New Perspectives on Issue Analysis—One‐Sided Preferences as a Strategic Source in Multi‐Issue Negotiations

Proper negotiation preparation is multifaceted; negotiators must identify issues, define interests, and analyze the other party or parties (Lewicki, Barry, and Saunders 2010). A crucial part of negotiation preparation entails structuring negotiation issues and topics (Pendergast 1990; Peterson and Shepherd 2011). There is consensus in previous research that issue analysis is a central method of this type of preparation (Fells 1996; Peterson and Lucas 2001; Lewicki, Barry, and Saunders 2010; Kirgis 2012). In this article, and in line with Kirgis (2012), “issues” are understood as the subjects over which opposing sides negotiate. “Issue analysis” identifies different valuations that help the parties to determine negotiation strategies, potential trade‐offs, and alternative ways of reaching agreement (Naquin and Paulson 2003; Giacomantonio, De Dreu, and Mannetti 2010; Lewicki, Barry, and Saunders 2010).

Given the importance of negotiation preparation, it is not surprising that there is comprehensive research on preference analysis in negotiations. Preference research is multidisciplinary and includes work in microeconomic theory, behavioral economics, cognitive psychology, and the neurosciences (Gimpel 2007). Insights from behavioral economics and reference‐dependent aspects are most relevant for studying preferences about (multiple) issues in negotiations, as they are most likely to be related to negotiators’ individual assessments of the importance, priority, or relevance of specific issues (Bendahan et al. 2005; Giacomantonio, De Dreu, and Mannetti 2010; Albin and Young 2012). These subjective values for issues influence a negotiation process and its results (Curhan, Neale, and Ross 2004; Gimpel 2007; Lewicki, Barry, and Saunders 2010). When considering preference positions on an aggregated level of a negotiation, research generally distinguishes between two types of preference positions, namely, one‐sided or two‐sided. This division depends on whether both or only one of the negotiating parties considers the issue to be important and therefore seeks to achieve the best possible outcome concerning it. The negotiators’ individual assessments can be of different strengths or even opposite (Lax and Sebenius 1986; Vetschera 2007). The existence of two‐sided preferences on at least one of the negotiation subjects is obviously one of the reasons that there is a negotiation in the first place.

One‐sided preferences (where an issue is important to only one party) have not received much attention. This is probably due to the assumption that parties only negotiate on matters that are of concern to both sides (Peterson and Shepherd 2011). However, when striving for a comprehensive deal, agreement must be reached on all issues, including those with a one‐sided preference. Especially against the background of the need for integrative strategies in complex multi‐issue negotiations and for cooperative business relationships (Lebow, Pruitt, and Rubin 1987; Rubin, Pruitt, and Kim 1994; Curhan, Elfenbein, and Xu 2006; Patton and Balakrishnan 2010), it is becoming more necessary to look for additional options. To the best of our knowledge, there is no research on two‐sided preferences in cases where only one party has a preference toward an issue and the other has none at all (0‐preference).

Following the approach of logrolling (Froman and Cohen 1970; Pruitt 1983), a negotiator with no (instead of lower) preference could take advantage of the preference imbalance by using the issue for trade‐offs on other issues, assuming that the opposing party is more dependent on that issue due to its higher relevance rating. Consequently, the counterparty might need to concede on other (more preferred) issues to realize gains on the issue with a one‐sided preference. Thus, we may hypothesize that a 0‐preference could function as leverage during a negotiation.

Examples from negotiation practice in the business‐to‐business environment reinforce these assumptions. A typical negotiation issue with a one‐sided 0‐preference is the timing of payments. Large companies usually have high liquidity, and it does not matter much when payment is made (before, during, or after performance). For small or young companies, on the other hand, it can be very important to receive at least 50 percent of the payment at the time the contract is signed. The more liquid company could therefore accommodate the counterparty—often at the end of the negotiation—and receive a last concession on an issue that is more important to them, since this step means a large, unexpected concession for the counterparty.

Another typical 0‐preference can be identified if a change in one negotiation issue does not result in any change in the value of the overall deal for one of the parties. This is often the case with the inclusion of general quality standards in the final contract. For some companies, it could be very difficult to meet these contractual requirements (pain point) and they would extremely benefit from a weakening of the general standards. The counterparty, on the other hand, may have no problem with this weakening, as long as the minimum quality can be guaranteed. This issue can also be played strategically when there is a risk of a breakdown in negotiations, as a concession on it represents a major added value for one party and increases its willingness to engage in further negotiation rounds.

A broader scope of strategic chances becomes particularly relevant for negotiation performance where one party is disadvantaged. A power imbalance between negotiators is one of these scenarios, as it affects the negotiation process and performance measures differently (Mannix 1993; Mannix and Neale 1993; Pinkley, Neale, and Bennett 1994; Mannix, Neale, and Northcraft 1995; Kim and Fragale 2005). Several researchers have agreed that a higher power position positively affects a negotiation’s individual outcomes (Pinkley, Neale, and Bennett 1994; Giebels, De Dreu, and van de Vliert 2000; Wolfe and McGinn 2005). Situational power enables negotiators to influence a negotiation process favorably, mainly due to the control of resources and the ability to influence the reward structure (Greenhalgh, Neslin, and Gilkey 1985). Low‐power negotiators are disadvantaged, as they must often adapt instead of proactively assert their own interests (Giebels, De Dreu, and van de Vliert 2000; De Dreu and Van Kleef 2004; Anderson and Kilduff 2009; Wiltermuth, Tiedens, and Neale 2015).

Consequently, negotiating parties strive to increase their power. For example, researchers have observed the efforts of weaker countries to match the power of stronger countries (Rubin and Zartman 1995). Kim, Pinkley, and Fragale (2005) have emphasized that a negotiation’s power structure is not fixed. Unfortunately, there has been little research about appropriate opportunities to increase power in a negotiation. We suggest that the presence of a 0‐preference for an issue (chance issue) will offer a good opportunity to overcome the more limited room for maneuvering of weaker parties and allow them to improve their negotiation performance. This is because weaker negotiators are often the ones who are already under a lot of pressure during a negotiation, as the counterparty has a better BATNA (Wolfe and McGinn 2005). It is particularly difficult for weaker negotiators to offer counterparties options that are attractive to the counterparties but still benefit themselves.

