Kaiser Permanente: Using Interest-
Based Negotiations to Craft a New
Collective Bargaining Agreement
Negotiation Journal, January 2004
Robert B. McKersie, Susan C. Eaton,
and Thomas A. Kochan
Building on a new labor–management partnership, Kaiser Permanente
and its nearly seventy thousand union employees negotiated
a five-year contract agreement in 2000 based on the principles of“interest-based negotiations.” The people who made this remarkable
achievement happen, as well as the historic background of the case, are
described and analyzed. A key element to the success of this initiative
was the back-and-forth work of many different groups, including joint
labor–management committees, coalitions of unions, bargaining task
groups focused on particular subject areas, and local and national
leaders of the company and its unions. Using illustrative comments
from actual participants in this complex, nearly year-long negotiation
process, the authors explore how the parties crafted their agreement.
In 1997, Kaiser Foundation Health Plan and Hospitals and the Permanente
Medical Groups (hereafter Kaiser Permanente, or KP) and a coalition of
twenty-six local unions representing at the time nearly fifty-seven thousand
Kaiser employees created what is now the largest and most ambitious
labor–management partnership in the United States. In 2000, the parties
faced the major challenge of negotiating their first labor agreement under their new Partnership agreement.They designed and implemented what is
also the largest and most complex interest-based negotiations (IBN) process
carried out to date in the field of labor–management relations.We describe
this case here both to provide a historical account of the process and
to highlight the lessons that might be learned from how these parties
addressed a series of generic challenges encountered when introducing IBN
principles into collective bargaining.
The Historical and Institutional Setting
Kaiser Permanente (KP) is America’s leading nonprofit health maintenance
organization (HMO) and hospital and health care delivery system, and is the
third largest integrated health organization in the U.S. (following the
Veteran’s Administration and Mayo Clinic).Nationwide KP serves 8.6 million
members across nine states and the District of Columbia; fully 80 percent
of its operations are still in California, where it began. It was in 1938 that
Henry Kaiser created the nation’s first prepaid group health practice
and insurance programs to provide coverage for the sixty-five hundred
workers building the Grand Coulee Dam; and together with unions representing
blue collar workers, Kaiser expanded the health organization in California,
Oregon, and Washington during World War II. Advance
payment of union dues helped provide the funds to expand into Southern
KP consists of a partnership between two organizations: (1) Kaiser
Foundation Health Plans and Hospitals; and (2) the Permanente Medical
Groups. The latter are composed of physicians and other health care
providers; the former, as the name suggests, is made up of the HMO insurance
plan, and the twenty-nine medical centers and many other health care
facilities owned by KP. The eight medical groups operate as professional
corporations or limited partnerships, contracting services solely to Kaiser
Foundation Health Plans and Hospitals; at the same time, Kaiser Foundation
Health Plans and Hospitals retains its nonprofit status.
KP workers (including nurses, technical workers, service and maintenance,
and many clerical workers) were unionized shortly after Henry
Kaiser created the nonprofit organization during World War II, and unions
continue to represent most eligible workforce members.Today, KP employs
approximately one hundred thirty thousand individuals, of whom almost
eighty thousand are represented by one of eight different national or international
unions and one local independent union.
Because KP’s operations are highly decentralized, so too has been
collective bargaining. Traditionally, bargaining had taken place separately
with more than thirty different local unions and more than fifty bargaining
units, governed by separate labor contracts with different expiration dates.
Because of its origins, in part, management worked to develop positive
relationships with the unions representing its workforce. Although labor–management relations have had their ups and downs over the years,
they have been mostly positive during KP’s history. Part of the reason for
this was that until the 1980s, KP could use some cost-plus pricing, thereby
passing on some costs of improvements in its labor contracts to its customers,
or “members.” KP and its unions sought for the organization to be
the health care “employer of choice,” in their words.
However, in the late 1980s and early 1990s, KP began experiencing
severe competitive challenges in its markets, particularly from for-profit
health care providers aggressively seeking to increase market share. KP also
decided on an expansion strategy around the country, including in predominantly
nonunion areas such as Atlanta and North Carolina.With these
new pressures,management implemented a tougher labor relations strategy
that produced a series of layoffs, strikes, collective bargaining concessions,
perceived “de-skilling” (such as substituting less highly trained and lowerpaid
employees to do the work of registered nurses), and an increasingly
demoralized workforce.“This is not the Kaiser we came to work for”was a
comment often heard from frontline workers.
Developing a Partnership
By 1995, a crisis was building. Concession contracts had angered many
union locals and more trouble loomed ahead. The largest single international
union at KP, the Service Employees International Union (SEIU), convened
all its local unions with KP union members to discuss strategy.“What
was happening was contrary to how we wanted to build relationships and
build the [health care] industry,” recalled Margaret Peisert, then an SEIU
researcher and now assistant director of the Coalition of Kaiser Permanente
Unions (CKPU).SEIU then turned to the Industrial Union Department (IUD)
at the American Federation of Labor-Congress of Industrial Unions (AFLCIO)
and asked its staff to call a meeting of all the unions representing
workers at KP. Some KP nurses, represented by the American Federation of
Teachers (AFT ), joined the SEIU in this request.The IUD had many years of
experience in coordinating bargaining efforts, and so Peter diCicco, then
president of the IUD,welcomed the opportunity to address the critical problems
at Kaiser. Peter diCicco was widely respected by his union peers for
his thirty years of experience in negotiations with General Electric (GE) as
business agent of the International Union of Electrical Workers’ (IUE) local
representing GE workers in Lynn, Massachusetts as well as his success in
leading coalition bargaining for the AFL-CIO’s IUD. Peter diCicco describes
how the various parties got started:
We took our normal approach.We called an initial meeting of all
principal unions. More than one hundred people attended.We
knew from experience that we had to get all the unions on board
with a clear strategy for how to deal with Kaiser. It became
evident, given the negative attitude of the public toward strikes in health care,we had to consider other options, and so we began
looking at other means to achieve bargaining strength — corporate
campaigns and such. I went to the international unions for a
supplemental budget to fund the corporate campaign. They
accepted the supplemental budget and we staffed up and started
the corporate campaign — a successful one.
