

two regional forums that involved
county officials, GIS experts,
MC

staff
, and
representatives of cities, school districts, water management organizations, the regional
mosquito control district, park

boards, and
the
Metropolitan Airport Commission
. The
forums produced consensus that local governments were ready to cooperate with the
council in setting up and maintaining a
geographic information
system.
Randall

Johnson
soon became a determined champion

of developing a system that was technically
advanced, met local and regional government needs
,

and
helped the region become an
example of how pooled geospatial information could be used to foster wise development
and improve public services.



In
December

1995, Johnson and his staff worked with faculty at the Humphrey
Institute of Public Affairs (including
this paper’s second author
) to conduct a Strategic
Planning Forum that included 18 representatives of government, nonprofit and business
organizations.
The
author

facilitated a
strategy mapping

session
(Bryson et al., 2004)
that
helped the group agree on a set of strategic issues for the GIS effort
, elements of a
mission statement,

a statement of intent to pursue creation of a shared system
, and a set of
what came to be called “
strategic

projects
.


Interviewees agree
d

that the strategy
mapping session was a signal event in the creation of MetroGIS.

In April 1996, a formal mission statement,
goals, guiding principles,
five
strategic

projects

(called Strateg
ic Initiatives)
, and an initial organiz
ational structure were created
and agreed by key stakeholders, including the
formally established
Coordinating
Committee and the MC. In other words, what had been several strategic “matters of
concern” involving direc
tion, organization, and governance evolved into “matters of fact”


11

(Latour, 2005
, p. 22
).
The original
, officially approved
mission of MetroGIS grew
directly out of
(meaning much of the language was taken directly from) the

strategy
mapping exercise
. The or
iginal mission was: “
To provide an ongoing, stakeholder
-
governed, metro
-
wide mechanism through which participants easily and equitably share
geographically referenced data that are accurate, current, secure, of common benefit and
readily usable. The desire
d outcomes of MetroGIS include:

i
mproved participant
operations
; r
educed costs
; and s
upport for cross
-
jurisdictional decision making


(
www.metrogis.org
/about/history/mission.shtml).


Many persons interviewed

for thi
s
article
commented on the importance of the
guiding principles for developing and sustaining the organization across many different
kinds of boundaries. Many noted the principles are frequently referred to, and are clearly


even
emphatically



inclusive,

participatory and democratic.
Interviewees

also
emphasized the importance of tapping and creating shared knowledge and understanding.
The principles are as follows (slightly modified in recent years from the original)
(
http://www.metrogis.org/about/index.
shtml#principles
):



Pursue collaborative, efficient solutions of greatest importance to the region when
choosing among options.



Ensure that actively involved policy makers set policy direction.



Pursue comprehensive and sustainable solutions that coordinate
and leverage
resources: i.e., build once, make available for use by many.



Acknowledge that the term “stakeholder” has multiple participation characteristics:
contributor of resources, consumer of the services, active knowledge sharer, potential
future con
tributor, potential future user, continuous participant, and infrequent
participant.



Acknowledge that funding is not the only way to contribute: data, equipment and
people are also valuable partnership assets.



Rely upon voluntary compliance for all aspects

of participation.



Rely upon a consensus
-
based process for making decisions critical to sustainability.



Ensure that all relevant and affected perspectives are involved in the exploration of
needs and options.



12



Enlist champions with diverse perspectives whe
n implementing policies and carrying
out activities.

The
key stakeholders

also established the M
etroGIS Coordinating Committee
that
would
guide the system’s development.
The Coordinating C
ommittee
agreed to
undertake five design projects:



eliciting endorse
ments from stakeholders



executing data and cost
-
sharing agreements



implementing an Internet
-
based data search and retrieval tool



identifying common information needs



creating a business plan and organization structure.


By the end of 1996,
the group had ob
tained substantial stakeholder buy
-
in. Eleven
key
stakeholder organizations had adopted resolutions endorsing creation of MetroGIS
and designated a representative for the MetroGIS Policy Board that
was
created

to
set
policies

for the system.
The policy boa
rd consists
of public officials
from the
MC

and
counties, as well as associations of local governments (municipalities,

school districts,
watershed districts).
Victoria Reinhardt, a member of the Ramsey County Board of
Commissioners, became chair. (St. Pau
l is in Ramsey County.)
The
C
oordinating
C
ommittee became more formalized,
and

David Arbeit, director of the state’s Land
Management
Information
Center (LM
I
C),
was
elected
as
chair. Other members,
including Will Craig, a GIS expert at the University of Min
nesota,
provided early and
continuing leadership within the committee
.

A Technical Advisory
Team

also was
established.
Staff secured preliminary data
-
sharing agreements with representatives of
metro area counties and other data producers, and they began ef
forts to identify common
information needs.

During the next four years the MetroGIS staff, the technical advisers,
the
Coordinating Committee,
and the
Policy B
oard worked hard to put together more
permanent data and cost
-
sharing agreements and
to
establis
h rules and standards for data
collection and dissemination. An important product was the DataFinder Internet tool,
which provided a quick data search and retrieval mechanism for MetroGIS partners.


13

When a University of Minnesota team assessed the benefits
of MetroGIS during t
his
period, it found that the system had given stakeholders easier access to useful data,
fostered better communication among GIS users across the region, and increased trust in
sharing geographic information with the user community

(
Cr
aig & Bitner
,
1999)
.