This article aims to highlight the strategic potential of a specific type of issue analysis (chance issue analysis) for negotiation scenarios with issue preference differences and to examine its benefits for unbalanced power conditions (one side has high power and the other low power). Our work investigates this new perspective on issue preference analysis and its impacts. Following a review of relevant literature on preferences and power in negotiations, we present an experimental study we designed to examine the impact of one‐sided preferences.

Uncovering preference differences when systematically analyzing issues during preparation enables stakeholders to gather important information and helps uncover structural and behavioral patterns that become critical to the process and outcome (Peterson and Lucas 2001; Lewicki, Barry, and Saunders 2010; Kirgis 2012). For example, information about the other party’s evaluation and prioritization of issues allows conclusions on the underlying interests (Giacomantonio, De Dreu, and Mannetti 2010). This information enables better planning and the use of strategic steps as well as higher process continuity in a negotiation (Peterson and Lucas 2001; Peterson and Shepherd 2010). By undertaking issue analysis, negotiators are able to prioritize the identified issues in an upcoming negotiation according to the relevance of the issues both to themselves and the opposing party (Lewicki, Barry, and Saunders 2010). The analysis reveals preference differences that can be due to different resource access, utility, or cost structures behind the issues (Thompson 2015). Through issue analysis, a party may examine how it benefits from the counterparty’s preferences and how such preferences may enable it to obtain the best outcome from the negotiation.

Issue analysis can identify three main types of preferences:

• Two‐sided preferences:

  • a)

    Both parties have high preferences for this object (e.g., price); both parties are striving for the best individual outcome on this issue.

  • b)

    Both parties have preferences, but they differ in intensity (e.g., payment terms). Both parties are striving for the best individual outcome on this issue, which can be different depending on the underlying benefit structures.

• One‐sided preferences: Only one party has a preference for this issue; only the party with a preference tries to achieve an outcome that accords with its underlying value evaluation.

• No party preferences: These items are usually not part of a negotiation.

Negotiations with preference differences are of particular interest, as the negotiators have to agree on different issue manifestations. The resulting complexity (Northcraft et al. 1998) is one of many reasons for a closer look at preference differences and the impact of these differences in negotiation.

Preference differences can affect the nature of an upcoming negotiation, which is often determined by the number of involved issues but also their suitability for joint value creation (Lewicki, Barry, and Saunders 2010; Kirgis 2012). Negotiation issues can be compatible (both parties agree on the same option), distributive (an opposing party gains everything the other party gives up), or integrative (various options may be available to increase the joint outcome). The nature of a negotiation itself is therefore characterized as distributive or integrative and entails either a fixed or a variable win set (Raiffa, Richardson, and Metcalfe 2007).

Distributive negotiations usually result from different preferences toward issues if they describe the exact opposite—for example, one party strives for high values on that item and the other party for low values (Raiffa, Richardson, and Metcalfe 2007). In this case, it is difficult to reach an agreement because the total benefit available to the parties (colloquially known as “the size of the pie”) is fixed and must be split between the parties (Peterson and Lucas 2001). In contrast with distributive bargaining, integrative bargaining requires preferences that are different but not conflicting, thus allowing for trade‐offs without losing overall value (Lewicki, Barry, and Saunders 2010). In integrative bargaining, logrolling—a process whereby a party offers a concession on one item that is considered less important in order to gain an advantage on another item with more value (Tajima and Fraser 2001; Thompson 2015)—is recommended. In this way, opportunities to maximize the win set for both parties can be created. A negotiator’s choice of behavior and tactics will therefore differ, depending on the nature of a negotiation (Kirgis 2012) and the underlying preference structure for issues (Northcraft, Brodt, and Neale 1995; Curhan, Neale, and Ross 2004). To summarize, an examination of the parties’ underlying preferences, especially the identification of differences in preferences, enables better strategic planning and preparation before a negotiation.

Research has shown that preferences are primarily shaped by each negotiator’s subjective evaluation. The type and intensity of subjective evaluation for single negotiation issues are generally seen as crucial indicators of different dynamics and dependencies in a negotiation process and also predict negotiation outcomes. A higher individual evaluation of an attribute leads to higher offers, a lower willingness to concede, and higher outcomes (Vetschera 2007; Lewicki, Barry, and Saunders 2010). The lower a party values an issue, the less resistant it is to concede on that issue, whereas a higher evaluation will elicit greater resistance and will push the opposite party to move away from its resistance point (Lewicki, Barry, and Saunders 2010). On the other hand, having a higher preference for an issue could also be disadvantageous, as the party will be less flexible on the issue and consequently more dependent (Lax and Sebenius 1985). This can eventually put pressure on the party to adapt on other issues.

Although these studies provide a great deal of insight into the interaction of negotiation behavior and negotiation performance under two‐sided preference conditions, other context‐specific questions remain. For example, no research has dealt with a scenario where only one party has a preference toward an issue and the other has none at all. More concretely, it could be assumed that there might even be issues where one party has no preference (in mathematical terms, a “0”‐valuation or 0‐preference). This issue would neither add nor subtract value for the party during a negotiation, but will remain on the negotiation table if the opponent has a preference for it. In this specific case of one‐sided preferences, only one party regards an issue as important. This issue might afford a chance for one negotiation party and is therefore called a chance issue. Performing an issue and preference analysis according to this novel perspective requires a new approach. So far, preferences in numerical terms have most often been described with value functions and individual utilities (Sebenius 1983; Lax and Sebenius 1986; Hindriks, Jonker, and Tykhonov 2006). Nonetheless, it seems unrealistic that negotiators are always aware of a concrete mathematical function regarding their preferences or issue valuation. It seems more applicable to work with pay‐off matrixes or with a comparison, which structures the issues according to different priorities (Watkins 2003; Lewicki, Barry, and Saunders 2010). In doing so, negotiators subjectively weigh and prioritize issues before a negotiation to calculate potential trade‐offs and develop strategy (Lax and Sebenius 1986; Keeney, Raiffa, and Meyer 2003). We want to introduce a specific approach of issues analysis—the chance issue analysis—that particularly helps us to identify issues on which the opponent has a one‐sided preference.

