Three Cheers for Teaching Distributive Bargaining Open Access

I'm told that back in the 1990s, business school professors at an Academy of Management conference debated the propriety of teaching distributive bargaining to their students. The particulars of that exchange are lost in the mists of time, but at the end of the session, a straw poll apparently was taken. A huge majority of the attendees disapproved of exposing their impressionable pupils to the reality that in some negotiations, more for one party means less for the other.

I gather that the consensus view rested on the notion that distributive bargaining is brutish, perhaps even immoral. Perhaps negotiation teachers wanted to see themselves as surrogate peacemakers and problem solvers, bringers of value‐creating light to the world through the future good works of their charges. They certainly didn't want to see themselves as pit bull trainers.

I was perplexed by such a position at the time and remain so today, although I've certainly heard such views expressed a number of times between then and now. Perhaps like Emily Litella, I have misunderstood the question. Or I suppose it's possible that I've got the question right, but have arrived at the wrong answer. But I don't think so. Furthermore, I'm worried that those people who are skittish about teaching distributive bargaining are unwittingly doing their students a disservice.

It turns out that I'm in excellent company. At the 2011 meeting of the American Bar Association's Section of Dispute Resolution, a panel of legal educators tackled the same question their business colleagues had addressed approximately fifteen years earlier, but, for a variety of reasons, these law school professors came to the opposite conclusion.

With this brief note, I will introduce stimulating articles by the three panelists: Jennifer Brown of the Quinnipiac University School of Law, Paul Kirgis of the St. John's University School of Law, and Nancy Welsh from Penn State University's Dickinson School of Law. They come at the issue in different but complementary ways, each focusing on what and how to teach negotiation students.

In the first piece, for example, Brown notes that many students who elect negotiation courses are conflict averse and shy about advocating for themselves — and thus, by implication, could prove less than effective in advocating for their clients. She uses the Thomas‐Kilman Conflict Instrument to prompt students to see and confront their normal tendencies. She hopes her students will make conscious choices about how they negotiate rather than act reflexively out of unexamined habit.

Kirgis uses a sequence of simulations to much the same end. He is interested in the use of hard bargaining more broadly, not just in how the pie is divided. He wants his students to be prepared to defend themselves (and their clients) when others use tough tactics. He also gives explicit attention to ethical challenges, so that students experience the potential tension between achieving substantive success for a client and honoring their own personal values.

Welsh likewise focuses her article on how she teaches distributive negotiation, which she notes she has done more of in recent years. Her students' grades include a component for how well they perform substantively in simulations, but she has added an interesting twist by also giving students credit for the score they earn on a “reputation index.” Students thus have to balance short‐ and long‐term rewards, much as they will have to do in real life. As you will see, crafting such an index turns out to be an interesting challenge.

To frame their pieces, let me briefly offer my own three cents, specifically a trio of arguments on why we must teach distributive negotiation. Each is sufficient in its own right to make the case, though the points are tightly linked. I look forward to hearing counterarguments, of course, as well as concurring opinions.

First, we need to be clear about what distributive bargaining entails and what it means to teach it. To ground the former in an example, we can imagine a transaction between a seller who won't part with an item unless she gets at least $75 for it and a would‐be buyer who is willing, if need be, to pay as much as $100. To make this a purely distributive situation, we'll assume this is a private, one‐time encounter. Because the parties have no prospect of a future relationship and the transaction is anonymous, they also have no reputational concerns. Thus, in this stark case, we're assuming that each party is concerned only about her own financial welfare; hence the buyer wants to push the price down while the seller wants to push it up. And just to keep the early analysis clean, we'll posit that each party knows the other's exact “walk‐away” number.

Other than that last stipulation, such situations occur in real life. They are often termed “zero‐sum,” but that it is a misnomer. As Howard Raiffa has noted, these are more accurately described as “quasi‐constant‐sum” games. The distinction is important. In our example, a potential surplus of $25 could be divided between the parties.

But here's the rub: there will be no surplus unless the parties agree on a price that each sees as preferable to either getting nothing or to continuing to haggle. (By contrast, a “dictator” game in which one player has to decide how much to give the other — if any — is truly zero‐sum, while the “allocator” game is not because the recipient can veto a deal.) The good news is that people often do come to agreement in such situations. The process may be quick or protracted. The bargainers may find a lofty principle to justify a particular figure, or they may just flip a coin. But whatever means they use to reach a settlement, they have certainly created value relative to their alternatives. An economist might argue that in doing so they have also created social value by finding an outcome that is efficient, in that it is superior to deadlock.

Too often, we gloss over the value‐creating nature of distributive bargaining. After all, because a bargaining range or ZoPA (zone of possible agreement) exists doesn't guarantee that parties will reach agreement. (We see some of our own students getting stuck, and the world is all too full of examples of negotiators who manage to snatch stalemate from the jaws of victory.) Economists would say that in our example, there is no stable equilibrium. Two computers wired together would melt before they could settle on a price. So hip‐hip‐hooray for bumblebees that fly and for real‐life people who are able to create value, theory be damned!

Starting our courses with this kind of bare‐bones scenario gives students a better chance of understanding core concepts. Often they confuse BATNA (best alternative to a negotiated agreement) with walk‐away (or reservation) price. The former is a course of action, while the latter reflects how we value that option subjectively.