The
MC

hired a consulting firm to produce a business plan for sustaining the
system through 2003
. The plan, which built on the findings of the
u
niversity assessment,
was approved by the Policy B
oa
rd and the
MC
; it identified priority

functions, projected
costs, and outlined long
-
term funding issues.
By 2001, MetroGIS was attracting favorable
attention from national GIS advocates and policy makers. It received
a federal grant

to
improve th
e DataFinder website. Victoria
Reinhardt

was ap
pointed to the board of the
National GeoData Alliance created by the
U.S.
Department of
the
Interior
.

During the next six years, the project leaders focused on improving and
further
institutionalizing the system.

To improve the ability of the DataFinder t
o respond to user
needs, a consultant team solicited input from users, state agencies, and the US Geological
Survey.

The expanded tool
allowed

much more customized usage. A working g
roup of
MetroGIS stakeholders and a consultant also developed a performanc
e measurement
program.

Additionally
MetroGIS
came up with

“endorsed regional solutions” for
the
13
highest priority i
nformation needs of the region and

established

mechanisms for quick
distribution of relevant data during emergenc
y and homeland security ev
ents.

D
uring this period, MetroGIS had to survive a major challenge from within the
MC to its continued existence via
the

program evaluation audit. According to MetroGIS
Policy Board Chair Virginia Reinhardt, the organization had been called on to justify
its
existence as a public entity by MC decision makers; when a program audit found that
MetroGIS produced benefits far in excess of its costs, the challenge disappeared (Virginia
Reinhardt interview 2008). In June 2006
,

the MC endorsed MetroGIS and guarant
eed its
continued existence.

By 2006, the system had won national and international recognition for its
accomplishments. Its leaders recognized that now that the system was on a solid footing,
they should plan for the next five years. They convened stakeho
lders once again to
consider new technological opportunities and growth dynamics for the region. In 2007,
system managers and policy makers participated in a Strategic Directions Workshop
,


14

again

facilitated by this paper’s
second

author
,

in order to contin
ue providing public
value and be a leader in using geographic information systems to meet public needs.
An
expanded mission for the system

and new goals and strategies
emerged from the
workshop

(and from further consultations)
.

The new mission is:
“…
to exp
and
stakeholders' capacity to address shared geographic information technology needs and
maximize investments in existing resources through widespread
collaboration of
organizations that serve the Twin Cities

Metropolitan Area” (
www.metrogis.org
). T
he
new mission represents a significant change from the previous mission and helps
MetroGIS “go to the next level” (Randall Johnson, personal communication 2007; Group
interview 2008). Previously

the purpose of the organiz
ation was to create a mechanism
for sharing GIS information. The new mission states the purpose is to expand
stakeholders’ capacities to address GIS needs, maximize investments in existing
resources, and foster widespread collaboration of organizations


n
ot just governments


that serve the metropolitan area.
The interviewees agree
d

that t
he mapping process
helped clarify for key MetroGIS stakeholders that the organization had outgrown its
previous mission.
In other words, mapping again made a difference


and was in essence
an

actor
” in its own right (Latour, 2005, p. 71)
.


The mapping exercise
also
resulted in a focus on eight major activity areas for the
next three to five years, beginning in 2008.

The activities are as follows
:




Develop and maintain re
gional data solutions to shared information needs



Expand endorsed regional solutions to include support and development of application
services



Facilitate better data sharing
through making more data available, having more uses, and
improving processes




Pr
omote a forum for knowledge sharing



B
uild advocacy and awareness of the benefits of collaborative solutions to shared needs



Expand MetroGIS stakeholders



Maintain funding policies that get the most efficient and effective use out of available
resources and

revenue for system
-
wide benefit




15

While collaboration may be necessary or desirable, the research evidence
indica
tes that it is hardly easy (Agra
noff,

2008; O’Leary and Bingham, 2009
)
. The
MetroGIS story fits this pattern
. The organization
has survived se
rious challenges over
the
course of its history, including particularly
the

concerted effort
noted above
by some
within the
MC

to eliminate it
. They argued that the GIS service should be

provided by the
business

sector, or not at all.

A
dditional controvers
ies have not yet been settled. Some
MetroGIS founders have moved on; other
s

may soon, so the issue
of

leadership
transitions

is ongoing
. In addition, cities are big users of data and applications, but are not
major providers of them, so maintaining an equi
table balance of benefits and
contributions

remains an important issue
.
F
inally, one member county appears to be
ambivalent about
supporting
development of applications and
allowing
participation by
businesses.


A Framework for Understanding Leadership and

the Creation
and Maintenance of
Cross
-
Sector
Collaborations




In

this section each part of the
framework
(see Figure 1)
will be described
theoretically and summary propositions will be offered, with illustrations as appropriate
from the MetroGIS case. Th
e
parts

are: initial conditions, processes

and practices
,
structure and governance, contingencies and constraints, and outcomes and
accountabilities.
Integrative leadership involves leading across boundaries at individual,
group, organizational, and broade
r levels. In this case, Randall Johnson, Richard Johnson,
Victoria Reinhardt, Will Craig and other leaders had to work with their partners across
individual, intra
-
organizational, inter
-
organizational and sector boundaries to create and
maintain MetroGIS

a
s

a sustainable cross
-
sector collaboration that can help meet
important public needs and advance the common good.
To repeat, the illustrations are
meant simply to ground the propositions and are not presented as part of a test o
f

the
propositions.
Other
vi
ews and examples will also be presented as appropriate.