Using a concrete example, the idea can be explained as follows: First, each issue that is at stake for a negotiation should be evaluated separately, always considering both negotiation parties. Consider a buyer–seller negotiation over a punch press. Three issues are up for negotiation: price, payment terms, and delivery date. Another potential aspect of negotiation could be the extension of the quality guarantee. This issue should also be considered, as it may function as a chance issue. Each issue then is analyzed based on the question of whether any preference exists, described by the numeric value “1” if a preference exists or “0” if there is no preference. For issues that are evaluated with a 0, regardless of what is agreed on, value for the party is neither added nor subtracted (Table One).

In this example, two issues (payment terms and quality agreement) are at stake, which differ in preference positions on an aggregated level (one‐sided preference). More concretely, the existence of preference for issues from both parties should be set in relation to each other, mapped in the form of a matrix. As such, a chance issue can be identified (Figure One).

Identifying preference positions with a chance issue analysis brings to light potential ways of dealing with the negotiation issues. For example, a negotiator with a 0‐preference could take advantage of the preference imbalance by using the chance issue for preferred trade‐offs and thereby higher realized gain, assuming the opposite party is more dependent on that issue due to its relevance evaluation. The counterpart in this scenario faces a risk, as it has a preference for this issue and is probably not readily willing to concede. Contrary to the existing research, which stresses that high valuation is beneficial as it can lead to a higher individual performance (Lax and Sebenius 1985; Vetschera 2007; Lewicki, Barry, and Saunders 2010), we can assume it might have the opposite effect. If so, there could be opportunities for negotiators facing a 0‐preference to improve their own negotiating position, the identification of which would constitute a valuable contribution to existing research in preference analysis and negotiation strategies. Against this background, our first hypothesis is:

Both economic and noneconomic outcomes are measures of negotiation success. Higher objective results are not necessarily perceived as such by the negotiators (Galinsky, Gruenfeld, and Magee 2003; Neale and Fragale 2006) and noneconomic outcomes are known to be more sustainable (Graham 1987; Patton and Balakrishnan 2010). Three variables were identified to measure alternative (non‐numerical) negotiation outcomes: satisfaction with the negotiations, perceived efficiency, and perceived effectivity. Satisfaction with the negotiation process and agreement determines whether further negotiations or cooperative behavior will follow, promoting long‐term relationships (Graham 1987; Rubin, Pruitt, and Kim 1994; Curhan, Elfenbein, and Kilduff 2009; Patton and Balakrishnan 2012). This seems desirable, as strong and cooperative business relationships become increasingly important, especially in the context of international business (Saee 2008; Chang, Wang, and Kao 2010; Patton and Balakrishnan 2010; Tangpong et al. 2015). The evaluation of negotiation results can also be affected by the perceived efficiency and effectivity of the bargaining process. These variables describe how well and under which circumstances negotiation goals or solutions were found (Pruitt 1981; Butler 1999; Hyder, Prietula, and Weingart 2000; Purdy, Nye, and Balakrishnan 2000) and are therefore useful indicators for a negotiator’s success.

Following our argumentation above, we believe that the strategic potential of 0‐preferences will lead to higher rankings in satisfaction with outcomes, higher perceived efficiency, and effectivity. This seems logical due to the added options when facing a 0‐preference: it reveals higher opportunity to proactively influence the negotiation process and to make use of the strategic chance by striving for preferred options (Froman and Cohen 1970; Thompson and Hastie 1990; Tajima and Fraser 2001) as well as to overcome potential deadlocks (Pruitt 1983). In addition, it enables quasi‐guaranteed agreement on at least one item, which—if played tactically—can lead to the fulfillment of objectives in other items, and therefore higher satisfaction. Therefore, our second hypothesis is:

It seems reasonable to assume that more options for agreements will promote joint performance, as long as there is enough integrative potential in a negotiation. If one party has no preference for an issue, there are more options for finding agreement or exchanging ideas. This is similar to our findings on preference differences and logrolling, where information exchange enables trade‐offs and more integrative behavior (Weingart et al. 1990; De Dreu, Koole, and Steinel 2000) and brings us to our third hypothesis:

One can also suppose that the beneficial impact of the 0‐preference depends on the negotiation. After all, we know that besides preference differences, the power balance between negotiators determines the negotiation process in general and negotiation tactics and behavior in particular (Van Kleef et al. 2006; Magee, Galinsky, and Gruenfeld 2007; Greer and Bendersky 2013). Bargaining power is defined as an advantageous position during negotiation that allows claiming the larger amount of the overall value due to resource control and enables assertiveness regarding one’s own interests (Magee, Galinsky, Fragale, and Gruenfeld 2007; Lewicki, Barry, and Saunders 2010; Fragale, Overbeck, and Neale 2011; Belkin, Kurtzberg, and Naquin 2013). Many researchers agree that a higher power position affects the individual outcome of a negotiation positively (Pinkley, Neale, and Bennett 1994; Giebels, De Dreu, and van de Vliert 2000; Wolfe and McGinn 2005). It is also suggested that the absence or loss of bargaining power does not necessarily entail a decrease of demands regarding negotiation issues (Wolfe and McGinn 2005). Consequently, striving for more power during negotiations seems desirable and Kim, Pinkley, and Fragale (2005) affirm that negotiators can indeed change their power position. However, the question as to how negotiators can appropriately go about changing their power position remains unanswered.