Where did the seller's $75 come from we might ask? Does she have a firm offer from somebody else or a mere expectation? If a good friend was the one who made the offer, does the buyer here only have to better it by a penny — or by much more? And what of the buyer's $100? If she thinks she can get an equivalent item for that price, has she figured in the hassle of another negotiation — or, for that matter, the possibility that another, less demanding buyer might materialize? We thus can spin off many hypotheticals, but unless students understand the basic framework, their thinking will be muddled.

Then, by relaxing the assumptions, we can explore process issues. Let's assume that the parties have been thoughtful about calculating their own walk‐aways, but what if they are utterly in the dark about what's acceptable to the other side? Bargaining then becomes a search in which parties try to determine if there is room for agreement and, if so, how much. They take soundings by making offers, though those offers are also signals that will be read (or misread) by their counterparts. As teachers, we can bring in auction theory here, in regard to balancing risk and rewards that come with making sealed bids. The value of information — and thus why it's often closely guarded — can be explained in this context.

In short, the first argument for starting a negotiation course with the simplest form of distributive bargaining — two parties, one issue, no relationship, or reputational concerns — is that doing so establishes an analytic foundation for our students. When we strip away the simplifying assumptions, one by one, we then offer students the opportunity to develop a richer view of the process.

The second argument for teaching distributive bargaining is that lessons from cognitive and social psychology are powerfully illustrated by single‐issue cases. To put it another way, when we introduce complex behavioral issues, it makes sense to keep the rest of the story simple. Let's take, for example, Ian Ayres and Peter Siegleman's (1995) classic studies of car buying. We only have to show students these data, and we're off and running. We can challenge them: why is it that salesmen (and they were mostly men) ask women to pay more than male customers, and likewise, blacks more than whites? Because we're talking strict dollars and cents, the bias question can't be ducked. But if distributive bargaining is off limits, students won't have the chance to confront such issues as directly.

Burgeoning interest in how people's actual behaviors diverge from models of rational decision making has kept a generation of social scientists off the streets and busy with their lab subjects. Again, many of these experiments are set in distributive contexts. I'm less enthralled by research on anchoring than are many in the negotiation field, but I wonder nonetheless how we would demonstrate that phenomenon if, in debriefing exercises, it were a no‐no to ask students whether their various first offers affected subsequent concession patterns.

The simple architecture of distributive bargaining thus can be the grain of sand in which much of the universe of negotiation can be revealed. That brings me to my third argument. To the extent that antipathy about teaching distributive bargaining rests on qualms about its moral implications, avoiding such cases ironically enables teachers and their students to duck difficult and important questions.

Let me illustrate that point with a story about a colleague, an economist, whom I like and respect. Several years ago, he was a member of the team that taught the required introductory negotiation course at my school. In the course of developing the standard curriculum that all of us would use, he got in a protracted dispute with some of his colleagues — me included — about having our students undertake the familiar “Win‐As‐Much‐As‐You‐Can” exercise. This is a four‐person prisoner's dilemma game played over ten rounds. Each player has two cards (X for defect; Y for cooperate). If all four cooperate on a given round, they enjoy modest gains. If they all defect, they suffer modest losses. If there is a mix, the defectors gain (sometimes substantially), and the cooperators lose. Players are instructed that this is an individualistic game. There is no reward for helping someone else, nor is there penalty for hurting them. One's sole concern is maximizing one's own bottom line.

My colleague hated the game because defectors sometimes end up doing quite well and some cooperators get burned. The rest of us nevertheless love the game because it wonderfully sets up issues of trust, communication, the tension between self‐interest and the group's interest, and, yes, ethics.

His objection was that it “teaches the wrong lesson.” Our students, he said, should not be left thinking that they can screw others over and not pay the price one day. What comes around, eventually goes around, he contended. I wish that were always true, as the world would be a better place. I don't think I'm being cynical, however, in saying that there's plenty of evidence to the contrary.

My problem with my colleague's view, however, is not his optimism. Rather it is that his formulation fudges a fundamental moral question: namely, what do we owe others beyond merely acting in our enlightened self‐interest (as he implicitly would have it)? When you and I are grappling over an orange, because you want the peel to add zest to your cake and I, in turn, covet the juice, that's neither a negotiation game nor a moral problem. It's simply a matter of communication and coordination.

By contrast, hard moral choices usually involve sacrifice. If we go back to our early example, what does the seller owe the buyer (and conversely) in regard to distributional fairness? Moreover, what is the source of that obligation, and does it apply across the board or is it context dependent? Then, how should we conduct ourselves in dealing with others who feel no such duty?

These aren't easy questions, and they cannot be wished away. Yes, pies often can be expanded to some extent, but there are limits. If we are implying that as pies get bigger, we can afford to be more generous, we may be guilty of sloppy thinking. One could just as well argue that when pies are small and there is less to go around, equity is all the more important.

The distinction between distributive and integrative bargaining goes back at least to Richard Walton and Robert McKersie's classic, A Behavioral Theory of Labor Negotiations (1965). Negotiation teachers have been debating the propriety of what and how to teach their students ever since. That the panelists at the ABA ADR Section meeting came to a different conclusion than did participants at the Academy of Management meeting fifteen years earlier may simply reflect differences between two groups of people in two different places. But I suspect that negotiation pedagogy has shifted, as well. I'm not suggesting that we have gotten meaner or crankier as we've aged. Far from it. Instead, as Brown, Kirgis, and Welsh demonstrate in these pages, some negotiation teachers have developed imaginative ways to grapple successfully with the tough course design and classroom dilemmas that the issue of distributive bargaining presents.