Insert Figure 1 About Here

In the frame
work
, initial conditions are thought to affect directly both processes
and practices and structure and governance. Processes and practices
, on the one hand,
and



16

structure and governance
, on the other hand,

are

seen as
intimately entwined
; as
also
affected by contingencies and constraints
; and as also
directly affect
ing

outcomes and
accountabilities.
We

also posit

a

direct
effect
of initial conditions on outcomes
and
accountabilities. The
two
-
way
linkage between processes and practices
, on the one hand,

and structures and governance
, on the other,

represents

insights from structuration theory
about how practices and structures are created and recreated through acti
on (Giddens,
1979, 1984;
Feldman, et al., 2006; Pentland and Feldman, 2008;
Orlikowski, 2009
)
.

The framework also incorporates insights from actor
-
network theory
(ANT)
(Latour, 2005
;
Law and
Hasard, 1999
). Latour

asserts that ANT takes as its challenge
ac
counting for new associations


i.e., “the
tracing of associations
” (
2005, p.
5, italics in
original)


without
a priori

assuming any fixity to social aggregations. In other words, the
“social” (
e.g.,
existing and new
data
-
sharing
networks, communication p
atterns,
stakeholder relations, coalitions,
collaborations
) is what must be explained, not assumed.
Further, tracing associations also means accounting for connections among “things that
are not themselves social


(loc. cit.); for example,
strategy maps,
p
lans,
integrated
computer networks, and worksites.
“Association” includes far more than, for example, a
communication l
ink between nodes in a network. A
ssociations may also
be

shared
understandings, affective responses, identity
-
based or
-
forming linkages,

agreements,
commitments, resource flows, and host of other possible connections, including causal
connections.

In the
article
’s concluding section, we will say more about the aptness of
structuration theory and ANT for studying leadership and cross
-
sector

collaboration.


Initial conditions


At the outset of a change effort
,

initiators
confront
the challenges of
adapting
leadership
to

context and
making use of the talents of particular kinds of leaders to push
the action forward
(Crosby and Bryson, 2005). T
hat is, they must pay attention to
contextual forces that affect the change effort and they need to understand the people
(including themselves) who bring assets and liabilities to the leadership work.
They must
seek sponsors
of and
champions

for the chang
e effort
.

In terms of context, s
tudies of cross
-
sector collaboration have emphasized the
importance of
system turbulence in general
,

and
institutional and competitive forces

in


17

particular
,

in launching successful collaboration (
Sharfman, Gray, and Yan, 199
1;
Bryson, Crosby, and Stone, 2006).

Changing conditions provide a perturbation that offers
opportunity for
the
system to re
-
organize (
Burns, 1978;
Senge
1990
; Allen and Cherrey,
200
0
; Uhl
-
Bien,

Marion, and McKelvey, 2007)
.
Particularly important are insti
tutional
arrangements


political, economic, and social


that can aid or hinder collaboration.
Effective integrative leaders must take
account of
these institutional arrangements, but
also recognize that changing conditions within and outside the institut
ions can offer
opportunities for new institutional arrangements.
Leaders must demonstrate a kind of

systems thinking
” (Senge, 199
0
; Senge, et al., 2008
)

in order to understand the
turbulence as well as the driving and constraining forces
. S
ystem
s

thinking

also
involves
seeing
existing flows of information and other resources among relevant organization
s
,
and
noting where
desirable
flows are
negatively
constricted by
intra
-
organizational
, inter
-
organizational

and
sector

rules and
boundaries. A sense of link
s and gaps can help leaders
think about who and what must be integrated
(
and
perhaps
dis
-
integrated
)
.
Randall
Johnson and Rick Gelbmann demonstrated systems thinking in their ability to see
connections among local government information needs

and emerging
technologies (the
I
nternet and GIS) to address them.

In a related vein,
stakeholder identification

and
analysis
efforts
are also important
in order to understand the political contours of the
context
:
who
has information and other resources (authority, tec
hnical expertise and
commitment or enthusiasm
)

for tackling a public problem or championing a solution
, and
the
what
avenues for change
are likely to
garner sufficient stakeholder support (
Luke,
1998;
Bryson, 2004)
.
Resources may show up

in unlikely places
, particular
ly

in sectors
with which one has least familiarity.

In this case, political institutions offered the most important constraints and
drivers. The fragmentation of local governments was definitely a constraint on efforts to
build a unified
geogra
phic
information system
for the metropolitan area. At the same
time, the
MC

had been established in the 1960s precisely to deal with problems that
spilled across local government borders and affected the entire region; it had
responsibility for planning ma
jor regional functions such as open space,

transportation,
and sewer service. The council also had its own taxi
ng authority. Still, it had

relatively

little power to force local governments to follow

its guidance. Thus as Richa
rd Johnson


18

and GIS liaison Ra
ndall

Johnson
launched

the
Metro
GIS project, they knew that
obtaining local government buy
-
in was a must.


Several forces drove the collaboration in this case


population growth in the
metro area, increasing needs of local government for information abou
t existing and
future development
as well as public services, and the availability of
increasingly
sophisticated GIS technology. In addition to local governments, utility companies
,
retailers and other businesses

wanted more reliable information about grow
th trends in
the region. These forces provided the “case” that metropolitan GIS advocates could use to
justify asking for
MC

funds and staff to launch the effort. Moreover, if the advocates
could convince local governments that the potential GIS system wou
ld be carefully
designed to meet their needs, the basis for buy
-
in would be there.

Significantly, there did
not appear to be any potential competitors, particularly from the
business
sector.

Proposition 1:

Like all inter
-
organizational relationships, cross
-
sector
collaborations are more likely to form in turbulent environments.

Leaders
will have more success at launching these collaborations when they take
advantage of opportuniti
es opened up by driving forces
(including helping
create or
favorably alter
ing

them),
while remaining attuned to constraining
forces
.