Bacharach and Lawler (1981) mention some power change tactics, whereas Fisher, Ury, and Patton (2011) also provide general recommendations, including on how to improve one’s own alternatives. Nevertheless, there is agreement that power is mainly determined by a party’s dependence on the opposite negotiator and that to overcome a weak negotiation position that is due to low power, the interdependence ratio of the negotiators has to be balanced out (Bacharach and Lawler 1981; Wolfe and McGinn 2005; Druckman and Olekalns 2011). Thus, negotiators should reconsider decreasing their own dependence and instead attempt to increase the other party’s dependence (Lawler 1992) to gain higher situational power by having more influence in resource allocation (Greenhalgh, Neslin, and Gilkey 1985). Kim and Fragale (2005) affirm that a strategy to increase power should involve enlarging the opponent’s dependency by escalating its valuation of its own negotiation behavior and thereby stimulating concessions.

The identification of chance issues (with relevance for the opponent but not for oneself) might be useful. Recall that according to preference theory, negotiators are less flexible in adjustments of important issues and strive for satisfaction of preferred agreement levels, particularly on these important issues (Lax and Sebenius 1985; Vetschera 2007; Lewicki, Barry, and Saunders 2010). More concretely, one can assume that the negotiator is not fully independent due to the valuation of these important issues, even if the negotiator is generally in a high‐power position. Consequently, it is apparent that this particular negotiation provides a valuable opportunity for the low‐power party to influence the interdependence ratio, and in that way also the power imbalance. The fact that high‐power negotiators are even more likely to communicate their preferences (Overbeck, Neale, and Govan 2010) could particularly help the weaker party to identify or reevaluate its own estimations of chance issues during a negotiation. It can be assumed that the effect of a 0‐preference is beneficial for high‐power negotiators in any case, as it might simply add to their existing power. For this reason, we want to focus on parties in low‐power positions. The difference effect of having a 0‐preference or not having a 0‐preference could be particularly relevant for them, as one might conclude that the advantageous leverage seems to be stronger here. Following the discussion above and concurring with our argumentation from Hypotheses 1 and 2, we therefore assume:

Summarized, whether a negotiator has high or low power affects the negotiator on an individual level. It also affects the overall character of a negotiation, depending on whether there is a balanced or unbalanced power situation when considering the power relationship type on an aggregated level. There is comprehensive research on behavioral patterns in negotiations according to the power position and relationship (Giebels, De Dreu, and van de Vliert 2000; Galinsky, Gruenfeld, and Magee 2003; De Dreu and Van Kleef 2004; Van Kleef, De Dreu, and Manstead 2004; Magee, Galinsky, and Gruenfeld 2007; Overbeck, Neale, and Govan 2010; Jeuken et al. 2015). When comparing balanced‐power negotiations with unbalanced‐power negotiations, some findings are worth discussing for this work, especially with regard to joint performance goals.

The existing research is not consistent as to whether it is easier or harder to reach integrative agreements in balanced or unbalanced power situations. Some researchers have found that a balanced power situation facilitates cooperative and effective negotiations as a power balance increases one’s motivation for understanding the opposite side and its interests, as well as the degree of information sharing (McGinn and Keros 2002; Wolfe and McGinn 2005). At the other end of this continuum, a power imbalance leads to lower negotiation efficiency (Giebels, De Dreu, and van de Vliert 2000; Zartman and Rubin 2000; Wolfe and McGinn 2005) and more competitive behavior (Lawler 1992; Greer and Bendersky 2013). According to these researchers, power asymmetry hinders negotiators from focusing on integrative bargaining behavior (Zartman and Rubin 2000) and the stronger party often gets the bigger share of the “negotiation pie” at the expense of the low‐power party (Wolfe and McGinn 2005).

However, several other studies have found that cooperative choices are more often made in unbalanced power relationships, leading to higher joint gains in such negotiations (Komorita, Sheposh, and Braver 1968; Tedeschi, Bonoma, and Novinson 1970). Mannix and Neale (1993) concluded that this is mainly due to the weaker party acting more integratively out of a need to realize some gains. The weaker party carefully considers available options, which leads to more efficiency in unbalanced power scenarios. We believe that 0‐preferences could particularly help low‐power parties to realize these integrative approaches and will promote higher joint gains in addition to increased individual outcomes. It is therefore logical to assume that low‐power parties would strive to use the chance issue for beneficial strategic moves to realize gains in a negotiation. We therefore assume:

To further our investigation of a new perspective on preference analysis and the identification of strategic options, and against the background of the literature review, we designed an experimental study to examine the impact of one‐sided preferences.

A total of 168 business and law students (46 percent of them female) from two large state universities participated in a simulated online negotiation with the incentive of extra course credits. Negotiation performance was debriefed afterward in class, and this was communicated beforehand, creating extra motivation to perform well. All students had prior bargaining experience due to exercises and lectures in negotiation classes. They therefore met criteria of comparable negotiation practice from practitioners (Herbst and Schwarz 2011). Data from ten participants were excluded from the analysis, as they ended in an impasse or failed the manipulation check, leading to a sample size of 158 participants. Each negotiation had two participants, rendering 79 negotiation dyads.

A mixed method approach was used for data collection, combining a negotiation simulation with a post‐experimental questionnaire. The experiment took place as online negotiations, using a closed online chat‐based negotiation platform. The negotiation exercise was based on the “New recruit” case (Polzer and Neale 1995)—a distributive multi‐issue negotiation with a fixed bargaining zone, but with the potential for integrative trade‐offs. The participants had to negotiate a job contract with eight issues (signing bonus, division, vacation days, starting date, reimbursement of moving costs, insurance, salary, and location). Each role description entailed a pay‐off table on which points were assigned to every issue and its variations for the five potential agreement levels to represent the negotiators’ preferences. The bargaining zone ranged from a minimum of −8,400 points to a maximum of 13,200 or 10,800 (depending on preference manipulation) points.