But it became clear to us if we proceeded with the campaign,we
would lose control of all this. [So] . . . I went to the international
union presidents and told them these guys [KP] are not the worst
of employers we deal with, and we might do permanent damage
to them and to our sixty-five thousand union members if we
mounted an all-out corporate campaign or used the information
we amassed for short-term advantage or leverage.Was there an
alternative? How could we work to improve quality without damaging
the already stellar quality reputation?
Based on his experience in Lynn, diCicco decided to try a “partnership
approach” with KP and enlisted John Sweeney, then president of SEIU, now
president of the AFL-CIO, to propose a labor–management partnership to
David Lawrence, then chairman of the board and CEO of Kaiser Foundation
Health Plan and Hospitals.
Lawrence describes the inception of the national partnership from his
At some point Al Bolden [a labor relations official at KP] and I
talked about the need to get together and try something different.
And I was willing to try anything at that point because it was
clear that the path we were on . . . was a dead end.We were going
to be facing labor strife in every corner of our organization.We
had fifty-four labor contracts and thirty-six unions; at the same
time we were in a fair amount of conflict between the Medical
Groups and the Health Plan — what I saw was an organization
that was starting to balkanize in very serious ways. External things
were driving a lot of this; a lot of it was being driven by changes
we were trying to make in the organizations at a strategic level.
I agreed . . . that we would meet with labor representatives
privately at Dallas–Fort Worth airport. It was almost a make-orbreak
meeting. What I remember thinking about at that meeting
was:We’ve got nothing to lose being forthcoming about what I
believed needed to happen in terms of the relationships . . . [and]
about the kind of collaboration that I think is required to deliver
modern medical care in all of its complexity.We had nothing to
lose in acknowledging the fact that there are no answers to these
things; they grow out of the collective effort of teams of people
who are working on specific areas of medical care delivery in
terms of how you best organize.
So I said these things. Peter [diCicco] said I took away all of his
thunder because he was prepared to say all of those things.And
it turned out that there was almost a revelatory session. . . .
However, KP is a complex organization, and getting approval to discuss
a national partnership strategy with most or all of its labor unions was not
a simple process.The overall KP board of directors took six months to consider
the proposal and discuss it at length. One board member, a former
chair of Northwest Power and Gas, had had a positive experience with
labor–management partnership and encouraged Lawrence to explore the
idea. Eventually all the unions formed a coalition, now known as the
Coalition of Kaiser Permanente Unions (hereafter the Coalition or CKPU),
and the board agreed to support partnership discussions. But this was just
One of the first things the parties did after agreeing to pursue more
collaborative strategies was to look for a consulting firm with significant
experience in facilitating labor–management partnerships. After interviewing
a number of candidates, they chose the Washington D.C.-based firm
Restructuring Associates, Incorporated (RAI), founded by Tom Schneider.
Facilitator and consultant John Stepp led RAI’s efforts.
“Walking through the steps,” as diCicco called it, to produce a real
partnership agreement was actually an intensive negotiation and problemsolving
process that took most of 1996 and into 1997,led by senior KP executives
and union leaders, with significant assistance from RAI’s Stepp and
Schneider. The toughest issues involved employment security, union security,
and the scope of shared decision making. An approach to shared decision
making was proposed by RAI using a continuum ranging from, at one
end, unilateral decision making with management informing union leaders
of actions to be taken, to full participation and consensus decision making
at the other end.
Employment security proved to be both a particularly difficult issue
and also one of the most critical. The language finally agreed to by the
parties stated that one partnership goal was to “provide [KP] employees
with the maximum possible employment and income security within [KP]
and/or the health care field.” This was later to require additional clarification,
but it was an essential element of the initial agreement.
“People felt it was fundamental,” says Margaret Peisert of SEIU. “You
couldn’t ask people to step up to the plate,change the way they were doing
things, get involved in joint decision making, redesigning work, and figuring
out new ways of delivering services, or finding efficiencies, if they were
going to be putting themselves or a coworker out of a job.”
Once the labor and management leaders (and respective principals)
had agreed on the key provisions of the proposed labor–management partnership,
it was submitted to a vote of the membership of the twenty-six
unions. Before this vote, however, an intensive process of education of
unionized frontline workers took place.
Most union members had never heard of a labor–management partnership,
nor did they have any idea how it would affect their interests.CKPU held a national teleconference to brief local and regional union leaders,
and produced videos describing the partnership that featured AFL-CIO President
Sweeney describing his vision for what a partnership of this size and
scope could mean for the future of labor relations in America and for the
labor movement.The partnership was approved by 92 percent of the local
union members voting, with high turnout.The only major union choosing
not to join the partnership was the California Nurses Association (CNA) representing
approximately eight thousand registered nurses in northern
California; the leaders of CNA,which is not an AFL-CIO union,chose to withdraw
from discussions before the partnership was negotiated in final form,
in part because of ongoing disputes with KP.
