Artful Paltering: The Risks and Rewards of Using Truthful

Journal of Personality and Social Psychology

2017, Vol. 112, No. 3, 456-473 Public copy created by the Student Social Support R&D Lab

Artful Paltering: The Risks and Rewards of Using Truthful Statements to

Mislead Others

Todd Rogers, Richard Zeckhauser, Francesca Gino,

and Michael I. Norton

Harvard University

Maurice E. Schweitzer

University of Pennsylvania

Paltering is the active use of truthful statements to convey a misleading impression. Across 2 pilot

studies and 6 experiments, we identify paltering as a distinct form of deception. Paltering diﬀers

from lying by omission (the passive omission of relevant information) and lying by commission (the

active use of false statements). Our ﬁndings reveal that paltering is common in negotiations and

that many negotiators prefer to palter than to lie by commission. Paltering, however, may promote

conﬂict fueled by self-serving interpretations; palterers focus on the veracity of their statements

(“I told the truth”), whereas targets focus on the misleading impression palters convey (“I was

misled”). We also ﬁnd that targets perceive palters to be especially unethical when palters are

used in response to direct questions as opposed to when they are unprompted. Taken together, we

show that paltering is a common, but risky, negotiation tactic. Compared with negotiators who tell

the truth, negotiators who palter are likely to claim additional value, but increase the likelihood of

impasse and harm to their reputations.

Keywords: deception, lying, negotiation, paltering, risk

Supplemental materials: http://dx.doi.org/10.1037/pspi0000081.supp

Cite as: Rogers, T., Zeckhauser, R., Gino, F., Schweitzer, M., & Norton, M. (2016). Artful

Paltering: The Risks and Rewards of Using Truthful Statements to Mislead Others.

Journal

Personality

and

Social

Psychology,

112

(3), 456-473.

Jim Lehrer: “No improper relationship”—deﬁne what you

mean by that.

President Bill Clinton: “Well, I think you know what it

means. It means that there is not a sexual relationship, an

improper sexual relationship, or any other kind of improper

relationship.”

Jim Lehrer: “You had no sexual relationship with this

young woman?”

President Bill Clinton: “There is not a sexual relationship—

that is accurate.”

—“NewsHour” With Jim Lehrer, January 21, 1998

Referring to his relationship with Monica Lewinsky, U.S.

President Bill Clinton claimed “there is not a sexual relation-

ship.” The Starr Commission later discovered that there “had

been” a sexual relationship, but that it had ended months

This article was published Online First December 12, 2016.

Todd Rogers and Richard Zeckhauser, Harvard Kennedy School, Har-

vard University; Francesca Gino and Michael I. Norton, Harvard Busi-

ness School, Harvard University; Maurice E. Schweitzer, Wharton

School of Business, University of Pennsylvania.

Correspondence concerning this article should be addressed to Todd

Rogers, Harvard Kennedy School, Harvard University, 79 JFK Street,

Cambridge, MA 02138. E-mail: Todd

og ers@hks.harvard.edu

before Clinton’s interview with Jim Lehrer. During the inter-

view, Clinton made a claim that was technically true by using

the present tense word “is,” but his statement was intended

to mislead: Jim Lehrer and many viewers inferred from Clin-

ton’s response that he had not had a sexual relationship with

Monica Lewinsky. We categorize Clinton’s claim as paltering:

the active use of truthful statements to create a false impres-

sion. We distinguish paltering from both lying by omission

and lying by commission, document the prevalence of palter-

ing, identify important consequences of paltering, and explore

why people prefer paltering to lying by commission.

Deception pervades human communication and interper-

sonal relationships (Bok, 1978): DePaulo et al. (1996) found

that people tell, on average, one or two lies per day. Though

many lies are harmless, some are signiﬁcant and consequen-

tial. One domain in which deception can substantially change

outcomes is negotiations (Bazerman, Curhan, Moore, & Val-

ley, 2000; Boles, Croson, & Murnighan, 2000; Gaspar &

Schweitzer, 2013; Koning, Van Dijk, Van Beest, & Steinel,

2010; Lewicki, 1983; Olekalns & Smith, 2009; Schweitzer &

Croson, 1999; Shell, 1991; Tenbrunsel, 1998). Negotiations

are characterized by information dependence, and negotia-

tors can often exploit their counterpart by using deception

(Lewicki & Robinson, 1998; O’Connor & Carnevale, 1997).

Prior deception research has distinguished lying by com-

mission, the active use of false statements (e.g., claiming the

faulty transmission on one’s car works great), from lying by

omission, the passive act of misleading by failing to disclose

relevant information (e.g., failing to mention any information

about a faulty transmission). We make a novel contribution to

the deception literature by identifying a third, and common,

ARTFUL PALTERING IN NEGOTIATION ROGERS, ZECKHAUSER, GINO, NORTON, AND SCHWEITZER

form of deception: paltering (a term initially highlighted in

this context by Schauer and Zeckhauser [2009]). Rather than

misstating facts (lying by commission) or failing to provide

information (lying by omission), paltering involves actively

making truthful statements to create a mistaken impression.

Though the underlying motivation to deceive a target may

be the same, paltering is distinct from both lying by commis-

sion and lying by omission. Unlike lying by omission, pal-

tering involves the active use of statements, and unlike lying

by commission, paltering involves the use of truthful state-

ments. Like lying by omission, paltering can involve failing

to disclose relevant information, but unlike lying by omission,

paltering involves the active disclosure of true but misleading

information: paltering enables would-be deceivers to actively

inﬂuence a target’s beliefs.

We investigate the prevalence of paltering and conjecture

that deceivers will often choose to palter rather than lie by

commission or lie by omission. A critical antecedent to en-

gaging in deception is self-justiﬁcation (Schweitzer & Hsee,

2002), and a related concern is the need to preserve one’s

moral self-image (Aquino & Reed, 2002). By using truthful,

but misleading statements, those who palter may be able to

eﬀectively mislead others while justifying their behavior and

maintaining a positive self-image. As a result, many deceivers

may prefer to palter than lie by commission. In addition, un-

like lying by omission, paltering does not require thetarget to

be complicit by failing to demand relevant information. That

is, paltering may be a common tactic, because deceivers have

abundant opportunities to palter, and because paltering is

relatively easy to justify.

Our investigation advances our understanding of deception

in several important ways. First, our research makes a con-

ceptual contribution by challenging the paradigm that has fo-

cused on the dichotomy between lies by omission and lies by

commission. Whereas prior work conceptualized deception as

involving dishonest statements (lying by commission) or rel-

evant omissions (lying by omission), we identify paltering as

a distinct and important type of deceptive behavior.

Second, we show that paltering is subject to prediction er-

ror. Palterers hold a mistaken mental model, failing to antic-

ipate how negatively the targets of their palters will perceive

them should they detect their deceit. As a result, paltering

can promote conﬂict. By failing to anticipate its adverse con-

sequences, disputants may be tempted to engage in paltering

and fail to appreciate how paltering can escalate conﬂict. As a

result, developing our understanding of paltering contributes

to our understanding of both deception and conﬂict.

Third, we describe an important distinction between two

diﬀerent types of paltering: prompted (by a question) and un-

prompted. Targets perceive palters to be more unethical when

palters are oﬀered in response to a direct question or prompt,

as opposed to when they are unprompted. This distinction

has implications for deception research, more broadly. Our

ﬁndings suggest that, at least in some cases, deceiving tar-

gets in response to a direct question is more unethical than

deceiving targets proactively.

Finally, we explore paltering in negotiations, a context

in which deception poses a particularly important challenge.

Our ﬁndings contribute to the negotiation literature by doc-

umenting a prevalent behavior that impacts both the negoti-

ation processes and negotiator outcomes. Though our exper-

iments focus on negotiation, we expect paltering to pervade

interpersonal interactions from romantic relationships to for-

eign aﬀairs, whenever individuals or groups are tempted to

mislead others.