Each negotiation lasted thirty minutes. Participants signed up for the negotiation exercise and the system randomly matched participants in negotiation dyads and to their role of recruiter or candidate. Prior to the negotiation, participants received a short user manual with general instructions, the negotiation case, and their specific role description via email. In order to meet relevant requirements for the successful implementation of an online simulation game (Asakawa and Gilbert 2003), the user manual contained information about the use and procedure of the online negotiation, with notes stressing careful preparation and confidentiality of the individual role descriptions and individual pay‐off table. Online negotiations were administered and participants received information about the official start time, the remaining time, and the end of the negotiation. Once thirty minutes expired, participants were presented with an online post‐negotiation questionnaire.

The experimental study had a 3x3 factorial design with within‐subject manipulation: (a) high power vs. low power vs. balanced power and (b) 0‐preference vs. one‐sided preference vs. two‐sided (both) preference. Participants were randomly assigned to a role and power as well as a preference condition, regardless of their role assignment. Power was manipulated by providing either an attractive BATNA for the high‐power condition (an alternative contract offer with 6,400 points overall) or an unattractive BATNA for the low‐power condition (an alternative contract offer with 2,300 points overall). Additionally, participants in these conditions received a specified role description (Van Kleef et al. 2006). The 0‐preference (chance issue) was manipulated by assigning zero points to one issue and its variations in the pay‐off table for one negotiation side, and entailed a short explanation in the role description. As no significant differences were found for the candidate or recruiter role (Pinkley, Neale, and Bennett 1994; Polzer and Neale 1995), all analyses were collapsed across roles.

Individual performance was measured relatively by the total point score out of the maximum possible number of points (13,200 or 10,800 if it was a 0‐preference manipulation) to ensure comparability among different preference manipulations.

Joint performance was measured on a dyad level and relative to the maximum possible integrative gains. The integrative potential for each issue (the best solution with the highest points, when comparing both pay‐off matrixes) was therefore identified and added to an overall level of potential integrative gain (13,200 points in all conditions). As an example, a joint outcome (the sum of each negotiator’s individual points) of 6,600 points equals an integrative exploitation rate of 50 percent.

Noneconomic outcome was measured by three separate constructs (Voeth, Herbst, and Gawantka 2006), namely, general satisfaction with the negotiations, efficiency, and effectiveness within a post‐questionnaire on five‐point Likert scales. Each measure was calculated by averaging all respective items to one overall score, leading to individual ratings for the three outcome dimensions as means. A low value reflects a positive perception of the relevant construct (more satisfied, more efficient, and more effective negotiation process). Satisfaction was measured with five items, for example, “I am fully satisfied with the result achieved” (scale reliability: α = 0.944; excellent). Efficiency was measured with four items. For example, “Overall, the negotiation can be classified as very goal‐oriented and structured” (scale reliability: α = 0.724; acceptable). Effectiveness was measured with four items, for example, “In the present negotiation situation, no better result would have been possible” (scale reliability: α = 0.840; high). For data analysis (H2, H4), a mean score from all three variables was calculated to one variable, namely “noneconomic outcome” (scale reliability: α = 0.807; high).

Participants completed a post‐negotiation questionnaire that also contained manipulation checks.

Power: Participants’ sense of power was measured with two constructs, namely experienced and relative power, adapted from Van Kleef et al. (2006). Experienced power was measured with four items on a five‐point Likert scale, with ratings from 1 = strongly agree to 5 = strongly disagree, for example, “I felt I had a strong negotiation position during the negotiation.” Two questions were reverse‐coded. The four items were averaged to a single index for experienced power (scale reliability: Cronbach’s alpha α = 0.779; acceptable). Relative power was measured with four items, with ratings on a scale of 1 = definitely me to 5 = definitely my opponent, for example, “Who do you think was most dependent on the other negotiator?” These four items were also averaged to a single index for relative power (scale reliability: α = 0.770; acceptable). To check for BATNA manipulation, participants had to evaluate power due to having an alternative.

Preference: To check whether participants were aware of their preferences, they had to rank all eight negotiation issues in order of importance for themselves and then for their opponent. The questionnaire also included two questions about the perceived strategic benefit due to preference differences (1. “I felt I could strategically use preference differences regarding negotiation issues during the negotiation” and 2. “I felt I could strategically (and especially) use issues that were less important to me during the negotiation”) and were rated on a five‐point Likert scale of 1 = strongly agree to 5 = strongly disagree.

The manipulation check for power was conducted using a one‐way analysis of variance (ANOVA) to examine evaluation differences for the variables experienced power and relative power between the three power groups. The results show a statistically significant difference between all condition groups regarding the experienced power (p < 0.001). A Tukey HSD post hoc test revealed that the participants in the high‐power group assessed their own power as statistically significantly higher (M = 2.64, SD = 0.70) than the participants in the low‐power group (M = 3.57, SD = 0.82, p < 0.001) and higher than the participants in the balanced power (control) group (M = 3.03, SD = 0.73, p < 0.001). Significant differences were also found for relative power (F(2, 155) = 9.37, p < 0.001). A post hoc test revealed significantly higher evaluations regarding relative power between high (M = 2.64, SD = 0.44) and low (M = 3.01, SE = 0.46, p < 0.001) power conditions. These results indicate that power was successfully manipulated. A common assumption in the “New recruit” case is that the role of the recruiter automatically entails a higher power position than the role of the candidate (Overbeck, Neale, and Govan 2010). As the participants in this study were students, with little experience in job interviews, it is even more likely to apply here. To counter this, power manipulation was implemented with an alternative (BATNA) and the relevant role description. To check on this, the correlation between assigned role and self‐rated feelings of power was calculated. The results confirmed that perceived power was independent of the role as there was no correlation (r (158) = 0.025, n.s.). Evaluations of relative power were also independent of role (r(160) = 0.020, n.s.).