Cross
-
sector collaborations also appear to be fostered by recognition that no one
sector can solve an important public problem on its own

(
Bozeman
, 2007
;
Bryson and
Crosby, 2008)
. In this case, neit
her the
MC

nor local governments had enough GIS
expertise and money to establish their own
full
-
fledged
systems.
Local government
planners were increasingly frustrated by
the
inaccuracy of
MC

projections and the
barriers to exchanging information with each

other.
Utility companies
and other
businesses
were in a similar situation
,

and GIS technology firms did not have the
authority, resources, or credibility to develop a unified regional system.
Faculty and
technical experts at the University of Minnesota ha
d crucial skills for putting together
such a system, but also lacked needed authority and funding. Leaders of the GIS effort
recognized the need to involve stakeholders from multiple sectors to provide the
necessary combination of skills, money, and expert
ise.

Proposition 2:

L
eaders

are most likely to try cross
-
sector collaboration if they
believe that separate efforts by several sectors to address a public problem have
failed and the actual failures cannot be fixed by a separate sector alone.



19



Linking m
echanisms


such as existing
cross
-
boundary groups or organizations,
and general agreement on the problem to be solved


also contribute to the formation of
cross
-
sector collaboration (Bryson, Crosby, and Stone, 2006). In this case, local
governments,
MC

s
taff
,
and
university faculty had worked together in the past

on a
number of tasks. At the university, geography and planning faculty in particular had
played a strong role in making the case for regional governance, planning
,

and geospatial
information gat
hering. Of course, the
relations between local government
and the
MC

weren’t always smooth, since local governments sometimes viewed the
MC

as attempting
to impose its will on them.
Nevertheless
the GIS advocates were able
at the outset
to
assemble a small

group of local government partners who did share a general agreement
that they needed the ability to gather
data and information

systematically across the
region.

One GIS partner commented, “Many of us felt a common need, so it was easier to
reach a commo
n vision.”

In spite of the common rhetorical emphasis on public
-
private partnerships, the
literature seems to indicate that
sector

boundaries between governments and nonprofits
(including academic institutions) may be easier to bridge tha
n

those
between go
vernment
and business

(Salamon, 2002)
. Part of
the
difficulty
is
inherent in the popular label

public
” sector,
which is
usually applied to government
, but also
occasionally to
nonprofits
,
and its contrast, “private sector.” The contrast presumes conflict
as much as it
might imply the need for cooperation to solve public problems.
T
he deeper challenge
appears to be the competing “institutional logics” between government and business
(
Friedland and Alford, 1991
)
. Though Randall Johnson has been fairly succes
sful in
involving technical experts from for
-
profit enterprises,
the
Policy Board has been
somewhat stymied about how to give businesses full
-
fledged access to
the
system.

An important part of the analysis of initial conditions should include
look
ing

for
i
ndividuals who can exercise personal leadership



that is, deploying personal assets
and
structural position
to lead change efforts.
Major change
efforts are unlikely to get off the
ground
and pro
sper unless at least
one committed champion and one committe
d sponsor

support them (Crosby and Bryson, 2005). By
champion

we mean a person who is a
tireless, process
-
savvy organizer and promoter of the change effort;
in contrast,
a
sponsor



20

is less involved in the process
,

but deploys authority, money, or connection
s to move the
change effort forward.

For example
, Richard Johnson deployed his formal authority to
gain
MC

members


support for
the project.
Randall

Johnson
played a champion role

by
bringing
planning expertise
, political savvy,

and enthusiasm to the proje
ct.

Both sponsors and champions will need to see

the problem as significant
and
capable of solution or remedy
, b
ut they should not be too wedded to specific problem
definitions or solutions or they
may not be able to engage in the collaborative problem
-
so
lving process
requir
ed to
enroll

diverse stakeholders

in the effort
and set directions
likely to result in widely beneficial changes
(Nutt, 2002
; Huxham and Vangen, 2005
)
.

Champions (or a set of champions as a group) are especially likely to need to be mul
ti
-
lingual translators
as
they seek to bridge differing organizational and practice cultures

(Carlile, 2004)
, and
will need
to be seen as legitimate
across camps
(
Maguire,
Hardy,
and
Lawrence, 2004
)
.

Randall Johnson
had been a local planner and critic of
M
C

projections,
so he
could speak the language of local planners

and be seen as a legitimate champion of
better regional solutions

that addressed local needs and concerns.

In short, he could work
well with policy makers, planners, and technical personnel.
W
ill C
raig
brought
legitimacy as
an academic
GIS
expert at the
u
niversity who was able to translate

technical

GIS
-
speak
” to
more general audiences.

Other champions also played important
translator roles across various boundaries.


In some ways this case re
presents a classic case of policy entrepreneurs taking
advantage of a window of opportunity (Kingdon, 1995; Roberts and King, 1996; Crosby
and Bryson, 2005
)
. Demand for change was spreading among local governments,
technological advances made change attrac
tive, and the right people were in positions
with the authority to provide needed approvals and resources, which they did.

Proposition 3:

Cross
-
sector collaborations are more likely to succeed when one
or more linking mechanisms, such as powerful sponsors

and champions
, general
agreement on the problem, or existing networks are in place at the time of their
initial formation.