Across our studies, we investigate paltering in diﬀerent

contexts (including face-to-face and online negotiations) and

participant samples (including experienced negotiators). We

show that paltering is common and that deceivers prefer to

palter than to lie by commission. In our studies, negotiators

who paltered obtained similar outcomes to those who lied

by commission, but deceivers found paltering easier to jus-

tify than lying by commission. Importantly, although those

who palter believe paltering to be more ethical than lying

by commission, once deceptions is exposed targets judge the

ethicality of the two forms of deception very similarly.

In some respects, however, paltering is very similar to other

forms of deception. Paltering enables deceivers to claim value,

but also increases the odds of an impasse and once exposed,

can cause signiﬁcant reputational harm.

Deception in Negotiation

Deception is an integral part of human communication

(O’Sullivan, 2003). Though people often deceive others for

pro-social reasons (e.g., “Your haircut looks great” or “I en-

joyed reading your manuscript”; Levine & Schweitzer, 2014,

2015), self-serving deception is common. This is particularly

true in negotiations, because individuals can gain a strate-

gic advantage by misleading their counterpart (Anton, 1990;

Aquino, 1998; Carr, 1968; O’Connor & Carnevale, 1997;

Schweitzer & Croson, 1999; Schweitzer et al., 2005). De-

ception is also diﬃcult to detect in face-to-face interactions

(Bond & DePaulo, 2006; Ekman & O’Sullivan, 1991; Ek-

man, O’Sullivan, & Frank, 1999). The opportunity to ex-

ploit a counterpart combined with the diﬃculty targets have

in detecting deception makes deception a pervasive feature

of negotiations (Murnighan, Babcock, Thompson, & Pillutla,

1999; Schweitzer & Croson, 1999). In one study, Murnighan

et al. (1999) found that over one-third of experienced nego-

tiators engaged in deception, and in a related study, Aquino

and Becker (2005) found that 53% of negotiators lied to their

counterpart.

Though several studies have distinguished lies according to

their content (e.g., Anton, 1990; Lewicki, 1983)

, the most

important distinction scholars have made with respect to de-

ception has been between lying by commission and lying by

omission (Bok, 1978; Gaspar & Schweitzer, 2013; O’Connor

& Carnevale, 1997; Spranca, Minsk, & Baron, 1991). Lying

by commission involves the use of false statements, whereas

lying by omission involves the omission of relevant informa-

tion (see Table 1). In general, lying by commission is viewed

more seriously, both legally and morally, than lying by omis-

sion (Shell, 1991; Spranca et al., 1991), and people appear to

be more willing to lie by omission than by commission. For

example, when individuals are not asked about a critical issue,

they often omit information, but when asked a direction ques-

tion they become far more likely to honestly reveal the critical

For instance, Lewicki and his colleagues (Lewicki & Robinson, 1998;

Lewicki & Spencer, 1991) identiﬁed ﬁve categories of questionable

negotiation tactics: (a) traditional competitive bargaining (e.g.,

high demands, low concessions); (b) misrepresentation of informa-

tion (i.e., misleading arguments); (c) bluﬃng (i.e., misleading in-

tentions); (d) information collection (i.e., trading favors or gifts for

information); and (e) inﬂuencing an opponent’s professional net-

work.

ARTFUL PALTERING IN NEGOTIATION ROGERS, ZECKHAUSER, GINO, NORTON, AND SCHWEITZER

Table 1

Dimensions on Which Lying by Omission, Paltering, and Lying by Commission Diﬀer

Dimension Lie by omission Palter Lie by commission

Aversiveness

Veracity of speciﬁc claim(s) None True False

Deceiver’s actions Passive Active Active

Eﬀectiveness

Addresses the relevant issue No Yes, indirectly Yes, directly

Attempt to inﬂuence beliefs No Yes Yes

information (though some do resort to lying by commission;

Schweitzer & Croson, 1999; see also Olekalns & Smith, 2009).

For example, in a negotiation simulation involving the sale of

a computer with a faulty hard drive, Schweitzer and Croson

(1999) found that buyers who did not ask direct questions

were never informed about the faulty hard drive.

We introduce and investigate a novel form of deception:

paltering. To illustrate this form of deception, consider a re-

cruiter interested in hiring a new marketing manager. After

carefully sifting through more than 100 applications, the re-

cruiter identiﬁes a small set of candidates with suitable back-

grounds to interview. Though many of the applicants are im-

pressive, after interviews have been completed, the recruiter

decides that only one candidate, Claire, has the requisite skills

and qualiﬁcations for the job. When the recruiter oﬀers the

job to Claire and begins to negotiate the oﬀer, the recruiter

could convey the false impression of having strong alterna-

tives (and hence more leverage). For example, the recruiter

might say, “There is a great deal of demand for this position

from a large number of impressive individuals. We received

more than 100 resumes, and I interviewed the ten with the

strongest credentials.” Though these statements are truth-

ful, they convey the false impression the recruiter has several

other well qualiﬁed candidates should she not accept the job

oﬀer.

In contrast to paltering, lying by commission involves the

active use of false statement to mislead a target. For example,

the recruiter might lie by commission by stating that, “There

are three other individuals to whom I would happily oﬀer this

position.” Both paltering and lying by commission are active.

They involve the use of statements to create a mistaken belief.

We conceptualize both paltering and lying by commission as

active acts of deception. In contrast to paltering and lies by

commission, lies by omission omit relevant information. For

example, if Claire were to remark, “You must have many other

qualiﬁed candidates,” the recruiter could lie by omission by

remaining silent and failing to correct Claire’s mistaken belief.

The active nature of paltering and lying by commission—

making statements—may cause targets and observers to judge

these acts more harshly than lies by omission (see Knobe,

2003; Leslie, Knobe, & Cohen, 2006).

Paltering as a Distinct Form of Deception

By engaging in deception, individuals can often advance

their economic interests, at least in the short-run. Deception,

however, entails risks and costs. If discovered, reputations and

long-term relationships will suﬀer, and the use of deception

can trigger a psychological cost. Individuals strive to maintain

a positive self-concept (Adler, 1930; Allport, 1955; Kruger

& Dunning, 1999), and the overt use of deception interferes

with their ability to preserve a moral self-image. Mazar et

al. (2008) introduced the term self-concept maintenance to

account for the competing concerns of self-interest and self-

concept to explain why people curb their dishonest behavior.

We build on self-concept maintenance theory to develop

our understanding of paltering. We postulate that potential

deceivers balance two competing concerns: how eﬀective their

deceit will be and how aversive the use of deception will be

to their moral self-concept.

We conceptually distinguish paltering from both lying by

omission and lying by commission with respect to their eﬃ-

cacy, and the extent to which they challenge the deceivers’

moral self-concept (see Table 1). With respect to eﬃcacy, we

consider the extent to which deception is likely to eﬀectively

inﬂuence the beliefs of the target individual. We expect active

forms of deception that use statements to shape false beliefs

(such as lies by commission and paltering) to be more eﬀec-

tive than passive deception (lies by omission), for two reasons.

First, lies by omission require targets to be complicit in the

deception. For lies by omission to succeed, targets must hold

and make known relevant mistaken beliefs and/or fail to ask

relevant direct questions. Second, active forms of deception

harness conversational norms to shape false beliefs. Both pal-

tering and lies by commission communicate information that

is relevant and informative. Even if targets were initially un-

sure or skeptical, paltering and lies by commission can lead

targets to form false beliefs.

We also predicted diﬀerences among the three forms of de-

ception with respect to their tendency to allow deceivers to

maintain their moral self-concept; the three forms of decep-

tion aﬀord diﬀerent opportunities for justiﬁcation, which is a

critical antecedent for the deception process (Gino & Ariely,

2012; Schweitzer & Hsee, 2002; Shalvi, Gino, Barkan, & Ayal,

2015). We expect passive deception, lies by omission, to pose

the smallest challenge to one’s moral self-image. Individuals

can justify lying by omission by pointing to the fact that they

did not change the target’s beliefs, that the target was respon-

sible for failing to investigate the relevant issue, and that they

did not actively mislead the target. In general, passive forms

of deception are considered more moral than active forms of

deception (Spranca, Minsk, & Baron 1991).