A manipulation check for preferences was conducted in the same way, using a one‐way ANOVA to examine evaluation differences for importance ranking of the issue with 0‐preference (starting date) between the three preference groups. Participants with 0‐preference condition correctly evaluated the 0‐preference issue (starting date) as the least important issue (M = 7.77; ranking number 8), whereas the two‐sided (M = 4.83, ranking number 5) and one‐sided preference conditions (M = 4.49; ranking number 5) evaluated the issue correctly as middle ranked. This difference was significant (p < 0.001) when comparing groups of 0‐preference with the two‐sided and one‐sided‐preference conditions. Generally, participants ranked the negotiation issues accurately with importance due to gained points per issue and levels of agreement on that issue. It can therefore be assumed that the manipulation was successful. The manipulation check on preferences and strategic use revealed a correlation between preference condition and higher perceived usefulness of preference differences in general (r (158) = 0.180, p < 0.05, low effect), and of less important issues in particular (r (158) = 0.243, p < 0.01, medium effect). A one‐way ANOVA showed that the overall perceived strategic use of preference difference (considering the sum of the means of both variables) was significantly higher in 0‐preference condition (M = 1.96, SD = 0.95) compared to the two‐sided preference (control group) (M = 2.55, SD = 0.81; p < 0.001) and higher than the one‐sided preference (M = 2.00, SD = 0.73, n.s.). It can therefore be assumed that this manipulation also triggered participants in the 0‐preference condition to perceive a higher strategic potential of preference differences, particularly of issues with no preference (starting date).

In line with the structure of the hypotheses, we followed a two‐step approach to analyze the data. We first tested Hypotheses 1 to 3 by conducting an ANOVA to determine the impact of the issue preference on the individual economic outcome (H1), the individual noneconomic outcome (H2), and a t‐test due to dyad level for joint performance (H3). In a second step, we analyzed the moderating role of power (Hypotheses 4 to 6) by conducting a two‐way ANOVA to measure the impact of the two independent variables—issue preference and power position—simultaneously on the relative individual performance (H4), noneconomic outcome (H5), and joint performance (H6) as dependent variables.

Individual economic outcome: A one‐way ANOVA was conducted to test the effects of preference condition on the relative individual outcome (H1) and revealed marginal significant differences between groups (p < 0.1) (Table Two).

A post hoc test (Table Three) revealed that negotiators facing 0‐preferences achieved significantly (p < 0.1) higher individual outcomes than negotiators facing two‐sided preferences and higher outcomes than one‐sided preferences, proving Hypothesis 1.

Noneconomic outcome: A one‐way ANOVA revealed significant differences of noneconomic outcomes (satisfaction, efficiency, and effectivity). Table Four shows that the different preference conditions were a significant source for the individual noneconomic outcome differences (p < 0.05).

A post hoc comparison (Table Five) revealed that negotiators facing 0‐preferences reported significantly (p < 0.05) higher noneconomic outcomes than negotiators facing two‐sided preferences and higher outcomes than one‐sided preferences, proving Hypothesis 2. When checking on the three separate categories, efficiency and effectivity differed the most, especially when comparing 0‐preference condition with a two‐sided‐preference condition. Generally, 0‐preference condition led to the highest values in all three categories.

Joint performance was measured on a dyad level, not an individual one, leading to two groups with either an asymmetrical preference condition (one‐sided or 0‐ preference) or a balanced preference condition (two‐sided preference). A t‐test was used to compare means and revealed significant (p < 0.01) differences in joint performance between the two groups. Negotiation dyads where one negotiator had no preference for one issue reached significantly higher joint performance (M = 0.41, SD = 0.09) than negotiation pairs where both negotiators had a preference for the issue (M = 0.34, SD = 0.08; p < 0.01). We can therefore accept Hypothesis 3 (Table Six).

To evaluate the moderating role of power on the relationship between the preference condition and individual relative performance, we first conducted a two‐way ANOVA. We compared the impact of this relationship on components with a high, low, and balanced degree of power. This analysis allowed assessing the main effects of the two independent variables as well as the interaction effects between them. Regarding the main effects, the results as shown in Table Seven indicate that the type of power condition was the most significant source for the deviation between individual performances. However, the type of preference also had a significant effect.

Our analysis did not suggest significant interaction between the type of issue preference and the type of power situation on the deviation of relative individual performance. High power always led to higher individual performance than balanced power or low power, regardless of the type of issue preference. However, 0‐preference was also always the best issue preference condition with regard to individual outcome. Significant differences were found between high‐power (M = 0.54) and low‐power (M = 0.35) conditions. However, when comparing outcome differences from a two‐sided preference condition with 0‐preference condition, a performance increase is obvious. For high‐power parties the performance growth is 22.7 percent, for balanced power parties it is 24.2 percent, and for low‐power parties it is 25 percent. Thus, even though low‐power parties still have lower absolute performance scores when facing 0‐preference compared to the other power groups, the performance increase (leverage effect) is slightly higher for low‐power than balanced and high‐power parties.

To summarize, whether a 0‐preference leads to success or not does not depend on the power position. It always leads to a better result. The same applies to the effect of power: high power will always lead to a better result, regardless of the preference condition. However, low‐power parties profit from 0‐preference and the leverage effect was highest for them when comparing it to the other power groups. We can therefore confirm Hypothesis 4.

To evaluate the moderating role of power on the relationship between the preference condition and noneconomic outcome, we also conducted a two‐way ANOVA. The results show that power has a statistically significant moderating role on the relationship between the preference condition and noneconomic outcomes (p < 0.001) and that both independent measures have a significant effect on the deviation of noneconomic outcomes (Table Eight). Figure Two shows the comparison between the power group and preference conditions.