Preparing for Interest-Based Negotiations
When the 1997 partnership agreement was signed, the parties made a decision
to keep the partnership activities separate from collective bargaining.
Union leaders felt that to do otherwise would risk losing local union support
for partnership activities.This is a challenge many labor–management partnerships
confront, because many start out with an agreement to hold the
collective bargaining agreement separate and apart from the partnership
The KP and Coalition Partnership met this challenge in 1999 and
addressed it in a creative way. Because the partnership focused on ways to
engage the workforce and train people in interest-based problem solving, it
was naturally more developed in some areas than in others. In the Pacific
Northwest region, a more confrontational management negotiation strategy
continued at the bargaining table, particularly with SEIU’s Local 49 in Portland,
Oregon. When the service workers went on strike, and some nurses
failed to go to work (although their union did not support staying out of
work), Peter diCicco flew out to Oregon to put out yet another “fire” that
had arisen from the ashes of previous poor relations. After a settlement was
reached, he and others in the coalition began to consider the value of a
national collective bargaining agreement so that the partnership could be
Some months passed before a proposal was made, with both labor and
management support, to the governing group called the Kaiser Permanente
Partnership Group, or KPPG (see Figure One). Much to the chagrin of the
proponents, KPPG members rejected the idea, perhaps because they were
not well briefed, or because they feared a national strike possibility that had
always been prevented by separate local agreements with different expiration
This was a serious setback.However, facilitator John Stepp worked with
Francis J. Crosson, MD, chair of the KPPG, and Leslie Margolin, senior vice
president of workforce development, and other KPPG leaders, as well as
with the union leadership, to fashion an alternative approach to national
bargaining with various “gates” that the parties would move through before
negotiating a national agreement. Either side could exit as it passed through
these gates if they felt the process was not moving in a constructive fashion.
For example, in view of the primary reason for the initial rejection of the
national bargaining idea,an important “gate”was agreement in principle that
local labor market rates would continue to be negotiated locally rather than
on a national wage basis. Another critical step involved training the many
potential participants in national and local negotiations in the concepts and
skills of IBN.
Certain provisions needed to be dealt with before negotiations could
commence.For management, a key issue around which they required assurances
from union leadership was maintaining the flexibility to pay local labor
market rates as well as preventing the possibility of a system-wide strike.
For the unions, it was essential to have a target date against which to
test the feasibility of reaching a national agreement so — absent closure —
they would be free to return to the status quo ante of negotiating
agreements one by one.In other words,they needed to have a viable fallback.
The task force of the KPPG and union leaders came back with a revised
proposal that called for the following: extensive use of IBN problem-solving principles and the necessary training to prepare the parties for this very
complicated process, a single integrated national negotiation that would
allow local agreements to retain their respective deadlines (thereby addressing
one of management’s fears of a common expiration date), and a series
of decentralized task forces that would focus on particular issues. All parties
approved this in February 2000, and the first big test of the partnership had
Chronology and Overview
The 2000 negotiations at KP thus far represents the largest and most innovative
and successful experiment with IBN processes conducted to date in
U.S. labor–management relations. The negotiations involved nearly four
hundred union and management representatives and more than twenty
neutral facilitators. The negotiations included eight international unions
with twenty-six locals, one of them an independent union local.
Table One lists the chronology of key steps in the negotiation process.
Those involved in the negotiations hailed from all the local and national
unions, and managers from all the participating Kaiser regions including
Northern and Southern California, the Pacific Northwest, the Rocky Mountains,
Ohio, and the Mid-Atlantic states. These representatives were then
trained in interest-based principles and problem solving. They gathered
for a large “kick-off” session with a central coordinating group, called the
Common Issues Committee (CIC), and established seven areas for exploration.
These groups focused on such ideas as wages, benefits, and work–life
balance, which are discussed in further detail later in this essay.The parties
committed to engage in discussions that were to be open for innovation
and joint problem solving.The negotiations were complicated by the logistics
of getting everyone together,the complex interactions between national
and local negotiations (because all local unions agreed to open their contracts
simultaneous with national bargaining), and the need for representatives
to check back with constituents between meetings as well as keeping
the organization “running.”
The CIC, (cochaired by Peter diCicco, director of CKPU, and Leslie
Margolin, senior vice president of workforce development for KP) in fleshing
out the timetable for the process, agreed that negotiations should be
concluded by September 1, 2000, with any tentative agreement then
recommended to the various memberships for ratification — and local
bargaining needed to mesh with this schedule.
In addition to a national agreement, new local agreements would also
be bargained separately,even though most were not approaching their expiration
dates.The CIC sorted through the recommendations of the bargaining
task groups (BTGs) and identified those that needed to be forwarded to
local tables and those that applied uniformly across the system and therefore
needed to be negotiated centrally by the CIC.
Dealing with economic issues, especially wages, took longer than
planned and some subjects required more data and research.As a result, in
the end the parties needed a few weeks beyond the September 1 target to
finalize a tentative agreement.