Compared to lying by commission, we expect paltering to

pose a lesser challenge to one’s moral self-image. Deceivers

who palter can justify their behavior by pointing to the fact

that they made truthful statements. Deceivers who lie by

commission, however, are unable to employ the same justiﬁ-

cation.

Taken together, we expect paltering and lying by commis-

sion to be similarly eﬀective, but because paltering involves

the use of truthful statements and lying by commission in-

volves the use of false statements, we expect lying by com-

ARTFUL PALTERING IN NEGOTIATION ROGERS, ZECKHAUSER, GINO, NORTON, AND SCHWEITZER

mission to be more aversive to would-be deceivers, and we

postulate that individuals will prefer to palter than lie by

commission in a distributive negotiation (Hypothesis 1 ).

A Broken Mental Model: Palterers See their

Behavior as More Ethical than Targets Do

We expect targets of paltering, however, to view paltering

as far less ethical than palterers do. Whereas palterers are

likely to focus on their use of truthful statements to justify

the ethicality of their behavior, we expect targets to focus

on having been actively deceived to conclude that the use of

paltering was unethical.

Deceivers who engage in paltering are likely to engage in

motivated reasoning. Self-interest often guides how individ-

uals perceive the morality of their own behavior (Kronzon

& Darley, 1999; Uhlmann, Pizarro, Tannenbaum, & Ditto,

2009). For example, people may engage in morally question-

able behaviors but rationalize their behavior in a way that

allows them to think of themselves as moral (Bandura, 1991;

Chance, Norton, Gino, & Ariely, 2011; Mazar, Amir, & Ariely,

2008). Deceivers who lie by commission are constrained in

their ability to justify their behavior, because they used state-

ments that were explicitly untrue. Deceivers who palter, how-

ever, can focus on their use of truthful statements and dis-

count the misleading consequences or attribute the misleading

inference to the target (who should have paid closer attention

to exactly what the deceiver was saying). Thus, deceivers who

palter are likely to justify their use of deception more readily

than deceivers who lie by commission. As a result, we expect

deceivers who palter to judge their behavior as more moral

than deceivers who lie by commission (Hypothesis 2).

Targets of deception, however, may be unlikely to reach

the same conclusion. We expect targets of paltering to blame

deceivers for actively inﬂuencing their beliefs and misleading

them. That is, we expect targets of paltering and targets of

lies by commission to similarly apportion blame when the de-

ception is revealed. We hypothesize that compared to telling

the truth, both paltering and lying by commission will cause

substantial reputational damage to would-be deceivers when

targets learn the truth (Hypothesis 3 ).

The Costs and Beneﬁts of Paltering

We expect negotiators who palter to gain at least a short-

term beneﬁt from misleading their counterpart. Negotia-

tions are characterized by information dependence (Kelley

& Thibaut, 1969); to reach agreement, negotiators must ex-

change information. At the same time, negotiators seek to

claim value. By misleading a counterpart, negotiators can in-

ﬂuence a target’s beliefs and increase their personal gains at

the target’s expense (Gaspar & Schweitzer, 2013; Schweitzer

et al., 2005). Thus, deception can enable negotiators to gain

an advantage (Chertkoﬀ & Baird, 1971) and claim a larger

share of total proﬁt (O’Connor & Carnevale, 1997; Schweitzer,

Brodt, & Croson, 2002). We expect paltering to be a partic-

ularly eﬀective form of deception because the use of active,

truthful statements is likely to both distort a target’s beliefs

and be very diﬃcult to detect. Thus, in distributive nego-

tiations we expect paltering to enable negotiators to claim

greater proﬁts than telling the truth (Hypothesis 4 ). In fact,

we expect deceivers who palter to gain an advantage similar

to deceivers who lie by commission. Speciﬁcally, we expect

deceivers who palter and deceivers who lie by commission to

extract similar concessions and claim similar amounts of sur-

plus.

Paltering, however, may also increase the likelihood of im-

passe when a positive bargaining zone exists. By distorting

the information-sharing process, paltering deprives targets of

complete and accurate information. As a result, negotiators

may fail to identify opportunities to reach a deal. In other

research, scholars have found that the absence of accurate in-

formation can increase the likelihood of an impasse (Pinkley,

Griﬃth, & Northcraft, 1995; Thompson, 1991). As a result,

we hypothesize that relative to truth tellers, for negotiations

with a positive bargaining zone

negotiators who palter in-

crease the likelihood of reaching an impasse (Hypothesis 5 ).

In our investigation, we also consider whether or not the

target of deception asked a direct question. In general, targets

of deception may or may not seek relevant information. For

example, a used car buyer may or may not ask about the past

accident history for the vehicle. We term cases in which the

target of deception seeks relevant information (e.g., by asking

a direct question) and the deceiver actively misleads the buyer

as prompted deception. We term cases in which the deceiver

proactively misleads the target as unprompted deception (e.g.,

the deceiver volunteers misleading information).

Given our main interest in paltering, in this paper we com-

pare prompted and unprompted paltering. We postulate that

prompted paltering will be judged to be less ethical than un-

prompted paltering. Prompted deception violates conversa-

tional norms, and actively shifts beliefs about a topic the

target identiﬁed as uncertain and relevant to their decision

process. As a result, we hypothesize that individuals will

judge prompted palters to be less ethical than unprompted

palters (Hypothesis 6).

The Present Research

We tested these hypotheses in two pilot studies and six

experiments. Across our studies, we contrast paltering with

both lying by commission and lying by omission, and demon-

strate that paltering is a distinct form of deception. We begin

by demonstrating that the concept of paltering is both readily

comprehensible and easy for laypeople to distinguish from ly-

ing by omission and lying by commission (Pilot Study 1). We

also investigate the prevalence of paltering among experienced

negotiators. We ﬁnd that experienced negotiators report that

they engage in paltering as often as they lie by omission, and

more often than they lie by commission (Pilot Study 2).

Supporting Hypothesis 1, we ﬁnd that would-be deceivers

prefer to palter than to lie by commission (Study 1). In a face-

to-face negotiation (Study 2), we compare the use of paltering

to the use of lying by commission and test Hypotheses 3, 4 and

5. We ﬁnd that paltering, like lying by commission, enables

negotiators to claim value, but that both paltering and ly-

ing by commission, once revealed, have substantial long-term

reputational consequences. In Study 3, we replicate an aspect

of Study 2 that tests Hypothesis 3 and ﬁnd that those who

palter can suﬀer reputational consequences akin to the conse-

quences of those who lie by commission when the deceptions

are detected. Studies 4a and 4b test Hypothesis 2 and ﬁnd

that palterers perceive their deception to be signiﬁcantly more

In negotiations with a negative bargaining zone, deception may

actually reduce impasses (Kray, Kennedy, & Van Zant, 2014).

ARTFUL PALTERING IN NEGOTIATION ROGERS, ZECKHAUSER, GINO, NORTON, AND SCHWEITZER

moral than their counterparts do. Finally, in Study 5, we con-

trast prompted palters and unprompted palters. Supporting

Hypothesis 6, we ﬁnd that participants judged palters that

were prompted by a question to be signiﬁcantly less ethical

than unprompted palters.

Pilot Study 1: Laypeople Distinguish between

Paltering and Other Forms of Deception

In Pilot Study 1, we examine whether laypeople recognize

that paltering is diﬀerent from two other common forms of

deception: lying by commission and lying by omission. In the

study, we provide participants with deﬁnitions of paltering,

lying by commission, and lying by omission, and then describe

a series of deceptive acts and ask participants to categorize

them.

Method

Participants. We recruited 100 participants (M

age

35.9, SD = 12.63; 48% female) via Amazon’s Mechanical Turk

using an announcement that oﬀered to pay them $0.75 and re-

quired that they be located in the United States. This sample

size was chosen in advance somewhat arbitrarily. We analyzed

these data after all participants had completed the study, and

we report all measures.

Design. We presented participants with the following deﬁ-

nition of each type of deception that can occur in a negotiation

setting:

There are at least three ways negotiators might mislead

their counterparts.

First, negotiators could mislead with active commission.