Integrative process
es and practices



A
process

is a

series of linked actions or proceedings
. A
practice

incorporates
process, but

is also a contextually situated, socially accomplished flow of organizational


21

acti
on
. Practices may be seen as “
a flow of organizational ac
tivity that incorporates
content and

process, intent and emergence, thinking and acting, and so on, as reciprocal,
i
ntertwined and frequently indistinguishable parts of a whole when they are observed at
close range (
Jarzabkowski, 2005
, pp. 7


8
, 11
).
I
mportant
i
ntegrative process
es

and
practices include: wise design and use of forums, arenas, and courts, including crea
ting
effective
boundary
-
spanning groups,
boundary experiences
,

and objects; building
leadership capacity; forging agreements;
building trust; and
building legitimacy.

Wise design and use of forums, arenas, and courts
.

Forums, arenas, and courts
are the ch
aracteristic setting
s

we humans use to creat
e

and communicate meaning (
in
forums), make
and implement
decision
s

(
in
arenas), and enforce
principles, laws, and
norms (
in
courts) in shared
-
power s
ituation
s

where no one
person
or group is fully in
charge (Cro
sby and Bryson, 2005
, pp. 401
-
426
).
N
o one is wholly in charge in forums,
arenas, and courts, either, since they involve shar
ing

power
, albeit
power
that
is hardly
ever shared equally (Flyvbjerg, 1998)
. Nonetheless, the
se

settings have a
huge

impact on
act
ion, because the ideas, rules, modes, media, or methods
that are design features of

the
settings serve to divide what is
conceivable

in terms of a set of potential decisions, issues,
conflicts, and policy preferences, into those that are
actually observed
,

and those that
are
sent off into a kind of public policy

never
-
never land


as non
-
decisions and non iss
ues,
latent or covert conflict, and unsupported policy preferences (Bachrach an
d Baratz, 1962,
1963; Luke, 1974; Flyvbjerg, 1998
).
For example, if your

issue or decision item does not
get on the agenda of a formal arena, the issue becomes a non
-
issue and the decision
becomes a non
-
decision, at least for the present.
As
practices, the three settings make it
possible for actors to draw on collective, struc
turally

based
rules and resources to
produce
actions that (typically epiphenomenally)
recreate the
underlying
rules and resources,
albeit often in modified form. In keeping with s
tructuration theory
,
practices and
structures are created and recreated throu
gh action.
S
tructures are only “instantiated”
through action and otherwise do not exist; there is thus a “duality” to structures as both
the product of action and the bas
is

for further action

(Giddens, 1984
).


Our previous research

indicates that the desig
n and use of forums in which shared
meaning is created
may be

more important than

the design and use of arenas or courts

when it comes to

public problem
-
solving
,

because
the design and use of forums



22

determines what is even considered a public problem, what

solutions are viable, and what
public programs, projects, and policies are discussed by policy makers (
Crosby and
Bryson, 2005)
. The creation and communication of shared meaning may be seen as the
primary work of visionary leadership.
Integrative leaders
will need to help stakeholders
within single organizations and from
different organizations and sectors develop at least
partially shared meanings and understandings of past and present conditions and a
desirable future


in other words, a shared vision


if
they are to agree on and implement
new projects, programs, and policies that advance the common good (
Kouzes and Posner,
2007; Morse, this issue
).

We discuss below how wise design and use of forums in
different phases of a change effort provide
cross
-
bo
undary groups with
cross
-
boundary
experiences typically involving the creation and use of boundary objects.


Cross
-
b
oundary groups
(
boundary groups
for short)
are “collections of actors who
are drawn together from different ways of knowing or bases of exp
erience for the purpose
of coproducing [cross
-
]boundary actions” (Feldman, et al., 2006, p. 95). Examples
include cross
-
boundary networks, task forces, and teams; coordinating committees; and
representative policy making bodies. MetroGIS is perhaps best un
derstood as a
set of
practices involving
boundary group
s
. In other words, as a completely voluntary assembly
of over three hundred member organizations who, in theory, can walk away whenever
they wish, MetroGIS is essentially
a “virtual organization” (Rand
all Johnson, interview,
2007
; group interview, 2008)
. Members

stay
,

in essence
,

because the benefits of
belonging to the cross
-
boundary groups exceed the costs.
The Policy Board, Coordinating
Committee, and Technical Advisory
Team
are all

boundary groups,
operating mostly as
forums, with each member representing a different organization.
As they become
formalized, structured and institutionalized, boundary groups become cross
-
boundary
organizations
, which
MetroGIS has essentially become, and in that regard
belongs in the
structure category discussed below.

Adeptly designed f
orums allow boundary groups to have b
oundary experiences
,
defined as
“shared or joint activities that create a sense of community and an ability to
transcend boundaries among participants
” (Feldman, et al., 2006, p. 94; Feldman and
Khademian, 2007).
Boundary experiences are important for helping participants develop
a shared perspective
that
they then can act upon (
Boland and Tenk
as
i, 1995
).
MetroGIS


23

was forged out of many such experiences

involving diverse participants and elements and
continues because of the self
-
conscious design and use of such experiences.

Indeed,

cross
-
boundary stakeholder involvement has been a hallmark of MetroGIS forums
.

Johnson’s office has reached out to several
businesses, including the major retailer
Target, land planning and engineering firms, software companies, and The Lawrence
Group (consultants that handle street data). Commenting on some of the forums, a
participant said, the “sessions really helped me to
understand other people’s concerns and
needs and many of the major issues.”

Boundary objects are
typically important in helping people create shared meaning

(Star and Griesemer, 1989)
. Boundary objects are
“physical objects that enable people to
understan
d other perspectives”
(Feldman, et al., 2006, p. 95).
Beyond that, boundary
objects can facilitate the transformation of diverse views into shared kno
wledge and
understanding (Carli
le, 2002
, 2004
). Of particular importance in the MetroGIS case is the
creat
ion and use of shared strategy maps (Bryson, et al., 2004) by key stakeholders

at the

two strategic p
lanning
f
orums
.
The strategy maps produced at both forums were view
ed

by interviewees as extremely important for the development of MetroGIS
.


