1

Feature Article

2 New tools for negotiators

A few simple ideas make it possible to construct powerful strategies for even the most complex deals.

Tera Allas and Nikos Georgiades

2001 Number 2

Negotiations are the stuff of business life, and volumes of advice tell managers how to prepare for and conduct
them. But most of the advice applies to deals that are simpler than those pursued today. Negotiations now
typically involve many parties that have an interest in the outcome—the "stakeholders"—and require decisions on
many complex issues. Immense sums of money may ride on the outcome.

Sophisticated support for negotiators is available—for example, computer simulation models based on dynamic
game theory. These models can predict the behavior of stakeholders in multiparty deals and guide negotiators to
winning strategies. As yet, however, few business leaders feel comfortable surrendering their personal judgment
on such crucial issues to a "black box."

We have therefore been developing a middle way: a set of readily understood tools to help decision makers in
complex multiparty negotiations.1 With the help of these tools, negotiators can develop strategies that are not
only favorable to them but also palatable to the other parties. This method rests on the same logic as
sophisticated simulation tools but doesn’t require an elaborate computer model. It has been used so far to
develop strategies for several real-life negotiations and is appropriate whenever many stakeholders have different
goals on a number of issues and the final decision emerges from bargaining among those stakeholders. The tools
are therefore relevant to mergers and acquisitions, partnerships and alliances, regulatory rulings, litigation, and
labor disputes—in short, to many of the items at the top of a chief executive officer’s agenda.

Who are a deal maker’s allies and enemies? What is the chance that a particular outcome will stick?

Applying the tools at the start of a deal should answer many crucial questions. Who are a deal maker’s true allies
and enemies? What is the chance that a particular outcome will stick? Who is worth lobbying for support on
particular issues? What bargaining chips can be traded for that support? Used together, these fact-based tools
suggest effective strategies for arriving at decisions that all parties can accept, and they help negotiators refine
their plan of action as circumstances change. To see how the tools operate, consider how they helped negotiators
in two real situations.

Power play

A European government decided to liberalize its electricity generation and supply industry. One major move in
carrying out that plan was the privatization of a near-monopoly utility, which we will call Power. Negotiations had
to be conducted on several issues, including the level of end-customer tariffs that Power could charge, the
wholesale-market structure most appropriate for competition, and the possible forced sale of Power’s generating
plants.

Power’s first task was to identify the key issues and the range of possible outcomes for each. On the tariff issue,
for example, the regulator proposed a one-off cut of 15 percent in tariffs charged to end customers. Power, by
contrast, aimed to keep the existing tariff levels. In addition to such obvious issues, Power considered several
others, such as "green" subsidies: higher tariffs for electricity produced at environmentally friendly plants. Some
of the additional issues at first seemed immaterial but later turned out to be important.

Next, Power had to identify all of the stakeholders that might influence the decision on any issue and to
understand the objectives of each. It identified a total of 16 parties, including the regulator, the Ministry of
Industry, the Treasury, labor unions, competitors, the free-market system operator, and Power itself. Although the
number of parties was large, a negotiator using our tools requires only three pieces of information about
stakeholders to predict their behavior on any issue.

1. Position: What is the stakeholder’s preferred outcome on the issue? Is the stakeholder arguing for one of
the extremes on the range of possible outcomes, for example? For a stakeholder who hasn’t stated a
position on a particular issue, what would the rational position be?
2. Salience: How important is this issue to the stakeholder as compared with all other issues? For example,
would the stakeholder drop everything else to attend a meeting on this issue? What do we know about
this person’s past interest in the issue—for instance, his or her willingness to spend resources on it?
3. Clout: As compared with other players, how much power does the stakeholder have to influence the
decision on this issue? In the case of Power’s future tariffs, for example, the regulator had the highest
clout, followed by Power and the Treasury.

It is helpful to express this information by assigning a number from 0 to 100 to represent each stakeholder’s
position. On tariffs, Power chose 0 to represent its own position and 100 for the regulator’s. The other
stakeholders fell somewhere between these two extremes (Exhibit 1). As for salience, a rating of 100 means that
this is the most important issue for the stakeholder, a rating of 50 that it is one of several important issues but not
the most important, and a rating of 0 that it is unimportant. To express clout, the stakeholder with the greatest
influence on an issue—in the case of tariffs, the regulator—would get a rating of 100; the others would get fewer
points.

The task of gathering all of this information may seem daunting. In our experience, however, it is possible to find
people who can provide answers based on their dealings with the stakeholders and their knowledge of the current
negotiations. If so, a day or two will suffice to question these people carefully, preferably in a workshop setting; to
calibrate their inputs; and to arrive at consensus estimates. Recording the results in an electronic format will be
useful for later analysis.

The first cut

Having undertaken this preparatory work, Power was ready to see how the negotiations might pan out on each
issue. The first tool, called an outcome continuum, graphs each stakeholder’s position on a given issue relative to
the two extremes. By weighting the players’ positions according to salience and clout and then calculating the
average, you can foretell the result of a pure compromise "vote," with no bargaining among the parties. This
analysis quickly showed Power that it faced tough opposition on the tariff issue, with a potential outcome far from
its ideal (Exhibit 2, part 1).