The final stages of the process were characterized by a number of
twists and turns (not revealed by the chronology).With the September 1
target in mind, in mid-August the CIC entered into what turned out to be a
nine-day marathon session to resolve all of the outstanding issues — if only
in principle so that local negotiations could proceed to the finish line. As
September 1 approached and it became clear that the task of codifying the
results of the marathon session would not be finished, several of the large
unions said that they would not proceed with their local negotiations until
they received appropriate guidance from the CIC (especially with respect
to the resolution of financial items).
Consequently, the CIC found it necessary to reconvene in early September.
Whereas the makeup of the CIC for the marathon session in August
had numbered forty, for the final phase the size of the CIC had dropped to
ten participants.This “end game” phase resembled traditional bargaining in
some respects. The pace intensified with twenty-hour-a-day sessions. After
six of these long sessions, agreement was reached on all but a few agenda
items, thereby providing the guidance that several local unions needed.
Completing the task at the local level required another week.
By late September, the national and local agreements were ready to
be presented and evaluated by both KP management and rank-and-file
members of the participating local unions.The union membership approval
process alone was a logistical marvel, and the national labor agreement and
local contracts were ratified by substantial (90 percent) margins.
The substantive terms of the national agreement included a five-year
contract with across-the-board wage increases between 4 percent and 6
percent for each of the five years of the agreement (providing for regional
adjustments and higher increases for registered nurses because of labor
shortages),and numerous specific changes in practices designed to redesign
and improve business systems, quality of patient care, and work processes.
In the 1997 National Agreement, Kaiser agreed to recognize requests for
union representation with a majority showing of support without an election.
Further, employment security was pledged — with a commitment to
retraining and redeployment of workers if it proved necessary. In the 2000
National Agreement, a new trust fund was created, partially financed at six
cents per hour (with an annual escalator) from employee wages after the
first year to support training and other efforts needed to diffuse the partnership
throughout the organization. Compared to other collective bargaining
contracts, the thirty-eight-page agreement is quite brief.
In return, KP gained five years of labor peace — a major achievement
given the record of numerous health care industry strikes during the preceding decade. Also, the promise of new HMO members was a significant
gain. Here, union leaders of the partnership would promote KP as a “health
care provider of choice” to their affiliates. And of course, the most innovative
opportunity was the improvement of patient care, and the ability to
deliver it in a more participatory cost-effective manner as a result of the
joint activities fostered by the partnership. The parties agreed to jointly
address several key issues in years two, three, and four of the contract.These
issues would be the foci of workplace programs, with potential additional
compensation tied to the accomplishment of specific performance objectives.
These areas later were agreed to include improved health and safety
(also tied to fewer errors and safer patient care), and joint staffing agreements.
These agreements still are novel in unionized health care and would
have been impossible in a nonunion health setting.
With this overview as background, several key aspects of the process
that contributed to the success of IBN at KP will now be described, specifically:
how participants were trained; the bargaining over how the parties
would bargain (e.g., makeup of the committees and ground rules); the role
and work of the BTGs; local negotiations; coordination activities and the
dynamics of the process (e.g., the role of third parties).
RAI had developed a systematic way of presenting the concepts and skills
necessary for IBN. Early in the KP Partnership journey, extensive training
took place in this approach to problem solving and negotiations. More than
four hundred individuals, who would be involved in local bargaining, the
BTGs, and the CIC,were trained by RAI in the principles and tools of IBN.
Their pedagogy drew its inspiration from the work of Fisher,Ury, and Patton
(1991).These authors believed the traditional approach to negotiations suffered
from: (1) too much focus on people (i.e., fixing blame); (2) the taking
of positions because solutions were advanced early in the discussions; (3)
limited opportunity for generating alternatives; and (4) heavy reliance on
As taught and refined at KP, IBN involved five steps: (1) defining the
problem; (2) determining interests; (3) developing options (often using the
technique of brainstorming); (4) agreeing on criteria; and (5) selecting a
solution.Training of the participants who would be involved in the negotiations
was conducted by professionals from RAI and the Federal Mediation
and Conciliation Service (FMCS), and took place over a three-day period in
early May 2000.The recollections of one of the trainers captures the flavor
of these sessions:
During the training, people were excited yet apprehensive.They
did not know what to expect. A lot of questions came from the
audience. People wanted to know what was going to happen to the issues after the national negotiations — when would local
bargaining teams get underway? What would be their input?
What would be the ratification process? There were loads of questions.
Every question was answered by Leslie Margolin and Peter
diCicco. There was very good, open communication at these
meetings, and a very positive atmosphere.
Membership on the Committees, Ground Rules,
Staffing the committees involved recruiting representatives from all sectors.
Identifying the appropriate leadership on the union side was more straightforward:
the individuals were those who held office and had participated
in previous negotiations (for the most part at the local level). On the management
side, the decision was made to designate a number of operating
heads and medical group administrators, in addition to the human resource
and labor relations professionals. For example, key operations leaders
for both Southern and Northern California Permanente Medical Groups
were actively involved in the negotiations. As one union participant
observed:“The level of people that management sent to the BTGs was really
In addition to a commitment to follow the principles of IBN, an important
ground rule for both sides was the option to pull out of the process at
any time.As it turned out, this option was never exercised.