The analysis revealed that the impact of 0‐preference on higher values in noneconomic outcomes is strongest for a balanced degree of power compared to a low or high degree of power. Negotiators in the balanced power condition and with 0‐preference in one issue (M = 1.89, SD = 0.47) reported significantly higher noneconomic outcomes than negotiators with one‐sided (M = 2.7, SD = 0.81, p < 0.05) or two‐sided preferences (M = 2.65, SD = 0.74, p < 0.05) in a balanced power situation. Interestingly, 0‐preference condition in a balanced power situation (M = 1.89, SD = 0.47) even led to significantly higher evaluations than in high‐power positions (M = 2.70, SD = 0.65, p < 0.05). Noneconomic outcomes were also significantly higher for low‐power negotiators when having 0‐preference (M = 2.36, SD = 0.66) compared to facing a two‐sided preference (M = 3.12, SD = 0.71, p < 0.05). Interestingly, this effect was different for negotiators with a high‐power position. Here, 0‐preference led to the lowest values in noneconomic outcomes. However, no significant differences for high‐power negotiators were found with regard to preference condition. Our results therefore partially confirm Hypothesis 5. A 0‐preference for an issue led to higher noneconomic outcomes, also for low‐power negotiators. The impact was however strongest in a balanced power situation and, interestingly, not advantageous for high‐power parties with regard to noneconomic outcomes.

To evaluate the moderating role of power on the relationship between the preference condition and joint performance, we conducted a two‐way ANOVA, similar to H3 considering the dyad level. Regarding the main effects, the results as shown in Table Nine indicate that only the type of dyad preference situation (two‐sided or one‐sided) was the most significant source for the deviation between relative joint performances. There was no effect for the dyad power group (balanced or unbalanced) and no interaction effect was found.

The one‐sided preference condition always led to a higher joint performance (M = 0.41, SD = 0.09) than the two‐sided preference condition (M = 0.34, SD = 0.08) in negotiation dyads (p < 0.01), regardless of the power situation (balanced or unbalanced). No significant differences were found for joint performance when comparing unbalanced with balanced power situations in negotiation dyads. Joint performance was, however, higher in unbalanced power dyads (M = 0.40, SD = 0.10) than in balanced power dyads (M = 0.37, SD = 0.79). To summarize, the fact that one‐sided preferences (chance issues) enable higher joint outcomes does not depend on the power situation in the negotiation dyad. We can therefore partially confirm Hypothesis 6.

A great deal of research has focused on two‐sided preferences and their differences, establishing that high valuations of issues have a positive impact on negotiator performance (Lax and Sebenius 1985; Vetschera 2007; Lewicki, Barry, and Saunders 2010). This experimental study extends the research on preference differences with a new perspective, that is, by preference differences toward negotiation issues due to a 0‐preference on one side of a negotiation. The results indicate initial verification that no preference at all for one issue could lead to higher individual performance and higher individual noneconomic outcomes. Joint performance was positively affected by 0‐preference and increased, even in unbalanced power situations. Although this research framework needs additional study with a larger and more generalizable sample, the initial results suggest the potential for a significant contribution to the existing research on issue analysis.

Potential reasons for better performance were suspected to result from a leverage effect. Importantly, participants acknowledged that chance issues could be used strategically and that they felt higher perceived power. Further study is needed to understand fully the mechanism by which these judgments were made from the participants’ point of view. To shed some light on this connection, a post hoc qualitative analysis of negotiation chat texts pulled from the negotiation platform was conducted. This pattern analysis suggested potential links between 0‐preferences and negotiation tactics, providing anecdotal support for at least a correlational relationship between preferences and tactics. This trend warrants further study.

Four main findings are relevant for this study. First, and as expected, the qualitative analysis revealed that the chance issue was often used for preferred trade‐offs. Consistent with logrolling (Pruitt 1983; Tajima and Fraser 2001), less important issues, in this case issues without any importance, were exchanged for other issues. Most often, concessions in particularly important issues were demanded, for example:

• salary, which was distributive and most important for both roles;

• insurance, with high value for the recruiter; and

• reimbursement of moving costs, with high value for the candidate.

This negotiation strategy seems reasonable and even more feasible in the case when facing a 0‐preference toward an issue, as the negotiator will not lose any value.

Second, the chance issue was also deployed in moments when agreement and the negotiation progress seemed to become difficult and stagnant. Research has shown that to overcome negotiation barriers, it could be helpful to introduce (add) new issues (Sebenius 1983). Of course, the issue could not be introduced in a completely new light in this study, as both negotiators were aware of all the negotiation issues from the beginning. Against this background, it should also be noted that a completely new introduction of issues and the new linkage with previous issues of the agreement could also jeopardize the overall result (Poast 2012), for example, if the demands in the “new” issue are too harsh. Although it is likely to assume that there is a willingness to reach an agreement on this issue based on the counterparty’s preference, the overall deal should not be changed too much as a result. Nevertheless, in situations where agreement on the discussed object seems impossible, one could redirect to a new and more positively perceived item, as a chance issue would be due to only a one‐sided preference for one negotiator and could facilitate further negotiation progress. Interestingly, negotiators often relented on the chance issue in the beginning. This might have led to a “smooth” start of the negotiation process, with the perceived intention to cooperate, and might have indirectly influenced the other side to respond similarly with concessions. This assumption is in line with previous research on negotiation behavior that discovered this typical reaction, following the principle of reciprocity (Putnam and Jones 1982; Weingart et al. 1990).

Third, behavioral patterns were found in the 0‐preference condition that are associated with distributive tactics and powerful bargaining. For example, it often happened that a negotiator feigned a preference on the issue of starting date (chance issue) and subsequently relented on that issue, declaring it a huge concession. In addition, bluffs about the actual starting date or pretenses to have strict regulations regarding the chance issue were used. This deceptive conduct is a typical high‐power behavior (De Dreu and Van Kleef 2004; Van Kleef et al. 2006; Kaufmann et al. 2018) and supports the assumption that using the 0‐preference promoted assertive and dominant behavior that might have led to perceived power from the opponent.

Finally, this study unintentionally replicated previous findings (Pinkley, Neale, and Bennett 1994; Giebels, De Dreu, and van de Vliert 2000; Wolfe and McGinn 2005) by confirming that a high‐power position during a negotiation leads to higher performance, which simultaneously justifies the ambition to improve the power position during a negotiation. Complementary to existing findings on the relationship of power and negotiation performance, our study contributes to a (so far) scarce body of literature about power change tactics with the introduced 0‐preference as a strategic potential to influence power relations during a negotiation.

