Alexander Dobrev

Concession Aversion was the winning entry in a 1999-2000 competition.

I must confess immediately, concession aversion does not really have much to do with food. But it provides a good starting point: Imagine a Double Whopper with cheese. No onions. Now imagine your Whopper accompanying you to the negotiation table, where I, your counterpart, take it from you and eat it. You have just lost a Whopper .

Imagine a second scenario. You come to the negotiation table with no food of your own and I, your counterpart, of my own accord, offer you that very same type of sandwich. You have just gained a Whopper. How would you value these otherwise identical sandwiches?

A loss aversion approach would predict that most individuals would "hate" the loss of the Whopper in the first scenario considerably more than they would "love" the gain of an equivalent product in the second. You may be wondering, "How can anyone compare the value of the loss of a Whopper with the value of its gain in terms of 'love' and 'hate'?" Indeed, love and hate, although certainly expressing value, are not readily quantified. Consider, therefore, the following alternatives to a love-hate relationship with your sandwich:

How much money would you require in order to lose your Whopper in the first scenario, if I asked for it?

If, on the other hand, I offered you a Whopper when you did not have one, how much would you be willing to pay in order to gain it?

The loss aversion prediction is that you would ask me to pay a considerably larger proportion of money to part with your Whopper than you would be willing to pay if I offered you mine.

So far one thing seems clear: it is not a good idea to take somebody's food. But other than good table manners, does loss aversion provide any practical guidelines for successful negotiation behavior? This question brings us to the concept that inspired the culinary theme of this paper -- concession aversion. Before you get a case of "concept aversion," however, let me assure you that concession aversion is nothing but loss aversion as it arises in certain negotiation contexts.

Let us go back to the Whopper example. Notice that in the case where you valued the Whopper more, you actually intended to eat it. More generally, the Whopper was something you intended to use. On the other hand, if the Whopper is intended strictly for exchange, no such drastic difference in value (asking and offer price) is likely to occur. That is, if you are in the business of buying and selling Whoppers (a Whopper broker of sorts ), and happen to bring a Whopper to the negotiation table, your asking price for that Whopper would not be as disproportionate to your offer price for my Whopper as it was when the Whopper was intended for use other than exchange, namely eating.

Concession aversion refers to the disproportionate value assigned to for-use type interests in a negotiation context, and particularly to the surrender of such interests. For example, let us consider our Whopper broker re-negotiating her contract with Burger King. The for-use interest in this hypothetical is vacation time. Let's say our broker has until now enjoyed a 20-day yearly vacation. Management has somehow decided on a two-day reduction of vacation time, for productivity maximization. The perception of such reduction as a loss by the employee will likely result in a valuation considerably disproportionate to the value management may have assigned to the same reduction based on productivity gains. The more disproportionate the values assigned, the less likely it is that a mutually satisfactory agreement will be reached.

The key feature of concession aversion is that the very perception of an exchange as a loss assigns extra value (or cost) to that exchange and, importantly, such assignment does not usually have any identifiably rational basis. To ground this assertion, let's consider the following actual experiment: an attractive object (starring a decorated mug) is given to a portion of the people participating in the experiment. These are the owners. They are asked to consider a list of prices and indicate for each price whether or not they would be willing to exchange the mug for that particular sum of money. No bargaining is involved.

A second portion of the participants act as choosers. As the designation suggests, they are faced with a choice: they must decide whether to keep a sum of money given to them (by the people conducting the experiment), or to offer all or a portion of that money to the owners to get a mug. Notice again, there is no bargaining between owners and choosers: both indicate their valuation of the mug on identical "price-menus" and an exchange occurs only when there is an overlap between what a chooser is willing to pay and an owner is willing to accept for the mug. Notice also that the owner has no rationally added value to the mug ( e.g., sentimental value due to memories, etc.). The only difference between the owners and the choosers is that the owners happen to be in physical possession of the mugs. Because there is no bargaining involved ( each owner and chooser evaluates a list of prices for the mug once ), a difference in the price assigned by each group cannot be attributed to strategic or tactical motives.

As you have probably guessed, there was, in fact, a difference in the median price given by each group: the owners valued the mug at $7.12 whereas the choosers thought it was only worth $3.12. This disproportionate valuation of the mug (which is a for-use item) can be attributed to the fact that the owner perceived parting with the mug as a concession of a for-use interest, whereas the chooser saw the acquisition of a mug as a simple gain.

I would like at this point to examine several ways in which awareness of the phenomenon of concession aversion can inform and guide a more successful strategy of "getting to yes." Two aspects of concession aversion -- ( 1) an individual' s overvaluation of losses and (2) his or her dubious rationality for such overvaluation -- provide the blueprints.

The fact that individuals tend to overvalue losses has immediate implications: the most cost-effective offers you can make in a negotiation are those that minimize your counterpart's losses of for-use interests. Let's go back to our Whopper broker. As you may recall, management was considering reducing our broker's vacation time by two days (from 20to 18) in order to maximize productivity. Let's say that such reduction would have improved overall productivity by a monetary equivalent of $250. Let's also assume that management was intending on increasing our broker's salary3 by $2,500. To achieve its productivity objectives, management can do (at least) two things: either reduce the broker's vacation time by two days or increase her salary by $2,250 rather than $2,500. The proportionate values are identical: two days represent exactly one-tenth of 20 days, and $250 are one-tenth of $2,500. From the employee's perspective, however, the reduction of vacation time is a loss of a for-use interest, whereas the reduction of salary increase is a reduction of a gain. Using the overvaluation predictions of loss aversion, we can conclude that the employee will see the reduction of(for-interest) vacation time as considerably more costly than the proportionate reduction in gains. Thus, there is a better chance to reach agreement when the employee is asked to accept a reduction in salary increase rather than in vacation time: management will be expected to concede less in return for raise reduction while at the same time reaching its productivity goals. We also noted in the "attractive mug" experiment that the overvaluation of for-use objects (or interests) in one's possession is not necessarily a rational one. In terms of negotiation strategy, this knowledge adds an extra "oomph" to a simple question: why? While it is almost always advisable to ask your counterpart to explain his or her position by giving rational reasons for it, it is perhaps even more important to ask them to do so when you know they have no such rational reasons. Parties on opposite sides of a negotiation table often see themselves as the only source of reasonableness. Raising awareness of possible latent irrationalities (in a tactful manner, of course ), may improve the chance of reaching a well reasoned agreement.

Endnotes

See Daniel Kahneman and Amos Tversky, Conflict Resolution: A Cognitive Perspective, as presented in Arrow, Barriers to Conf1ict Resolution, p. 55.