Turning to logistics, the mammoth size of the undertaking is captured
by two comments:
We had only three to four weeks to get hotels and everything for
the first meeting (kickoff on April 16, 2000). Normally it would
take four to six months to put together arrangements for three
hundred people.We only had four to six weeks.And then we had
meetings, one after another. I was working one hundred hours a
week. (From a management official)
The meetings were held in hotels in Oakland, Los Angeles, and
Chicago.The places just pulsated with activity.We tied up all the
conference rooms in these hotels for the meetings of the BTGs as
well as for caucuses.The schedule was very intensive, with many
meetings, lots of dinner meetings as well. There was a lot of air
travel and schedules kept getting revised many,many times. (From
an FMCS official)
Engaging in Interest-Based Negotiations
Once individuals had completed their training in IBN and committees were
staffed, participants began to engage in the interest-based process. In what
follows,we describe several key aspects of the IBN process at KP, including
the BTGs, local bargaining, the role of key leaders, coordinating the process, generating and exploring ideas, mediation and facilitation, and the role of
The Heart of the Negotiations: The Bargaining Task Groups
The selection of members for the BTGs was done carefully, and individuals
were recruited who were already champions of the partnership or who
might be converted as a result of participating in the intense process of
negotiations. In addition, some individuals with content expertise were also
brought on board.
BTGs on the following topics were formed: (1) wages; (2) benefits; (3)
work–life balance; (4) performance and workforce development; (5) quality
and service; (6) employee health and safety; and (7) work organization and
innovation. Each group engaged in an interest-based process of joint study,
problem solving, and negotiations. Guidelines from the CIC urged the BTGs
not to propose specific language but to develop guidelines and statements
embodying concepts and principles.
The BTGs were each staffed with two facilitators, one from FMCS and
the other from RAI.The facilitators intervened, especially when the parties
got stuck, by asking them to go back to the fundamentals of IBN: identify
interests (not positions) and generate new options.The facilitators, with the
help of the partnership’s internal consultants, also managed various lists (on
flip charts),prepared notes,and during the intervening night after each day’s
session produced a summary to help launch the next morning’s session.
Members of the CIC, except for Margolin and diCicco, were assigned
to each BTG.Their role was stated as follows:“They are not to be co-chairs;
they will play a leadership role and model the IBN principles and behavior;
they will help keep the BTG on task and on target; and they will serve as
the eyes and ears of the CIC.”
The BTGs also included line managers who could speak to the operational
feasibility of the options under consideration. In traditional negotiations,
the management contingent usually consists of labor relations
specialists and other staff with expertise in employee benefits and other“bread and butter” subjects. By contrast, the inclusion and participation in
BTGs of middle managers with significant organizational responsibilities
brought important knowledge and credibility to the process and enabled
the BTGs to tackle a wide range of subjects.
Most of the important issues were considered (except for the negotiations
over money) within the seven BTGs. Consider the experience of one
subgroup within the BTG, which tackled the subject of performance and
measurement.This group consisted of twelve people and included a person
from operations at one of the KP hospitals, a vice president from one of the
nursing unions, and other management and union leaders with direct experience
with and responsibility for the issues within this group’s mandate.
The schedule involved meeting many long hours over three consecutive days every other week. During the interludes, members of the committee
reflected on what had happened, consulted with constituents, and accomplished
behind-the-scenes liaison work that was necessary.
By early July 2000, the seven BTGs had finished their work and were
ready to report back to the CIC. All members of the BTGs assembled in one
room. Many members of the KPPG also participated. Each BTG presented
its work.The scene was incredibly energizing, according to more than one
individual; everyone came away from the session on a real “high.”
Local bargaining did not commence until August 2000, after sufficient
progress had been made in the work of the BTGs, allowing the CIC to
forward subjects with its recommendations to the teams that had been
assembled to conduct local bargaining. Consider this item from the minutes
of the CIC regarding the timing for local negotiations:
Locals will not have guidance on the specific economic parameters
of bargaining until late in the process. It is hoped that the CIC
will have a good sense of the issues that will be referred to the
local after the upcoming meetings in Chicago. Concern was
expressed that unless locals have early information on what will
be handled by the CIC and what is appropriate for local negotiations,
they may spend time working on issues that they will not resolve or not begin to work soon enough on difficult issues that will be referred to them.
For many of the small unions, local bargaining did not differ from previous
rounds because they always followed the pattern set by the larger
unions. But for large unions, having the money items negotiated centrally
by the CIC was a major change and created some unease.However, because
early in the negotiations, management at the national level had agreed to
put a significant sum of money on the table, local union leadership knew
that ratification was highly likely. So, the local unions could turn their attention
to allocating these funds and to addressing important nonmonetary
Considerable effort was made by leaders on both sides to establish a
context for local negotiations that would encourage the use of the new
process, rather than reverting to the well-traveled bumpy road of adversarial
bargaining. Consider this excerpt from discussions that took place within
the CIC, as local bargaining was about to commence:
The unions expect that as many as 90 percent of the union BTG
participants will be involved in local bargaining; they will bring
an understanding of the overall common issues bargaining
process and the interest-based approach. However, the union
believes that many of the management representatives in local
bargaining will not have been exposed to the partnership and are used to a very traditional approach to bargaining. The concern
was also expressed that some participants on both sides may not
be supportive of the common issues bargaining process or the
interest-based approach. It was agreed that both the unions and
management needed to communicate to their local bargaining
representatives the expectation that their process should be consistent
with the spirit of the partnership. Rigid adherence to the
interest-based process is not required. Concerns regarding possible
problems in local negotiations should be referred to Leslie and
The experience of a large union local illustrates how the process
unfolded at the twenty-six local bargaining tables.