That is, they could actively say untrue facts to their coun-

terparts. For example, imagine that over the last 10 years

your sales have grown consistently, but that next year you

expect sales to be ﬂat. If asked by your counterpart “How

do you expect sales to be next year?” misleading through

active commission would involve replying with something

like “I expect sales to grow next year.”

Second, negotiators could mislead with active paltering.

That is, they could actively say truthful facts that know-

ingly lead their counterparts to false conclusions. For ex-

ample, imagine that over the last 10 years your sales have

grown consistently, but that next year you expect sales to

be ﬂat. If asked by your counterpart “How do you expect

sales to be next year?” misleading through an active palter

would involve replying with something like “Well, as you

know, over the last 10 years our sales have grown consis-

tently.” This speciﬁcally does not highlight your expecta-

tion that sales this coming year will be ﬂat, but you know

it is likely to create the impression in your counterpart that

sales will grow.

Third, negotiators could mislead with passive omission.

That is, they could passively fail to correct a mistaken be-

lief that they know their counterpart holds. For example,

imagine that over the last 10 years your sales have grown

consistently, but that next year you expect sales to be ﬂat.

If your counterpart makes a statement like “Since sales have

gone up the last 10 years, I expect them to go up next year”

misleading through passive omission would entail you not

actively correcting this false belief.

Next, participants answered three comprehension check

questions presented in random order. The questions were:

“Which of the following involves saying truthful statements

that you expect will lead your counterpart to believe some-

thing that is untrue? Active paltering, passive omission, ac-

tive commission.” “Which of the following involves saying

untruthful statements that you expect will lead your counter-

part to believe something that is untrue? Active paltering,

passive omission, active commission.” And, “Which of the

following involves failing to correct an untrue belief that a

counterpart holds? Active paltering, passive omission, ac-

tive commission.” Participants then read descriptions of nine

diﬀerent deceptive acts (reported in supplemental online ma-

terials). After reading about each deceptive act, participants

were asked to classify each one as either an “active palter”

(palter), “active commission” (lie by commission), or “passive

omission” (lie by omission). For example, one of the scenarios

read:

Imagine that the following two statements are true

about a car that you owned for one year: a. Twice

in the last year this car would not start and both times

you had to have a mechanic ﬁx it; b. This car drives

very smoothly and handles very well. Just last week it

started up with no problems when the temperature was

−5 degrees Fahrenheit.

Next, participants read three diﬀerent negotiation ex-

changes that involved diﬀerent types of deception. We asked

participants to classify the deception as one of the three forms

of deception. For example:

1. Imagine that a potential buyer says, “This car seems

like it works perfectly. I expect it has not had any me-

chanical problems.”

If the potential then buyer asks you, “Has this car ever

had problems?”

If you replied “This car has never had problems” you

would be misleading by: Active paltering, passive omis-

sion, active commission.

2. Imagine that a potential buyer says, “This car seems

like it works perfectly. I expect it has not had any

mechanical problems.”

If the potential buyer then asks you, “Has this car ever

had problems?”

If you replied “This car drives very smoothly and han-

dles very well. Just last week it started up with no prob-

lems when the temperature was −5 degrees Fahren-

heit.” you would be misleading by: Active paltering,

passive omission, active commission.

3. Imagine that a potential buyer says, “This car seems

like it works perfectly. I expect it has not had any

mechanical problems.”

If the potential buyer then moves on to talk about a

diﬀerent topic.

If you did not correct the potential buyer’s mistaken

belief that the car has not had mechanical problems

you would be misleading by: Active paltering, passive

omission, active commission.

Across participants, we randomized the order of the sur-

vey pages, the scenario questions, and the multiple-choice an-

swers.

ARTFUL PALTERING IN NEGOTIATION ROGERS, ZECKHAUSER, GINO, NORTON, AND SCHWEITZER

Results

Eighty-three percent of participants answered all three of

the comprehension questions correctly. Sixty-three percent

of participants answered all nine scenario questions correctly,

and the most common number incorrect was one: on average,

participants answered 7.9 (SD = 1.38) out of nine questions

correctly.

Among participants, 73% correctly answered all three pal-

tering questions, 76% correctly answered all three questions

involving lies by omission, and 75% correctly answered all

three questions involving lies by commission. On average,

participants answered 2.58 (SD = 0.79) out of three palter-

ing questions correctly, 2.67 (SD = 0.67) out of three lie-by-

omission questions correctly, and 2.65 (SD = 0.7) out of three

lie-by-commission questions correctly. Participants’ accuracy

in classifying the type of deception did not diﬀer across the

three types of deception, Hotelling F (2, 98) = 0.99, p = 0.37.

Discussion

This ﬁrst pilot study reveals that laypeople can discrimi-

nate among paltering, lying by omission, and lying by com-

mission. Moreover, participants were as accurate in classify-

ing paltering as they were in classifying the two other forms

of deception.

Pilot Study 2: Experienced Negotiators Palter

Having conﬁrmed that laypeople can distinguish paltering

from lying by omission and lying by commission, we con-

ducted a second pilot study to assess how often experienced

negotiators palter and their evaluations of the ethicality of

this tactic.

Method

Participants. Participants in this study were enrolled

in an executive education course at Harvard Business School

that focused on advanced negotiation strategies. They were

all mid- to senior-level managers from a broad cross-section of

industries. All 184 executives, all of whom negotiate as part of

their normal activities, participated in the study. Because we

wanted to keep the questionnaire as short as possible, we did

not collect demographic information. Participants were not

compensated. We analyzed this data after all participants

had completed the study, and we report all measures.

Procedure. As in Pilot Study 1, participants read deﬁni-

tions of the three forms of deception. The only diﬀerence in

this study is that we eliminated the words active and passive

from the deﬁnitions we provided (but used these qualiﬁers

in the questions we asked afterwards). Participants then an-

swered three comprehension check questions. For each form

of deception (palter, lie by omission, lie by commission), we

then asked participants (a) how often in negotiation they mis-

lead their counterpart using this form of deception (response

options: most of their negotiations, some of their negotia-

tions, or few of their negotiations); (b) when they do mislead

their counterparts with each form of deception, whether they

think they are being honest or dishonest; and (c) when they

do mislead their counterparts using each form of deception,

do they do so because they think this will allow them to get

a better deal, or not.

Results

All participants passed the three comprehension check

questions, suggesting that they understood the diﬀerence be-

tween palters, lies by omission, and lies by commission.

Many of these experienced negotiators (52%) reported that

they paltered in some or most of their negotiations. About

the same proportion reported that they paltered in some or

most of their negotiations as reported that they lie by omis-

sion in some or most of their negotiations (57%) (McNemar

test of signiﬁcance: χ

= 2.08, p = 0.150)

. A much smaller

percentage reported that they lie by commission than pal-

tered in some or most of their negotiations (21%), (McNemar

test of signiﬁcance: χ

= 52.56, p < 0.001)

More of these negotiators indicated that they perceived

their palters to be honest (32%) than indicated that they

perceived their lies by omission to be honest (23%; McNe-

mar test of signiﬁcance: χ

= 4.92, p = 0.026). Similarly, a

greater proportion indicated that they perceived their palters

to be honest than indicated that they perceived their lies by

commission to be honest (5%; McNemar test of signiﬁcance:

= 48.08, p < 0.001). A greater proportion indicated that

they perceived their lies by omission to be honest than in-

dicated that they perceived their lies by commission to be

honest, McNemar test of signiﬁcance: χ

= 27.52, p < 0.001.

When they palter, most of these negotiators (88% of them)

do so in an attempt to get a better deal. This did not dif-

fer from the proportion of these negotiators reporting that

getting a better deal was their motivation when they lied by

commission (84%; McNemar test of signiﬁcance: χ

=2.00,

p=0.16 ). Similarly, this proportion did not diﬀer from the

proportion of these negotiators reporting that getting a better

deal was their motivation when they lied by omission (91%;

McNemar test of signiﬁcance: χ

= 1.47, p = 0.23).

Discussion

These results show that experienced negotiators commonly

use palters, that they believe palters to be strategically ad-

vantageous, and that they think palters are more ethically

acceptable than both lies by omission and lies by commission.