D
esigning an
d using
arenas and courts also mattered in the MetroGIS case. Each
participating organization’s decision makers, which often meant a decision
-
making

body

(
i.e.,
arena)
, had to authorize participation.
The
MetroGIS Policy Board does establish
policy and emp
hasizes consensus
-
based decision making. The
MC

members have
periodically had to formally endorse MetroGIS and allocate resources to the effort. By
emphasizing consensus
-
building and developing broad support, MetroGIS has sought to
make it easy for formal
decision makers to say yes when necessary.

Said differently,
t
hroughout the
creation and maintenance of MetroGIS
, integrative leaders needed to
achiev
e

desired decisions from policy
-
making arenas, such as the
MC
, county boards,
and city councils. Thus they

were careful to link forums to arenas
, so that consensus
-
based dialogue and deliberation preceded formal decision making, meaning that interests
were aligned in such a way that member organizations would continue to participate after
formal, authoritative

decisions were made
. Especially important was
the
inclusion of
local public officials
in forums, because they
ultimately
had to approve their city’s or
county’s participation in the data
-
sharing system.

The proposed system gained legitimacy


24

in the
se offic
ials’
eyes because they were helping to shape it and because they had a
chance to know the
MC

staff who would implement it. Victoria Reinhardt
,

as Policy
Board chair and Ramsey County

commissioner, and Randy Johnson
(no relation to
the
other
Johnson
s
)
,

as
Policy Board member and Hennepin County commissioner, were
politically savvy and powerful as representatives of the region’s most populous counties.
(Hennepin is the county that contains Minneapolis.)
Randall Johnson and Richard
Johnson also had honed thei
r understanding of how to influence policy makers during
many years of experience in government agencies.

In terms of courts, MetroGIS obviously must abide by the law, and formal court
s
have reinforced
the MC’s
authority when it has been

challenge
d

in laws
uits (Bryson and
Crosby, 1992).

Perhaps t
he most important court

has been the informal court of public
opinion. The guiding principles mentioned earlier act as strong norms guiding and
sanctioning behavior in relation to MetroGIS. As noted, many interviewe
es commented
on the importance of the norms for guiding MetroGIS’ work and the interactions among
participants in that work.

A
nother important role for formal and informal courts is helping integ
rative
leaders resolve residual conflicts that remain after
policy makers have approved new
policies and projects

that must still be implemented. Integrative leaders

will need to
employ sanctions that reward those who act in accord with the new policies and penalize
those who don’t. In th
e MetroGIS
case, the sancti
oning process
appears to have been
fairly informal. If a true cost
-
sharing system had been introduced, some means of
penalizing those who didn’t pay would have been important. The court of public opinion
could be important in bringing pressure to bear on l
ocal governments that fail to
participate in the system. Sanctions that could be imposed by the formal courts
also
are
important. For example, Metro GIS designers
have had to make sure

that data
-
sharing did
not violate privacy rights that would be enforced

by formal courts.

Proposition
4
:

Cross
-
sector collaborations are more likely to succeed when
sponsors, champions, and other leaders pay careful attention to the wise design
and use of forums, arenas, and courts
, including

the creation of helpful boundary

groups, experiences, and objects.





25

Forging

agreement
s
.

Informal agreements about the collaboration’s composition,
mission, and process can work (Donahue, 2004), but formal agreements have the
advantage of supporting accountability. The need for different

types of initial agreements
and reworking of agreements are likely to increase as collaborations grow to include
more geographically dispersed partners and diverse actors within a problem domain
(Kastan, 2000).

Possible elements of formal agreements incl
ude: broad purpose, mandates,
commitment of resources, designation of formal leadership, description of
eligible
members, decision
-
making structure, and built
-
in flexibility (such as allowing waivers)
for dealing with local conditions and changes (Crosby

a
nd Bryson, 2005; Arino and de la
Torre, 1998; Page, 2004). When partners do not completely agree on a shared purpose,
they may be able to agree on next steps (Huxham and Vangen, 2005). Studies of
collaboration highlight the importance of a drafting proces
s

that is highly participatory
and
involves

key stakeholders and implementers (Page, 2004
,

and this issue
). Less
powerful partners may have more difficulty than others in advocating for their interests in
this process, though
leaders

can use several techniq
ues to equalize power (Crosby and
Bryson, 2005
; Ospina and Foldy, this issue
).

In
the MetroGIS

case,
there were a series of “initial” agreements.
Randall

Johnson
’s office organized two
regional forums in
1995

simply to explore the need for a
multi
-
county
GIS unit.
The inclusion of
the University’s
C
enter for Urban and Regional
Affairs a
nd the
State’s
Land Management
Information
Center
as forum sponsors added
legitimacy to the gatherings and communicated the importan
t

message tha
t this effort was
not just R
andall

Johnson’s or the
MC
’s initiative.

As noted earlier, the forums attracted
numerous local government officials and staff, but also
a broad array of
people from t
he
sponsoring organizations and other interested parties.
Johnson posed two main questions

to the participants: Should interested parties purs
u
e a regional data
-
sharing system?

Should the
MC

take the lead?

Participants said yes to both.
The result was that
Johnson’s
office had a general mandate to proceed with exploring creation of a metropolit
an GIS,
many more stakeholders knew about
its potential benefits,
and some local officials were
primed to support it.