Each side was allowed to have a traditional opening statement;
however, instead of presenting demands, they were asked by the facilitators
to identify issues. Approximately fifty to sixty items were earmarked for
further attention. The lead negotiators then put the issues into four separate“baskets,” with volunteers assigned to the respective subgroups. On the
union side, fifty or sixty people (many of them stewards) were at the
opening session and available for assignment.However, on the management
side only five or six were in attendance, and as a result many other managers
had to be recruited to join the subgroups.
To enable the facilitators to keep track of what was happening, each
subgroup met on a specific day, different from the others. Typically, each
subgroup only met one day a week, usually for twelve hours. The parties
agreed on a timetable of six weeks to negotiate the local agreements.As it
turned out, local bargaining took longer because progress at the national
level fell behind schedule. Indeed, there was considerable moving of issues
back-and-forth between the local and the national levels.
Of the fifty to sixty issues, only a half-dozen could not be resolved by
the consensus method. One major substantive issue had to do with the allocation
of equity adjustments. Management members wanted the monies
applied to classifications that were hard to fill in order to be competitive in
the local labor market.The union,on the other hand,wanted to reduce some
regional differences within the local between rural and urban areas. In the
end, most of the money allocated for equity adjustments was used to lessen
these geographical differentials.
Role of Key Leaders
Margolin and diCicco had worked closely together since 1998, soon after
the signing of the first phase of the partnership, and as a result, had developed
a close working relationship. Thus, the choice to focus the partnership
initially on matters outside of contract negotiations gave the leaders
time and a low-risk setting to develop a solid working relationship.
Margolin and diCicco were seen as skilled leaders of their own respective
organizations. In fact, their success in negotiating the first national agreement in large part was the fruition of a much longer journey in developing
alignment across the disparate elements within their respective
domains. On the union side, diCicco had been able to weld together twentysix
separate unions into a working coalition during the partnership era starting
in 1996, a remarkable accomplishment given the diverse locations, sizes,
and approaches to collective bargaining. On the management side, developing
consensus was just as challenging as witnessed by the initial turndown
of the proposal by the KPPG to engage in national bargaining.
IIt is clear that diCicco’s leadership was instrumental in managing the
many pressures that could easily have brought matters to an impasse. As
negotiations progressed, several presidents of large local unions thought
they could do just as well by going back to the format of separate negotiations.
And in fact some leaders played “hardball” toward the end of bargaining
and threatened to pull out of the effort to reach a national
agreement. As one observer stated the challenge:
If we had not had the three years of experience with the coalition
working together under Pete’s leadership, we never would
have reached agreement. Pete had the confidence of this large
group of individualists and he knew how to keep them focused
on the objective of reaching a master agreement. He knew how
to spell out the advantages of working together and when a leader
went the other way (“let’s shake the trees and see if we can get
something better”), Pete stepped back and let some of the other
leaders who were committed to going forward apply a little collegial
Similarly, on the management side, it was critical to have a leader who
had the respect of both line executives and physicians within the Kaiser
Foundation Health Plan and Hospitals and the Permanente Medical Groups.
Leslie Margolin provided this leadership. She brought experience as a management-
side labor relations lawyer and a top manager at another health
care company, as well as the authority associated with being a member of
the top management committee (the KPPG). Margolin was as instrumental,
on the management side of the table, in bringing about agreement as
diCicco was for labor; in fact, they worked closely as a team to guide an
extraordinarily complicated process. Significantly, she was frequently asked
to attend union caucus meetings to explain management proposals and
to answer questions.This observer captures the important bridge role that
Leslie’s leadership was instrumental on both sides of the table.
She enjoyed complete trust with the union leadership and when
she said that management had “emptied its pockets”they believed
her. In many ways her toughest job was keeping her management
colleagues on the same page. Often when she would return from a meeting with the union leadership, she would find her teammates
embroiled in tense discussion and in the process of backing
away from positions that they had put on the table. For example,
toward the end of negotiations the legal office came back to the
management team and said that the commitment that had been
made to transmit COPE funds could not be done. As soon as Leslie
went to the union with this change, it opened the door for some
union presidents to press anew for agenda items that had been
dropped. She really had her work cut out for her in managing
closure and achieving consensus on the part of the management
While diCicco did not attend management meetings, on several occasions
he held off-the-record conversations with several key decision makers.
Leadership has many manifestations, and in addition to the two
cochairs, another important resource that enabled the parties to tackle a
wide range of subjects was the presence of line managers and employees
who could speak to operational feasibility of the options under consideration.
In the KP negotiations,middle managers with significant organizational
responsibilities brought important knowledge and credibility to the
On an earlier occasion, Anthony Gately, then an administrator who
would normally not get involved in frontline discussions, had observed:
What is key to doing this [IBN] well? It does not work as well
without the people who are affected being in the room. I was the
ultimate decision maker for many issues. If I had not been there,
I would not have felt or seen the passion, the hundreds of ideas,
I would not have felt the hurt. It was important for me to be there
and to feel and see this.
And Pete diCicco echoed the same point:
What really made this work is that this was a process that did not
have to filter things through labor relations representatives but
rather we had operations people at the table with us so that when
we began talking in both directions as the unions would try to
describe an area of interest and why we had that interest, it wasn’t
being filtered through a traditional labor relations function but
rather directly to the people who were making the decisions at
the time and similarly when we got the responses back, or when
we heard from a management perspective on this, it was directly
to our rank and file leaders and our local union leaders so that
they had that dialogue, it wasn’t being filtered if you will through
the union leadership in getting essentially what that bias may or
may not be.