Study 1: Paltering Is Preferred Over

(More Lucrative) Lying by Commission

In Pilot Studies 1 and 2, laypeople and experienced nego-

tiators distinguish paltering from both lying by omission and

lying by commission. Having demonstrated that paltering is

a distinct deception tactic and that it is often used in nego-

tiation, we now examine whether people prefer it over lying

by commission—as suggested by the ﬁnding in Pilot Study

2 regarding the frequency with which paltering is used. We

focus on lying by commission as a comparison since it is the

deception tactic known in the literature as leading to the rela-

tively best negotiation outcomes. Moreover, the opportunity

to palter or lie by commission is always present. In contrast,

the opportunity to lie by omission is limited. Deceivers can

only lie by omission when the target of deception holds and

communicates a relevant false belief and/or fails to ask a rel-

evant question. That is, compared to lies by commission and

We use the McNemar test of signiﬁcance because it is the most

appropriate version of a chi-square test for within-subjects data.

ARTFUL PALTERING IN NEGOTIATION ROGERS, ZECKHAUSER, GINO, NORTON, AND SCHWEITZER

paltering, deceivers have fewer opportunities to engage in lies

by omission, especially when they interact with experienced

counterparts.

In Study 1, we investigate whether potential deceivers ﬁnd

paltering to be more ethically acceptable than lying by com-

mission, as our ﬁndings in Pilot Study 2 suggest. To as-

sess this relative preference, we explore whether participants

would be willing to forgo proﬁt to palter rather than lie by

commission. That is, we investigate whether or not potential

deceivers are willing to pay to avoid lying by commission.

Discussion

Participants. We recruited 550 participants (M

age

35.15, SD = 11.08; female=53%) via Amazon’s Mechanical

Turk. The announcement oﬀered to pay participants $0.36,

and required that they be located in the United States. This

sample size was chosen to have suﬃcient statistical power as

determined based on pilot testing. We analyzed this data af-

ter all participants had completed the study, and we report

all measures and manipulations.

Design. Participants are told to imagine that they are try-

ing to sell a used car on eBay.com and that if they succeed in

selling the car they will receive $1.00 in bonus compensation.

Participants are told that they are motivated to make the sale

and that the following two statements are true: “Twice in the

last year this car would not start and both times you had to

have a mechanic ﬁx it” and “This car drives very smoothly

and is very responsive. Just last week it started up with no

problems when the temperature was −5 degrees Fahrenheit.”

Participants are told that a potential buyer emails and asks,

“Has this car ever had problems?” We randomly assigned

participants to one of two conditions. In each condition, par-

ticipants choose to send one of two email messages: an honest

message or a misleading message. In the Lie by Commission

condition, participants choose between the truth (“Twice in

the last year this car would not start and both times I had to

have a mechanic ﬁx it”) and a lie by commission (“This car

has never had problems”). In the Palter condition, partici-

pants choose between the truth (“Twice in the last year this

car would not start and both times I had to have a mechanic

ﬁx it”) and a palter (“This car drives very smoothly and is

very responsive. Just last week it started up with no problems

when the temperature was −5 degrees Fahrenheit.”).

Before the participants make a choice, they are told that

their answer will aﬀect their probability of success. Partic-

ipants in both conditions have a 30% chance of selling the

car and earning the $1.00 bonus payment if they choose to

tell the truth. In the Lie by Commission condition, partici-

pants have an 80% chance of selling the car and receiving the

$1.00 bonus compensation if they choose to mislead (i.e., lie

by commission). In the Palter condition, participants have

only a 60% chance of selling the car and receiving the $1.00

bonus compensation if they choose to mislead (i.e., palter).

A random number generator determines whether or not par-

ticipants “make the sale” and earn the bonus payment based

on the response participants’ decide to send. That is, partic-

ipants in the Lie by Commission condition reap an expected

value of $0.80 if they choose to lie by commission. Partici-

pants in the Palter condition reap an expected value of $0.60

if they choose to palter. The expected value of choosing to

tell the truth in both conditions is $0.30.

Participants then indicate how honest (1=Not at all honest,

7=Very honest), ethical (1=Not at all ethical, 7=Very ethi-

cal), and deceptive (1=Not at all deceptive, 7=Very decep-

tive) the misleading option they were oﬀered is. (In this study,

our ethicality index included “ethical” rather than “trustwor-

thy,” because it is a better ﬁt in our experimental context.)

Results

We ﬁnd that participants in the Palter condition choose

to mislead more often (71%) than did those in the Lie by

Commission condition (55%), χ

(1) = 14.58, p < 0.001. Mea-

sures of honesty, ethicality, and deceptiveness (reverse-coded)

were highly correlated in both the palter (α = .79, p < 0.001)

and the lie by commission (α = .88, p < 0.001) conditions

so we take the average of these measures to form an eth-

icality index. This index reﬂects participants’ perceptions

of the ethicality of the single option they were oﬀered for

misleading the prospective buyer. Participants in the Pal-

ter condition rated the ethicality of the palter option, while

participants in the Lie by Commission condition rated the

ethicality of the Lie by Commission option. Participants in

the Palter condition reported that the deceptive option they

were oﬀered to mislead the potential buyer was of higher

ethicality (M = 4.05, SD = 1.36) than did participants in

the Lie by Commission condition (M = 2.12, SD = 1.42),

= −1.93, p < 0.001, Cohen’s d=1.39.

When we add the ethicality index as an independent vari-

able in a logit model in which condition assignment is the

other independent variable and choosing the misleading op-

tion is the dependent variable, the relationship between condi-

tion assignment and choosing the misleading option became

insigniﬁcant (β

= 0.16, SE = 0.269, p = 0.410). At the

same time, the indirect eﬀect of ethicality remained signiﬁ-

cant (β

= 0.455, SE = 0.077, p < 0.001). To test the sig-

niﬁcance of the indirect eﬀect of ethicality on choosing the

misleading option, we used the Sobel-Goodman bootstrap-

ping method (Preacher, K. J. & Hayes, A. F. 2004). The

resulting 95% conﬁdence interval for the bias-corrected indi-

rect eﬀect did not contain zero (lower bound=−0.249, upper

bound=−0.111; conﬁdence interval without bias correction:

lower bound=−0.244, upper bound=−0.105). This is consis-

tent with an interpretation that perceived ethicality of the

option to mislead the potential buyer mediates the eﬀect of

condition assignment on choosing the option to mislead.

Discussion

Consistent with Hypothesis 1, participants preferred palter-

ing to lying by commission. When choosing between telling

the truth and lying by commission, participants were more

willing to palter than they were to lie by commission—even

though the expected value of paltering was substantially lower

than the expected value of lying by commission. This result is

particularly noteworthy given that this experiment was con-

ducted in an anonymous environment (Amazon’s Mechanical

Turk). Many factors that deter lying by commission in natu-

ral settings were not present in this experiment: participants

did not know the identity of their bargaining partners, they

were not in the physical presence of their bargaining part-

ners, and they had no chance to encounter their bargaining

partners again. Moreover, participants were faced with only

hypothetical choices. For these reasons, our ﬁndings oﬀer a

conservative test of the extent to which people ﬁnd lying by

commission less palatable than paltering. Importantly, we

also found that participants preferred paltering over lying by

ARTFUL PALTERING IN NEGOTIATION ROGERS, ZECKHAUSER, GINO, NORTON, AND SCHWEITZER

commission because they feel less ethical lying by commission

than paltering.

Study 2: Face-to-Face Negotiation

In Study 2, we extend our investigation by exploring the

potential costs and beneﬁts of paltering and lying by commis-

sion in a face-to-face negotiation. Both tactics involve active

attempts to mislead. We assess whether paltering, like lying

by commission, oﬀers short-term beneﬁts (gaining better out-

comes) but incurs longer-term costs if targets become aware

that they have been deceived (reputational harm). These

results would point to an ironic pattern: though negotia-

tors seem to prefer paltering over lying by commission, when

the deceptions are discovered negotiators experience the same

reputational cost from paltering as they do from lying by com-

mission.