Johnson emphasized that “asking for permission and getting buy in


26

at the beginning” were very important for laying the foundation for the
ultimate success
of the GIS system.

Proposition 5:

The form and content of a collaboration’s initial agreements, as
well as the processes
leaders
use to formulate them, will affect the outcomes of
the collaboration’s work.


Planning.

Two
contrasting

approa
ches to planning
have been associated with
successful collaboration
. One emphasizes deliberate, formal planning
that includes
c
areful articulation of mission, goals and objectives; roles and responsibilities; and phases
or s
teps, including implementation
(
Mattessich, Murray
-
Close, and Monsey, 2001). This

deliberate


approach
(
Min
tzberg, Ahlstrand, and Lampel
,

1998)
seems
most likely when
collaboration is mandated.
In the second
approach
,

a clear understanding of mission,
goals, roles, and action steps emer
ge
s

over time as conversations involving individuals,
groups and organizations grow to encompass a broader network of involved or affected
parties (Winer and Ray, 1994; Huxham and Vangen, 2005). This

emergent


approach
(
Mintzberg, Ahlstrand, and Lampel
,
19
98) seems most likely when
collaboration is not
mandated. Regardless of approach, c
areful attention to stakeholders clearly is crucial for
successful planning
(Bryson, 2004; Page, 2004), and t
he process should be used to build
trust and the capacity to man
age conflict effectively (Bryson, 2004). Planning is more
likely to be successful
if

it builds on the competencies and distinctive competencies of the
collaborators, including those arising from the distinctive sectors in which they operate
(Bryson, Ackerm
ann, and Eden, 2007).

The MetroGIS official history divides its
development into three main phases
(
www.metrogis.org/about/history/index.shtml
):
a

definition phase from 1995 to 1996, a
design phase from 1997 through 2001, and an implementation and transit
ion phase from
2001 to the present.
(
Note that even though MetroGIS put the implementation label on
the last phase, we should point out that some major implementation activities occurred
during the design phase, as is true of most strategy change efforts.
In other words,
implementation rarely waits until

all the planning has occurred [
Mintzberg, 1994;
Bryson, 2004
]
).
MetroGIS’
s
use of planning in the three phases exhibited characteristics
of deliberate as well as emergent planning, both of which seem to be
needed in successful
cross
-
sector collaborations (Bryson, Crosby and Stone, 2006). While major forums


27

usually involved emergent thinking


in other words, consensus on directions and next
steps emerged from the forum activities


the Coordinating Committee

did more
deliberate planning, for example, by adopting a mission statement and four goals for the
MetroGIS during the definition phase.


Early on,
after the two large regional forums
in 1995,
Richard Johnson

suppo
rted
Randall
Johnson
’s belief that the nex
t step should be a visioning and strategic planning
process
from which a possible vision, mission, and strategic options might emerge. The
exercise
would involve a wide array of potential GIS stakeholders.

So Johnson went
ahead in the face of some skeptici
sm from vocal participants in the regional forums.
The
first major event in this process

was the
December
1995 Strategic Planning Forum.

Participant numbers were at a level that easily permitted participation by all, and the
University of Minnesota facilit
ato
rs used a method called “
strategy

mapping,” to help the
group reach consensus on strategic issues and an initial agreement on how to keep
working together.
The group brainstormed and categorized a number of ideas in response
to three questions:

1.

What mi
ght a metropolitan GIS do?

2.

How might those things be done?

3.

And, what would result, or what would the consequences be, if those things were
done?

The result was dozens of ideas written on cards, posted on
a
flipchart
-
sheet covered wall,
and linked by arr
ows indicating what might lead to what. Out of the array of possibilities,
the group agreed
on a potential

mission of a metro GIS, what its goals should be, and
what major strategies and actions should be pursued to achieve the goals.
After further
discuss
ion and refinement, the Coordinating Committee adopted by consensus a
MetroGIS mission statement, goals, guiding principles and
a

set of “strategic initiatives”
to guide subsequent work.
The strategy map the group produced was a boundary object
that quite
literally helped guide development of MetroGIS over the next decade

and
more
; a reduced version of the map was posted
above

Randall Johnson’s desk
the entire
time
.


Also important was the willingness of MetroGIS leaders to plan anew once the
system had bee
n institutionalized.

The 2007 Strategic Directions workshop, another


28

major boundary experience
,

laid the groundwork for potential expansion and
restructuring of the system.

Pro
position 6
:

Leaders are more likely to guide c
ross
-
sector collaborations to
succ
ess
if they
help participants

combine
deliberate and emergent planning, with
deliberate planning
probably being
emphasized in mandated collaborations and
emergent planning
probably being
emphasized in non
-
mandated collaborations.

Proposition 7:

Leaders of

c
ross
-
sector collaborations are more likely to succeed
if
they ensure
planning
proces
ses include

stakeholder analyses, emphasize
responsiveness to key stakeholders, use the process to build trust and the capacity
to manage conflict, and build on the compe
tencies and distinctive competencies of
the collaborators.


Manag
ing conflicts.

Conflict in a collaboration emerges from the differing aims
and expectations that partners bring to a collaboration, from differing views about
strategies and tactics,
and from

attempts to protect or magnify partner control over the
collaboration’s work or outcomes. The mission of the collaboration may also affect levels
of conflict. For example, if the collaboration is mainly planning for systems change,
versus agreeing on how
to deliver a service, the level of conflict may be higher (Bolland
and Wilson, 1994). Furthermore, Gray (1996) has found that power issues, as prime
sources of conflict, vary by phases. As groups try to agree on the nature of the problem
that concerns them
,
they

are likely to
argue about who gets to be at the table
; as they
debate
approaches to solving

the problem,
they compete to shape

the collaboration
agenda and
control

information; once the implementation is underway,
collaboration
members may seek to m
aximize their authority, influence.
and
control of
resource
s
.