Third-party leadership was also critical in making the negotiations successful.
A key factor in the success of these negotiations was the facilitation by John Stepp and his colleagues from RAI as well as assistance from FMCS
staff.The key facilitators knew the parties well and were able to provide onthe-
spot training as well as help the parties stay focused on a mutual interest
agenda, using interest-based problem-solving techniques. Many union
and management leaders praised the RAI consultants for the multiple roles
that they played during the course of the negotiations. Aside from the
expected help in keeping the process on track, in some of the local negotiations,
the neutrals actually put language on paper to expedite closure
once the parties had reached agreement in principle.
Coordinating the Process
The CIC served as the main instrument for achieving coordination.The committee
cochairs, Leslie Margolin and Peter diCicco,were complemented by
thirty to forty key union and management officials. Senior mediators,
John Stepp from RAI and Barbara Pickett from FMCS, facilitated their work.
The full complement of the CIC met regularly each evening to review
reports from the various BTGs. A review of the flip charts (now archived)
reveals a highly systematic effort to keep track of all the issues under
From the outset, the CIC assured the BTGs that their proposals would
be seriously considered. But they also told them to be “blue sky” and visionary.
When the seven BTGs reported out to the CIC, almost four hundred
items were placed on the table for consideration. Many items were readily
agreed to at the main table but many had to be sent back for further work.
The RAI consultants recommended a system of “baskets” to guide the task.
In the first basket were placed those issues that needed to be settled before
the master contract could be signed, what were referred to as the nationally
mandated issues. In the second basket were items that were not mandated
but could be negotiated at local discretion, and the third basket
contained subjects for which the CIC did not hold views about how the
subjects should be resolved in local negotiations.
At times, the CIC was overwhelmed by many of the recommendations
that came to them.There were conflicts because many of the union representatives
believed their joint recommendations should have been agreed
to and automatically adopted.And of course management often came with
the orientation,“We have spent a lot of time and money, and worked on a
lot of conflicting priorities,and we can’t possibly do all these things at once.”
As one management official expressed it:
So there were probably many differences in terms of expectations.
But they got worked out as a result of a lot of hard work.
Even today, I’m not sure how we did this mammoth amount of
work within the CIC and then distilled it all down.
Generating and Exploring Ideas
According to Judith Saunders, director of national labor relations for KP
management, the idea was not to have proposals and counterproposals:
We did not want to approach these negotiations in a traditional
manner, where each side submits one hundred proposals directing
the discussion to singular solutions. Alternatively, we jointly
identified and agreed to broad subject matters, and structured
joint teams around each subject. The teams were charged with
identifying the issues related to their topic and, in an interestbased
manner, recommending solutions that met the interests of
The emphasis was on generating as many ideas as possible and keeping the
exploration away from positions, as the following quote illustrates:
Our approach was that there would be no proposals.We just had
subject matters, and asked the teams together to identify the key
issues and the interests we needed to work on.This was very different
from traditional bargaining where we would start with one
hundred fifty issues. (Management official)
All the participants had been trained in the methodology of brainstorming.
Interestingly, the management members of the BTGs swung into
the procedure differently than their union counterparts. As one person
reflected on the process: “The union folks had more difficulty expressing
interests and offered short sentences, while management tended to talk in
An effective exploration process requires the availability of substantial
information. And the BTGs were well supplied with this resource. As one
participant pointed out,“We had tons of information at our disposal.”
The work of the BTGs was thorough, with flip charts covering all the
walls of the meeting rooms. Typically they listed data needs as well as
options to be considered.
What was important about the process were the concerns that
were reflected on the walls (flip charts). So we did not have
minutes.We did not want to have a process where people could
at some point look back at some minutes and say,“You said soand-
so, and here are my notes to prove it.” (Management official)
One union official agreed about this process:
By design, not being on the record, people could speak more
easily, openly, and candidly, and explore more adventurously,
without worrying about being quoted or called to task by some
[people] after the fact.
The schedule also allowed sufficient time for exhaustive examination of
the issues. In reviewing the raw data from the process (flip charts and notes
from various meetings), one is struck by the thoroughness with which the
participants tackled their assignments. It is also clear from the debrief notes
that some thought that the process was “painfully slow”but as one participant
observed,“While we used a lot of time on the process, it did serve as a good
clearing house for us to look at all facets of the issues.”
Beyond sheer hard work, the parties formulated creative solutions to
some of the toughest negotiations process dilemmas.Take one area, communications:
the parties struck a fine balance between the ideal of joint
communications and the political reality that some communications needed
to be “in house.” And these communications were sufficiently detailed (as
much so on the union side as on the management side) to inform the
constituents but framed in a way to avoid creating unrealistic expectations.
Even communicating about the interest-based process required partnership
relationships to take priority, while each group felt bound to keep its
own members informed.
The first cut at the issue of wage increases took place within the BTG
charged with this agenda item. The participants agreed on the guideline
that KP should pay “darn good wages.” Ultimately, the resolution of the
wage increase issue had to be settled at the main table (the CIC), and at a
critical stage the union representatives accepted the reality (and the
assertion) that management had “cleaned out its pockets”and that there was
no more money to spend.This statement was accepted as credible when it
came from Leslie Margolin,KP’s chief negotiator,because she had built a high
level of trust with her union counterparts. Still,several union leaders wanted
to test this statement. It took considerable “arm-twisting” within the union
team to blunt the use of this tactic. Likewise,Margolin had to deter her management
colleagues from making new demands or backtracking on prior
Once both sides realized that the best financial offer was on the table
and acceptable,they could focus on the question of how to maximize value,
devising the best allocations for the distribution of the available resources.