Method

Participants. We recruited 80 students (M

age

22.11, SD = 3.49; 60% female) from a university in the north-

eastern United States to participate in a study in exchange for

a $7 payment and the opportunity to earn additional money

during the experiment. Participants reported an average of

2.75 years of work experience. We used an intensive schedul-

ing process that matched each participant with a confederate.

Each dyad negotiated face-to-face in a separate room. This

procedure enabled us to measure reactions to face-to-face de-

ception in a controlled environment, but the intensive nature

of the study limited our recruitment to 80 participants. We

recruited as many participants as we could schedule during

a summer session. We analyzed this data after all partici-

pants had completed the study, and we report all measures

and manipulations.

Design. After participants arrived at the laboratory, they

were informed that they would take part in a negotiation

study. They were told that they would be randomly assigned

to the role of either buyer or seller. In reality, we assigned ev-

ery participant to the role of seller, and we paired each seller

with a confederate buyer. We assigned participants to one

of three between-subjects conditions: Truth, Lie by Commis-

sion, and Palter. The confederate-buyers’ responses to par-

ticipants’ questions during the negotiation diﬀered depending

on condition.

Procedure. Participants and confederates arrived at the

experiment at the same time, and everyone checked in with

the experimenter. The experimenter then assigned each per-

son to a role. Though the process was designed to appear ran-

dom, the experimenter assigned every participant to the role

of seller and every confederate to the role of buyer. The ex-

perimenter paired each participant (seller) with a confederate

buyer and assigned each dyad to a group study room, where

both the confederate-buyer and the participant-seller received

packets of materials. The confederate-buyer’s packets con-

tained background materials, comprehension questions, a

Deal Sheet, a bonus payment sheet, a sealed folder containing

the answers to the comprehension check, and a sealed folder

containing a post-negotiation survey. The participant-sellers’

packets contained background materials, comprehension ques-

tions, a Deal Sheet, a bonus payment sheet, a sealed folder

containing the answers to the comprehension check, a sealed

folder containing an Immediate Post-Negotiation survey, and

a sealed folder containing a Full Disclosure Post-Negotiation

survey (see supplemental materials for all materials).

A common instruction sheet indicated that since the buyer

had one fewer sheet to complete, the buyer would be in

charge of managing time. This aspect of our design gave

the confederate-buyer some control over the negotiation pro-

cess. We asked both parties to read through their background

information. Both the participant-seller and the confederate-

buyer then answered a series of multiple-choice comprehension

check questions to ensure that they understood the material.

They used the answer sheet to self-check their comprehension

check answers. Once they had veriﬁed their answers, they

proceeded to the negotiation.

In the negotiation, we used a modiﬁed version of the Hamil-

ton Real Estate negotiation (Malhotra, 2010). We asked

participant-sellers to imagine themselves as the owner of a

property they were looking to sell. They were told that the

value of the land varied greatly depending upon what the

buyer planned to do with it. If the land were used for a res-

idential development, it would be worth $36-44 million. If it

were used for luxury condominiums, it would be worth $44-52

million. If it were used for commercial development, it would

be worth more than $60 million. We informed participant-

sellers that they would earn a bonus payment according to

the following schedule:

You will report your oﬀer: the amount you are willing

to accept to sell the property (between $38 million and

$60 million).

The BUYER will report her/his oﬀer: the amount

s/he is willing to pay (between $38 million and $60

million).

If the amount of your oﬀer is LESS than the amount

of the BUYER’s oﬀer, you will reach a deal.

The price will be the average of your two prices.

If the amount of your oﬀer is MORE than the amount

of the BUYER’s oﬀer, you will NOT reach a deal.

You will receive a cash bonus for a deal. The higher

the deal price, the bigger the bonus. You will earn

$0.50 in bonus cash for every $1 million dollars over

$38 million that the ﬁnal deal price is. For example,

if the deal price is $40 million dollars, you will earn

bonus cash of $1. If the deal price is $60 million

dollars, you will earn bonus cash of $11.

If you do not reach a deal at the end of the negotiation,

you will not earn any bonus cash.

We prompted the participant-seller to ask the confederate-

buyer two questions: (1) “Are you negotiating with other

parties as well?” (2) “Are you going to use the property

for commercial development?” The participant-seller knew

that the confederate-buyer ran a residential development ﬁrm.

They did not know that the confederate-buyer intended to

develop the land for commercial purposes.

The confederate-buyer’s response to this question fell un-

der one of three predetermined conditions. In the Truth con-

dition, confederate-buyer told participant-sellers that they

planned to develop the land for commercial use. In the

ARTFUL PALTERING IN NEGOTIATION ROGERS, ZECKHAUSER, GINO, NORTON, AND SCHWEITZER

Lie by Commission condition, the confederate-buyer told the

participant-seller that they planned to develop the land for

residential use. In the Palter condition, the confederate-buyer

emphasized their past experience in the residential market

and avoided stating a speciﬁc development plan. That is,

they used true statements to create a false impression. The

confederate-buyer received speciﬁc instructions and training

for how to deliver each type of answer (from one of the au-

thors). For example, in the Palter condition, their ﬁrst re-

sponse was “We have only ever developed residential property

before.” (See supplemental materials for more details on how

confederate-buyer was coached.)

After the negotiation ended, confederate-buyer and

participant-seller ﬁlled out separate Deal Sheets and turned

them face down so that the other party could not see what

they had written. These Deal Sheets determined the to-

tal payoﬀ for each participant, according to the incentive

scheme described above. The confederate-buyer was always

instructed to write $60 million as the highest price they would

pay for the property (their willingness-to-pay). We instructed

the participant-seller to write the lowest price they would be

willing to accept for the property.

After the participant and confederate completed the Deal

Sheets, the participant-seller completed the Immediate Post-

Negotiation survey to assess their perceptions of their

confederate-buyer. In this survey, they reported how much

they trusted their counterpart (their confederate-buyer) on 7-

point Likert scales (1: Not at all, 7: Completely), what they

thought their confederate-buyer intended to do with the prop-

erty (commercial, luxury condo, residential, no idea), how

certain they were of their answer regarding their confederate-

buyer’s intentions (7- point Likert scale ranging from Not at

all to Completely sure), if they would negotiate with their

confederate-buyer again in the future (7-point Likert scale

ranging from Deﬁnitely would not to Deﬁnitely would), and

some additional demographic questions.

After the participant-seller completed the Immediate Post-

Negotiation survey, they received more information about

the negotiation. Speciﬁcally, they opened an envelope that

revealed their (confederate) buyer’s true intentions: to use

the property for commercial use. That is, each participant-

seller then learned that their confederate-buyer either had

been honest or had misled them either by paltering or lying

by commission. The participant-seller then completed a Full

Disclosure Post-Negotiation survey.

In this survey, we again asked the sellers how much they

trust their counterpart (the confederate-buyer) on a 7-point

Likert scale (1: Not at all, 7: Completely), to rate their

confederate-buyer’s honesty (1: Very Dishonest, 7: Very

Honest), to report an open-ended overall impression of their

confederate-buyer, and if they would negotiate with their

confederate- buyer again in the future (7-point Likert scale

ranging from Deﬁnitely would not to Deﬁnitely would). Si-

multaneously, the confederate-buyer ﬁlled out their own post-

negotiation survey, which asked them to report if their

participant-seller asked them about their intentions for de-

veloping the property, how they responded to the question

(truth, palter, lie by commission), if they discussed speciﬁc

oﬀer values with their participant-seller during the negotia-

tion, and any additional comments they might have.

After the participant-seller submitted responses, both they

and the confederate-buyer ﬁlled out bonus payment sheets

and found out what their counterpart had written on the

Deal Sheet. After all of this paperwork was completed, the

confederate-buyer and the participant-seller gathered up all

of the materials, returned the materials to the lab manager,

were paid, and received debrieﬁng information.

Results

We ﬁrst report analyses of ﬁnal prices. Then, we report how

much participants trusted their counterpart immediately after

negotiating and again after they learned about their counter-

part’s true intentions.