Proposition 8:

Because conflict is common in partnerships, cross
-
sector
collaborations are more likely to succeed if
leaders

use resources and tactics to
help equalize power
, to avoid imposed s
olutions,

and
to
manage conflict
effectively.


In the MetroGIS case, t
he forums provided opportunities for people with

different
views of information
-
sharing problems and solutions to champion their ideas and come to
agreement. Some solutions were discard
ed and others survived.

Probably the most
significant conflict that GIS advocates faced was the issue of which entities would fund
the system and to what extent. Various groups developed recommendations for how costs
would be shared among the
MC
, local gov
ernments
,

and other users. While


29

recommendations were being developed over several years, the system was funded by the
MC
, which
award
ed grants to local governments to help cover their costs of participation
.

(The GIS unit also obtained some federal grants

for its work).

At one point a federally
funded study produced a well
-
researched “fair
-
share” f
unding
model that
would have
introduced subscriber fees for government agencies and access fees for other users. Local
government partners resisted paying, howev
er
; they argued

that paying for innovative
regional systems was really
MC’s

responsibility.
U
ltimately
the Policy B
oard
decided
that
a shared funding model was not feasible, so the council continued to be the system’s
main funder.

In effect, the Policy B
oa
rd
has

decided that retaining local
government
participation
i
s far more important



because of its direct impact on the mission


than
making
participants

pay
, a
less important
resource question related to implementation
.
Thus the decision was smart given

the interest of maintaining the partnerships

and
pursuing
its mission
, but it has made the system over
-
reliant on continued support from
MC

members and top administrators and posed barriers for charging business users.

Building l
eadership
.

Collaborations
provide multiple roles for formal and
informal leaders (e.g., Agranoff and McGuire, 2003). Examples of formal leadership
positions are: co
-
chairs of a steering committee, coordinator of a collaborative, and
project director. To be effective, these people n
eed formal and informal authority, vision
and long
-
term commitment to the collaboration, integrity, and relational and political
skills (Gray, 1989;
Waddock, 1986;
Crosby and Bryson, 2005). As noted earlier two key
leadership roles are “sponsors” and “cham
pions” (Crosby and Bryson, 2005). The
parceling out of formal leadership positions
is often a means of obtaining

buy
-
in by
collaborating partners;
partners that do not obtain these positions

may require other
assurances their interests will be taken into a
ccount (Alexander et al., 2001). The
development of informal leadership throughout a collaboration is likely to be especially
important, since participants often cannot rely on clear cut, easily enforced, centralized
direction.
To cope with the predictable

turnover of leaders
,

collaborating partners
should

prepare successors and build in ways to sustain the collaboration during changes in
leadership (Alexander et al., 2001; Merrill
-
Sands and Sheridan, 1996).

Proposition 9:

Cross
-
sector collaborations are mo
re likely to succeed if they have
committed sponsors and effective champions at many levels who provide formal
and informal leadership.



30

In MetroGIS, t
he multitude of forums provided settings in which many people
could exercise informal leadership, especial
ly by contributing to the development of a
shared vision for MetroGIS.

The creation of several
cross
-
boundary
governance and
working groups expanded opportunities for people to move into formal leadership roles
.
Interviewees said staff, elected officials,
and nongovernmental people provided
enthusiastic and persistent leadership. Said one, “People with various strengths and
experience all chipped in.” Another noted the importance of consistency over the long
term: “People stayed involved.”
Victoria
Reinhard
t

served as chair of the Policy Board
from its beginning

through s
ummer 2009
. Randy Johnson, the commissioner representing
Hennepin County on the Policy Board
,

is another long
-
serving member and consistent
supporter.
Several interviewees singled out the ex
tended service of
Randall Johnson
as a
process champion who applied his expertise in planning methods and kept “energy
focused on the vision.”
Randall Johnson described a high level of rapport among some of
the leaders
, to the point that


[w]e’re
able to f
inish each other’s sentences.” The
MetroGIS leaders also became active in federal efforts to build geo
spatial
data systems.
They
used the MetroGIS experience to
contribute to
and shape
the national effort and
brought ideas back from the national forums the
y attended.

Building trust.

Trusting relationships are
essential to

facilitat
ing

the work of
collaboration and they hold the collaboration together
, but collaborating partners initially
may not fully trust each other
. T
hus
trust
-
building is an ongoing requ
irement for
successful collaborations (Ring and Van de Ven, 1994
;
Huxham and Vangen,
2005
).

T
rust can comprise

interpersonal behavior, confidence in organizational competence and
expected performance, and a common bond and sense of goodwill (Chen and Grad
dy,
2005).

Collaboration partners build trust by sharing information and knowledge and

demonstrating competency, good intentions, and follow through; conversely, failure to
follow through and unilateral action undermine trust (Merrill
-
Sands and Sheridan,
1996;
Arino and de la Torre, 1998). For example, Huxham and Vangen (2005) emphasize the
effectiveness of achieving “small wins” together

as a trust
-
building practice. Successive
accomplishments, large and small, over the course of MetroGIS’ history appear
to have
been based in part on a certain amount of trust, but also appear to have built a substantial


31

amount trust in the organization, its leadership, its decision making, and its other
practices
.

Proposition
1
0:

Cross
-
sector collaborations are more likel
y to succeed if

leaders
make sure that
trust
-
building activities (including nurturing cross
-