Consider one aspect of this:
We decided to allocate one piece of money at the discretion of
each region and for each individual union in that region. For
example, they got across-the-board increases, and a pot that they
could use for equity, reclassifications, and shift differentials, whatever.
That process took quite a while, but it worked.(Management
One of the challenges in IBN, especially if many options are developed,
is how to converge and reach agreement. Here is an observation on this
A key is for the group to establish priorities.And the groups did
agree on what were the important standards.For example,staffing
was important to the union. Performance was important to the
company. Flexibility was also important to the company. So we
knew that there were certain things, or certain outcomes, that
were important to each side. That helped us focus on “how do
we shape a contract that we are sure includes these important
things?” (Management official)
An important innovation that RAI brought to the IBN process was the
notion that at the critical stage of focusing on agreement the parties needed
to identify their “make-or-break” agenda items. In their experience, RAI consultants
have found that bargainers find it difficult to agree on general standards
or criteria for settlement. Rather, the parties are asked to identify the
agenda items that must be resolved before agreement can be reached. Ultimately,
it is against this list of “must”items that any tentative agreement must
Mediation and Facilitation
The principles guiding labor mediators have been well established over the
years and require a mix of passive and facilitative skills. These range from
setting agendas and clarifying issues to such highly active efforts as making
substantive suggestions for compromises, urging parties to move off fixed
positions, and, at the right time, convincing each party that the best settlement
offers are on the table (Kochan and Jick 1978;Kolb 1983;Kressel 1972;
McKersie, Eaton, and Kochan 2003; Simkin 1971; Stevens 1963). Facilitating
IBN requires learning and then using new skills suited to the more problemsolving
mode of interaction.Yet,like the negotiators themselves,skilled labor
mediators/facilitators must be able to mix these different approaches as the
process unfolds and as the situation requires.The parties recognized this:
To do this process, you need a knowledgeable facilitator who
knows when to challenge,when to allow people to go off process,
when to keep them on process. It is not easy, and it takes constant
skills of facilitation, reminding people that “that was a position,”
or “that was an interest,”or “are we coming up with solutions
that are meeting everyone’s interests?”And if someone didn’t like
a particular solution, are they taking responsibility for finding
another solution? Individuals who serve as facilitators need to be
thoroughly knowledgeable about the interest-based process in
order to be able to do all of this. (Management official)
In an interview, one facilitator described the importance of being able
to mix interest-based facilitation with more conventional mediation:
Initially, I split management and the union into two groups to
determine their interests. On the union side, I found a real reluctance
to reveal their interests for fear this would make them seem weak. Since there were some problems of distrust between the
negotiator for the company and the staff representative for the
union, who both told stories about each other, I decided to
conduct shuttle diplomacy.
The Role of Caucuses
The IBN model often presented in seminars discourages caucuses.However,
in this negotiation the parties held different views. The unions wanted to
have caucuses so they could talk among themselves and iron out differences
across coalition members. Management found they needed to talk among
themselves as well. Given the size and complexity of the organization,many
diverse interests were present, for instance, the medical centers from different
regions that needed to work things out in private. Issues and solutions
differed across KP and representatives necessarily wanted to thrash
these out away from the main table.
In summary, what are the highlights of this complex negotiation story?
The parties trained and engaged more than four hundred management and
union leaders in joint problem-solving processes that focused on seven key
economic and organizational areas,producing a five-year national agreement
that enhanced the economic and employment security of the workforce.
They established a framework for sharing rewards from future performance
improvements, and positioned the parties to implement the partnership
principles into ongoing operations over the term of the agreement.These
negotiations stand as one of the signal accomplishments of the KP Partnership
to date.They are also likely to be recorded by future historians as
one of the most significant breakthrough negotiations in U.S. labor relations
of our time.
In analyzing the history of these negotiations, it is difficult to capture
the extent of the hard work, talent, and leadership of the leaders and participants
on both sides and the key roles played by the facilitators.This combination
of skilled people, resolute determination, and opportunity created
the new five-year agreement.
The parties are now involved in implementing the agreement and
putting into practice the concepts and skills that served them so well in
negotiating the national agreement. Of course the good experience of the
2000 negotiations has raised expectations that day-to-day dealings will be
transformed and that the principles of IBN will be very much in evidence
in ongoing decision making and interactions.
Fisher, R.,W. Ury, and B. Patton. 1991. Getting to YES: Negotiating agreement without giving in.
2nd edn. New York: Penguin.
Kochan,T.A. and T. Jick. 1978.The public sector mediation process:A theory and empirical evidence.
Journal of Conflict Resolution 22: 209–240.
Kolb, D. 1983. The mediators. Cambridge, MA: MIT Press.
Kressel, K. 1972. Mediation: An exploratory survey. Albany, NY: Association of Labor Mediation
McKersie, R.B., S.C.Eaton,and T.A.Kochan.2003.Interest-based negotiations at Kaiser Permanente.
Cambridge, MA: MIT Sloan School of Management Working Paper.
Simikin,W.E. 1971. Mediation and the dynamics of collective bargaining.Washington, DC: Bureau
of National Affairs.
Stevens, C.M. 1963. Strategy and collective bargaining negotiations. New York: McGraw Hill.