However, the outcome predicted by a pure vote usually doesn’t materialize in reality. When negotiators don’t like
what they see coming, they bargain. To estimate the probability that all of the negotiating parties would accept
the pure-voting outcome, Power applied the second tool, stability analysis, which examines how far the outcome
is from each party’s preference, how much each party cares about winning on the issue, and how important the
parties are in reaching an agreement on it. If too many important players are dissatisfied with the outcome, it
isn’t likely to be stable.

In Power’s case, stability analysis showed that a straightforward compromise on the tariff issue would leave Power
and the regulator very unhappy; both were likely to fight on (Exhibit 2, part 2). Power therefore had to apply the
next three tools to find subtle ways of tipping the balance in its favor.

Allies and enemies, influencers and bargaining chips

First, Power used a stakeholder classification, which identifies each stakeholder as an ally, an enemy, or in-
between on every issue (Exhibit 3). This exercise provides a useful overview of the main challenges and
opportunities in the negotiations as a whole; it showed, for example, that Power and the regulator were enemies
pretty much across the board. Power also found that on the issue of the wholesale-market structure, it could hope
to get its way fairly easily, since none of the stakeholders were outright enemies.
To see which of the potential allies would make the most useful friends, Power placed the stakeholders on a
negotiation landscape for each issue. Players with a lot at stake but little influence are "followers"—good to have
on your side but not, given their lack of clout, worth much effort to win. The "shapers," by contrast, both care
about the issue and can influence the outcome, so they are natural partners—or very strong enemies. Just as
important are the "influencers," who are not greatly concerned with the issue but have a good deal of influence
over it. If you can persuade them to support your position, you are much more likely to win.

In Power’s case, the shapers on the tariff issue were Power itself and the regulator, an enemy. But there were
several influencers—including the Treasury, the Ministry of Industry, and the system operator—that Power could
court (Exhibit 4). It now realized that it had to convince them that the tariff issue was salient for them as well.

In some cases, lobbying alone can work, but certain stakeholders want something in return for their support.
Moreover, Power still needed to get additional leverage with the most problematic stakeholder, the regulator.
Power discovered its best bargaining chips by identifying the issues for which it was itself an influencer. To find
them, Power plotted the salience of every issue for itself and for each other stakeholder on separate relationship
analysis matrices. The upper-left-hand quadrant contains the negotiator’s bargaining chips: issues that he or she
is willing to compromise on but that are important to the other party. 2 It turned out that on the issue of green
subsidies, which Power had initially regarded as insignificant, it could give a good deal of ground to the regulator
in return for a more favorable outcome on tariffs (Exhibit 5).
To build an alliance against the regulator’s tough stand, Power could try to increase the salience of the tariff issue
for the Treasury

By working through these five analyses, Power discovered early on that it had little chance of winning on the tariff
issue without hard bargaining. But it also found strong bargaining tactics. To build an alliance against the
regulator’s tough stand, Power could try to increase the salience of the tariff issue for the Treasury and other
stakeholders, which already favored Power’s position but would have to be persuaded to support it more actively.
Power could also bargain away green subsidies in return for higher tariffs. Although these strategies were
counterintuitive, they were practical and in marked contrast to Power’s initial confusion in the face of so many
issues and stakeholders.

Adding dynamics

These analyses can give you an exact snapshot of your own and your opponents’ negotiating strengths, along
with insights into ways of changing the dynamics of the game. Negotiators will have more confidence in their
chosen strategy if they can see what happens when they apply it to the next round of negotiations—something
they can do by rerunning the first two analyses with updated information and then tracking the results on a
spreadsheet.

A conglomerate we will call Alpha used this technique in negotiations concerning Coolstuff, a high-technology
company in which Alpha and another conglomerate, Beta, each had a 50 percent stake (Exhibit 6). The two
conglomerates had set up Coolstuff to share R&D costs and expertise, and the venture had successfully
commercialized many of its inventions. In contrast, Gamma, an independent start-up in the same field, was
brilliant at developing technology but less successful in the marketplace.
The management teams of Gamma and Coolstuff wanted the two companies to join forces so that they could
apply Coolstuff’s marketing know-how to Gamma’s unique technologies. For Coolstuff, the price of the deal would
depend on the relative valuation of the two companies and whether the deal was structured as an acquisition of
Gamma—with a corresponding premium—or as a merger. An acquisition would also raise the issue of whether
Coolstuff would pay in cash or in shares.

Irreconcilable differences?

Despite the commercial logic of a deal, differences in the expectations of the companies’ owners seemed likely to
make it impossible to consummate. Alpha hoped that Beta would in due course buy out its Coolstuff stake. Since
Alpha’s share of the combined Coolstuff-Gamma entity would be more valuable than the company’s share of
Coolstuff alone, Alpha naturally wanted a highly valued Coolstuff to merge with a relatively low-valued Gamma for
the lowest possible outlay of cash.