Seller willingness-to-accept. Three participant-sellers

in the Truth condition reported willingness-to-accept prices

that were higher than $60 million, the highest value in the

zone of agreement. We include these responses in our anal-

yses by recoding their willingness to accept as $60 million.

Excluding them does not change the primary results; see sup-

plemental materials. To control for the possibility that there

were main eﬀects by the research assistants who served as the

confederate-buyer, in all of our analyses we control for the

research assistant.

The three conditions resulted in signiﬁcantly diﬀerent

willingness-to-accept prices, F (2, 80) = 25.59, p < 0.001.

First, participant-sellers in the Lie by Commission condi-

tion (M=$45.23 million, SE=0.92) accepted a lower price

than those assigned to the Truth condition (M =$53.64 mil-

lion, SE=0.886), β = −8.41, SE = 1.28, p < 0.001, Cohen’s

d=1.57. Second, participant-sellers in the Palter condition

(M =$46.46 million, SE = .87) were willing to accept a lower

price than those in the Truth condition (M =$53.64 million,

SE = 0.89), β = −7.18, SE = 1.25, p < 0.001, Cohen’s

d=1.41. Third, participant-sellers in the Lie by Commission

condition were willing to accept amounts that were not signif-

icantly diﬀerent from the price sellers in the Palter condition

were willing to accept, β = −1.23, SE = 1.27, p = 0.33. That

is, participants who were deceived, whether by lying by com-

mission or by paltering, accepted lower prices. These results

support Hypothesis 3, which predicted that in distributive ne-

gotiations, paltering, as compared to telling the truth, would

enable negotiators to claim greater proﬁts. In fact, we hy-

pothesized, paltering would increase value-claiming as much

as would lying by commission.

Immediate post-negotiation seller perceptions. The

three conditions resulted in signiﬁcantly diﬀerent post-

negotiation seller perceptions, F (2, 80) = 106.31, p < 0.001.

Participant- sellers in the Truth condition were more likely

to believe that their confederate-buyers were going to use the

property for commercial purposes (M =92%, SE = 0.048)

than were those in the Palter condition (M =8%, SE =

0.047), β = 0.84, SE = 0.067, p < 0.001. Participant-sellers in

the Truth condition were also more likely to believe that their

confederate-buyers were going to use the property for com-

mercial purposes than those in the Lie by Commission condi-

tion (M = 3.0%, SE = 0.05), β = 0.89, SE = 0.07, p < 0.001.

Those in the Palter and Lie by Commission conditions did

not diﬀer in their beliefs regarding how likely their buyer was

to develop the property for commercial use, β = 0.05, SE =

0.07, p = 0.45. That is, lying by commission and paltering

similarly distorted participants’ beliefs.

Immediately after the negotiation concluded, and before

participants learned the truth, participants across the three

conditions were similarly interested in negotiating with their

counterpart in the future. Participant-sellers in the Truth

condition were equally interested in negotiating again with

their confederate-buyers (M =5.78, SE = 0.195), as were

ARTFUL PALTERING IN NEGOTIATION ROGERS, ZECKHAUSER, GINO, NORTON, AND SCHWEITZER

Table 2

Mean (and Standard Deviations) of the Main Variables Measured by Condition, Study 3

Condition Unethical Behavior Unethical buyer Trust in buyer Likely to negotiate again

Palter condition 4.30 (1.62) 4.49 (1.62) 2.54 (1.61) 3.44 (1.68)

Honest condition 3.24 3.58 (1.82) 3.54 (1.76) 4.22 (1.60)

Lie by Commission condition 4.61 (1.66) 4.45 (1.72) 2.18 (1.49) 3.45 (1.84)

those in the Palter condition (M =5.48, SE = 0.191), β =

0.30, SE = 0.27, p = 0.28. Participant-sellers in the Truth

condition were equally interested in negotiating again with

their confederate-buyers as were those in the Lie by Com-

mission condition (M =5.38, SE = 0.202), β = 0.40, SE =

0.28, p = 0.16. Those in the Palter and Lie by Commission

conditions also did not signiﬁcantly diﬀer, β = 0.10, SE =

0.280, p = 0.72. Analyzed using ANOVA, the three condi-

tions do not diﬀer signiﬁcantly from each other, F (2, 80) =

1.10, p = 0.34.

Before deception was revealed, participant-sellers did not

diﬀer across conditions in how much they trusted their

confederates-buyers. Participant-sellers in the Truth con-

dition did not trust their confederate-buyers (M =4.67,

SE=0.23) more than did those in the Palter condition

(M =5.05 , SE = 0.22), β = −0.385, SE = 0.32, p = 0.23.

Participant-sellers in the Truth condition did not trust their

confederate-buyers more than did those in the Lie by Com-

mission condition (M =4.82, SE = 0.23), β = −0.16, SE =

0.33, p = 0.66. Neither did those in the Palter and Lie by

Commission conditions diﬀer signiﬁcantly, β = 0.23, SE =

0.32, p = 0.48. Analyzed using ANOVA, the three condi-

tions do not diﬀer signiﬁcantly from each other, F (2, 80) =

0.74, p = 0.48.

Full disclosure post-negotiation seller perceptions.

The three conditions diﬀer signiﬁcantly in the full disclo-

sure post-negotiation seller perceptions, F (2, 79) = 43.5, p <

0.001.

After the confederate-buyers’ true intentions (to

redevelop the property for commercial use) were fully dis-

closed, participant-sellers in the Truth condition were less

likely to believe that their confederate-buyers were dishonest

(M =5.99, SE = 0.29) than were participants in both the Pal-

ter condition (M =2.98, SE = 0.29), β = 3.01, SE = 0.42, p <

0.001, Cohen’s d=1.57, and the Lie by Commission condition

(M =2.32, SE = 0.30), β = 3.67, SE = 0.42, p < 0.001. Those

in the Palter and Lie by Commission conditions did not diﬀer

signiﬁcantly, β = 0.657, SE = 0.42, p = 0.12. These results

support Hypothesis 3, which predicted that, after deception is

revealed, both paltering and lying by commission would incur

substantial reputational damage.

To assess how disclosing the truth aﬀected participant-

sellers’ interest in negotiating again with the confederate-

buyers, we conducted a repeated measures within-between

ANOVA where the question about the participant-sellers’

interest in negotiating again in the future with their

confederate-buyers is asked twice as the within-participant

factor. We asked participants about their interest in negoti-

ating with their counterpart before we disclosed the truth and

again after we disclosed the truth. The between-participant

One participant did not answer the honesty question, thus reducing

the sample size to 79.

factor is condition assignment. This analysis shows that af-

ter the truth had been disclosed, participant-sellers were less

likely to want to negotiate again in the future with their coun-

terpart in both the Lie by Commission condition (M =−2.12,

SE = 0.25) and in the Palter condition (M =−2.11 SE =

0.24). Participant-sellers’ interest in negotiating again in

the future with their confederate-buyers in the Truth condi-

tion did not change (M =−0.26, SE = 0.14); within-subjects

factor F (3, 79) = 31.4, p < 0.001, between-subjects factor

F (2, 79) = 7.89, p = 0.001.

Similarly, in order to assess how disclosing the truth af-

fected participant-sellers’ trust in their confederate-buyers,

we conducted a repeated measures within-between ANOVA.

The trust question is asked twice as the within-participant

factor—once before we disclosed the truth and once after—

and the between-participant factor is condition assignment.

This analysis shows that participant-sellers’ trust in their

confederate-buyers signiﬁcantly decreased after the full truth

was disclosed in the Lie by Commission condition (M =−.2.24,

SE = 0.20) and in the Palter condition (M =−1.85, SE =

0.27). Trust did not change in the Truth condition (M =.56,

SE = 0.13); within-subjects factor F (3, 79) = 31.52, p <

0.001, between-subjects factor F (2, 79) = 13.55, p < 0.001.

Discussion

Consistent with Hypotheses 3, 4 and 5, Study 2 shows that

paltering is a risky strategy for negotiators if the truth is likely

to become known and where reputation is important. Like

lying by commission, paltering enabled individuals to claim

value in the short-term. However, when the full truth was dis-

closed, paltering incurred signiﬁcant reputation damage. In

our study, as compared to telling the truth, the beneﬁts (in-

creased proﬁt) and the costs (reputation damage) of paltering

were similar to those of lying by commission.