The results of this study demonstrate the importance of identifying additional negotiation issues in general and chance issues in particular. This extends the bargaining zone and broadens the scope of potential actions and agreement options. This is known to promote joint gain but also to function as leverage for a negotiator’s own benefit (Sebenius 1983). The results confirmed that the strategic use of no preference for an issue was recognized. Negotiators were able to increase their individual performance and negotiation dyads reached higher joint performance. The preference change in only one item changed the overall integrative gains. It is therefore definitely worth searching for issues with no preference at all, because it enables more integrative gain, especially in the case of unbalanced power relationships in a negotiation.

If strategically used, the chance issue can help overcome a deadlock during a negotiation and restructure a problem that is blocking the process, for example, with preferred trade‐offs (Sebenius 1983; Sycara 1991). Fells (1996) has reinforced the importance of the identification of opportunities to handle potential deadlocks, preferably during pre‐negotiation preparation. These findings support the line of argument introduced in the beginning of this article regarding issue analysis and its practical relevance.

Especially for low‐power parties, there are implications concerning power adjustment options to diminish their degree of dependence on the opponent. The strategic potential of the chance issue for low‐power negotiators can be unfolded by increasing the dependence on the opponent (Lawler 1992) by means of the strong imbalance of issue preference. The opposing party faces a risk of potential loss in one issue due to its valuation, but a negotiator with a 0‐preference will not lose any value, no matter what concessions it decides to make regarding this issue. It should be noted that this also reveals enormous potential for the low‐power negotiator to adapt behavioral patterns that are then newly reassessed or even perceived as powerful by the opponent (Kim and Fragale 2005). Appropriate actions can, for example, range from appreciated concession to dominant or even deceptive bargaining strategies. That negotiators with high power performed significantly better when having a 0‐preference also indicates that it might be equally important to become aware of potential risk issues (which are simultaneously potential chance issues for the opponent). This becomes increasingly important to avoid a greater power imbalance that favors the high‐power negotiator, especially when finding oneself in a weak negotiating position.

It seems logical to ask if the artificial context of the experimental study limits the generalizability of the findings. However, research with and on electronic negotiations could show that negotiation simulations through online channels are indeed suitable for experimental designs such as this one (Citera, Beauregard, and Mitsuya 2005; Nadler and Shestowsky 2006). Nonetheless, power might have been perceived more obviously in person, which could then have led to a stronger effect of power change, as nonverbal cues like gestures and facial impressions were absent in the online negotiation chat format (Friedman and Belkin 2013). Against the backdrop of online negotiations becoming more and more relevant, it is a valuable contribution to a realistic negotiation research design (Naquin and Paulson 2003).

The use of student samples for negotiation simulations and their transferability to the professional setting is also controversial. However, some studies and discussions have concluded that results from negotiation simulations with nonprofessionals (like students) could be valid for practitioners as well (Neale and Northcraft 1986; Croson 2005). Previous training or theoretical introduction on negotiation for these students would further mitigate this issue of generalizability (Herbst and Schwarz 2011).

It might be worth investigating the interaction effect of power and preference in more detail. For example, future studies could explore if and how holding 0‐preferences in negotiations affects the perception and awareness of power potential during the bargaining process. Manipulation checks should ensure that participants give information about why they felt more powerful and what they did to act out their power position, as well as ask the opposing party why they perceived the other side to be powerful or not. Such procedures seem reasonable, as being in the more powerful position does not necessarily mean that the power is used. Because negotiators with a 0‐preference performed even better in this study when they used the chance issue more than once, it might be fruitful to conduct research examining these findings to determine the timing and order for introducing or discussing chance issue(s) in a negotiation agenda that would optimize the bargaining process.

Lastly, it should be noted that preference determination is regarded critically in negotiation research. It can of course be argued that asymmetrical information may lead to an inaccurate assessment of the opponent’s preference valuation. Nonetheless, it seems plausible that long‐term business relationships, in particular, provide relevant insights into potential chance (and risk) issues. There are critical voices when it comes to structuring preferences and preference changes for issues during a negotiation process (Gimpel 2007). Scholars have raised concerns about the assumption that preferences are static and remain stable during a negotiation. They encourage considering preference changes during a negotiation process instead, as the availability of relevant information can change and may potentially lead to a new and diverging evaluation or prioritization of negotiation issues. Weber and Johnson (2006) note research in psychology that indicates that preference formation happens in the course of the real decision process. These considerations give rise to the importance of conducting issue analysis not only in preparation for negotiation but also during a negotiation.

Tajima and Fraser (2001) have noted that there is a difference between prioritizing one issue over another and the intensity of preference for a specific issue. Research has explored nonlinear (convex or concave) value functions and their impact (for example, on information collection) during negotiations (Northcraft, Brodt, and Neale 1995; Northcraft et al. 1998; Vetschera 2006, 2007; Ito, Klein, and Hattori 2008; Fatima, Wooldridge, and Jennings 2009). This research could be expanded with additional inclusion of chance issues and would give more in‐depth consideration to the subject of value functions toward negotiation issues.

Our work on issue analysis demonstrates that both before and during a negotiation, negotiators should learn about the potential asymmetrical strengths of two‐sided preferences and identify chance issues and potential risks. The experimental study results confirmed that it is always beneficial to search for negotiation issues that are of no relevance for oneself but might have value for the opposing side. By using such issues strategically during a negotiation process, individual as well as joint outcomes can be improved. Identifying chance issues holds great potential, even for weaker negotiation parties. This finding is a significant contribution to previous research on power change tactics and strategies and has important practical implications. However, the concrete linkage between preference and power position, especially between 0‐preference for issues and specific strategies for gaining power during a negotiation, needs further study. To this end, future studies that can shed more light on this novel approach in negotiation research are needed.

Open access funding enabled and organized by ProjektDEAL.