Beta was in the business for the long haul, so it sought full management control of Gamma, with or without Alpha.
Beta therefore wanted Coolstuff to hold all of the equity in the new company, leaving Gamma’s original owners
with none. Rather than merge, Beta wanted Coolstuff to acquire Gamma—a course that would make it easier to
replace Gamma’s management. Although the price might include an acquisition premium, Beta cared more about
control and the long-term potential of the business. Beta was also determined to preserve the Coolstuff brand as
the public face of the new company.

The owner-managers of Gamma wanted it to be valued as highly as possible. They also wanted shares in the new
venture as payment so they could benefit from the future commercial success of the technologies they had so
passionately developed. They envisioned the new company as a merger of equals, branded Coolstuff-Gamma.

The search for common ground

Alpha ran the outcome continuum analysis, with dispiriting results. Beta, as the negotiator with the most clout and
salience, would probably wish to acquire Gamma for cash and to keep the Coolstuff brand name. Stability analysis
suggested that although Alpha might accept this arrangement, Gamma would be so dissatisfied that the deal
would never be consummated, and Alpha would have no chance to get out of its unwanted investment in
Coolstuff. Still, the game wasn’t lost: every stakeholder attached a different degree of importance to each issue,
so there might be room for bargaining.

The stakeholder classification reminded Alpha that Beta and Gamma might collaborate with Alpha on some issues
even if those companies were enemies on others. Gamma was Alpha’s natural ally on the merger-versus-
acquisition issue, for example, while Beta would support Alpha’s preference for a high relative valuation of
Coolstuff. The negotiation landscape analysis on each of the issues showed Alpha which party it should lobby for
support on the questions it cared about most. Alpha had a better chance of getting its way on the valuation, for
instance, if it could increase the issue’s salience to Beta. It could also try to get Gamma to push harder for a
merger rather than an acquisition. Both companies were predisposed to support Alpha on these issues, but it had
to persuade them to use their clout.

A relationship analysis showed Alpha what it might offer in exchange. The terms of payment and the name of the
future brand emerged as powerful bargaining chips for dealings with Gamma. Alpha could promise to wield more
of its own clout to support the desire of Gamma’s owners to retain stock in the new company if Gamma in return
accepted a relatively high valuation for Coolstuff. Alpha could also agree to rebrand the new company—an issue
much more important to Gamma’s management than to Alpha’s.

But the same analysis vis-à-vis Beta produced rather problematic results. Apparently, one of the most effective
ways of persuading it to accept a merger would be for Alpha to grant the company a cash-based deal. Alpha could
also help Beta get its way on the brand issue. Which party should Alpha support on which of these issues? What
would be the most effective use of Alpha’s bargaining chips?

And the winner is . . . everyone

Alpha needed only two days to gather and structure the data needed for these analyses and only half a day to
apply them

To find out, Alpha ran the stability analysis again, testing which of two strategies was more likely to yield a deal
among the parties. Should Alpha back Gamma’s desire for a combined brand name but persuade Gamma to
accept a cash deal? Or should it support Beta on maintaining the current brand but push for a share-based deal
more acceptable to Gamma? Rerunning the analysis showed that the latter strategy was the only way to make
everyone happy enough to agree to a deal; the other route would still leave Gamma highly dissatisfied. In
contrast, securing payment in shares would be just enough to persuade Gamma to go ahead with the deal and
pave the way for Alpha’s exit from the Coolstuff venture.

Alpha needed only two days to gather and structure the information for these analyses and about half a day to
apply them, with iterations. The effort yielded a practical strategy for resolving what had seemed to be a total
impasse. The deal went ahead.

More advanced game theory modeling can include—at a fairly low cost—an almost infinite number of dynamic
simulations, which allow a negotiator to identify and test strategies with greater precision. So far, few businesses
use this approach. Managers seem reluctant to apply mathematics to a process inherently concerned with human
judgment.

But past negotiations reveal that stakeholders behave predictably given their position, salience, and clout on each
issue. Negotiations are therefore susceptible to more rigorous analysis than traditional approaches allow. Five
simple tools—the outcome continuum, the stability analysis, the stakeholder classi-fication, the negotiation
landscape, and the relationship analysis—can give negotiators the best of both worlds.

About the Authors

Tera Allas is an associate principal and Nikos Georgiades is a consultant in McKinsey’s London office.

Notes

1
The methodology builds on the theoretical and empirical work carried out by Decision Insights Incorporated (DII),
a consulting company specializing in the computer-supported simulation of negotiations and political decisions. In
particular, DII’s dynamic-game-theory model for conducting multiparty negotiations gave us the concepts of
"position," "salience," and "clout," which are fundamental to analyzing the behavior of stakeholders in
negotiations.
2To find bargaining chips even more precisely, a negotiator could also take into account its relative clout on each
issue. The greater its clout, the more valuable the bargaining chip.

Site Map | Terms of Use | Updated: Privacy Policy | mckinsey.com

Copyright © 1992-2005 McKinsey & Company, Inc.