An important feature of this study is that participant-

sellers negotiated with a confederate who was trained to lie

by commission, palter, or disclose the complete truth. While

this constrained the study’s realism, it ensured that all nego-

tiations reﬂect the intended form of deception or disclosure.

Study 3: Counterparts Evaluate Paltering As

Severely As Lying By Commission

Study 3 leveraged a diﬀerent paradigm to replicate the ﬁnd-

ings in Study 2 regarding the reputational consequences of

paltering compared to lying by commission. Whereas Study

2 used confederates in a real face-to-face negotiation, Study 3

uses an online simulation. While it lacks to external validity

of a real negotiation, Study 3 ensures that confederates do not

communicate experimenter demands across condition. Study

3 examines how counterparts perceive paltering—when it is

disclosed to them—and how a reputation for paltering im-

pacts opportunities for future negotiations.

ARTFUL PALTERING IN NEGOTIATION ROGERS, ZECKHAUSER, GINO, NORTON, AND SCHWEITZER

Methods

Participants. We recruited 160 participants (M

age

30, SD = 8.96; 39% female) via Amazon’s Mechanical Turk.

We restricted our sample to participants in the United States,

and we oﬀered participants $0.50. Five participants failed an

attention check that occurred before we assigned them to a

condition, leaving a ﬁnal sample of 155 participants. The sam-

ple size of 160 was chosen to have suﬃcient statistical power

as determined based on pilot testing. We analyzed this data

after all participants had completed the study, and we report

all manipulations and measures.

Procedure. After passing an attention check, we pre-

sented participants with an adapted version of the materials

we used Study 2. Instead of asking participants to negotiate

as a buyer or Seller, we asked participants to imagine be-

ing responsible for the sale of a large piece of property. We

explained that the land would be worth 1.5 to 2 times as

much if it were developed for commercial use. Participants

also read that they would soon meet with a potential buyer

to negotiate a deal for the sale of the property, and that the

buyer represents a company that invests primarily in residen-

tial properties.

We then asked participants to imagine that during their

negotiation they asked the following question: “Do you plan

to develop the property for residential use?” Following this

question, they received one of the following three replies: In

the Palter condition, the buyer replied, “I have only ever de-

veloped properties for residential use before.” In the Honest

condition, the buyer replied, “No, I intend to develop the

property for commercial use.” In the Lie by Commission

condition, the buyer replied, “Yes, I intend to develop the

property for residential use.”

We then informed participants of the following, “You later

discovered that, at the time of negotiations, the buyer knew

that the property would soon be zoned for commercial devel-

opment, thus making the value of the property higher.”

Now, with the palter or lie by commission exposed, we

asked participants to indicate the extent to which they

thought the answer the buyer gave was unethical, dishon-

est, and immoral (α = 0.91) using a seven-point scale (from

1=Not at all, to 7=Extremely). Participants answered an-

other three questions rating how unethical, dishonest, and

immoral (α=0.92) they thought the buyer was using the same

scale.

Next, we asked participants to imagine that they had the

chance to engage in another negotiation with the same buyer.

Participants then indicated how likely they would be to nego-

tiate with the same buyer versus search for a diﬀerent partner

(1=Not likely at all, 7=Very likely). Finally, we asked partic-

ipants to imagine that they had to negotiate with the same

buyer and to indicate the extent to which they would trust the

buyer (1=Not at all, 7=Very much). Finally, we asked par-

ticipants demographic questions about their age and gender.

We asked participants several other demographic questions

that were never analyzed.

Results

In Table 2, we report the descriptive statistics of the main

variables measured in the study by condition. In Table 3, we

report the correlations among each pair of variables.

Unethical answer and unethical buyer. The three

questions on ethics were strongly correlated; hence, we aver-

Table 3

Correlations among the main variables measured, Study 3.

Variable M SD 2 3 4

1. Unethical answer 4.03 1.81 .82

∗∗

−.51

∗∗

−.56

∗∗

2. Unethical buyer 4.16 1.71 −.61

∗∗

−.59

∗∗

3. Trust in buyer 2.77 1.72 .76

∗∗

4. Likely to negotiate again 3.72 1.74

∗∗p < .001.

aged these items to create an ethicality index, see Table 3. As

expected, participants’ ratings of how unethical the buyer’s

behavior was varied by condition, F (2, 152) = 9.23, p < .001.

Participants rated the behavior of the buyer who paltered as

more unethical than the behavior of the buyer who answered

honestly (B = 1.06, SE = .36, p = .002), but no diﬀerent

from the buyer who lied by commission (B = −.31, SE =

.34, p = .36). Participants also rated the buyer’s response to

be less ethical in the Lie by Commission condition than in the

Honest (B = 1.37, SE = .34, p < .001) condition.

Similarly, participants rating of the ethicality of the buyer

varied by condition, F (2,144)=4.73, p=.01. Participants

rated the buyer who paltered or the one who lied by commis-

sion to be more unethical than the buyer who answered hon-

estly, (B=.91, SE=.34, p=.008 and B=.87, SE=.33, p=.01,

respectively). The buyer who paltered and the buyer who lied

by commission were regarded as equally unethical (B=.04,

SE=.34, p=.91 ).

Trust in the buyer and future negotiations. Sim-

ilarly, participants’ trust in the buyer varied by condition,

F (2, 152) = 9.92, p < .001. Participants had more trust

in the honest buyer than the one who paltered or the one

who lied by commission (B = −1.0, SE = .32, p = .002 and

B = −1.36, SE = .32, p < .001, respectively). The latter

two were regarded as equally untrustworthy (B = .36, SE =

.32, p = .26).

This lack of trust in those who palter and those who lie

by commission played out in Sellers’ likelihoods of voluntarily

negotiating again with the same buyer in the future. Those

likelihoods varied by condition, F (2, 152) = 3.65, p = .028.

Participants reported that they were more likely to negotiate

again with the buyer who answered honestly than the buyer

who paltered (B = −.78, SE = .33, p = .02), or the buyer who

lied by commission (B = −.77, SE = .33, p = .02). The latter

two had the same likelihood of securing another negotiation

(B = −.01, SE = .33, p = .97).

Discussion

Palters may never be discovered by counterparts. But when

palters are discovered, Study 3 (like Study 2) shows that they

can negatively aﬀect negotiators’ reputations, which in turn

impacts negotiators’ prospects for future negotiations. This

is consistent with the ﬁndings of Study 2. When individuals

discover that a prospective negotiation partner had paltered

to them in the past, they are less likely to trust that partner.

And since they are less likely to trust that person, they are

less likely to negotiate with that person again. Importantly,

palterers did no better in this respect than negotiators who

lied by commission. The counterparts of these individuals

assigned them equivalently poor ethical standing.

ARTFUL PALTERING IN NEGOTIATION ROGERS, ZECKHAUSER, GINO, NORTON, AND SCHWEITZER

Studies 4A and 4B: Palterers think Palters are

More Ethical than Do Targets

In Pilot Studies 1 and 2 and in Studies 1-3, we identify

paltering as a distinct form of deception that is frequently

used by experienced negotiators. We ﬁnd that people prefer

to palter than lie by commission; even though paltering and

lying by commission are similarly eﬀective, paltering is easier

to self-justify than lying by commission. Targets, however,

view paltering as unethical as they do lying by commission.

In Studies 4a and 4b, we extend our investigation of the di-

vergence between how potential deceivers and targets view

paltering.

Study 4A

In Study 4A, we investigate how palterers and targets eval-

uate the exact same behavior (thus directly testing Hypothe-

sis 2). We predict that palterers perceive their behavior to be

moral even though targets perceive revealed paltering to be

dishonest and immoral. Whereas a palterer is likely to focus

on the veracity of their statements, a target is likely to focus

on the mistaken impression that the statements successfully

conveyed.

Discussion

Participants. We recruited 258 participants (M

age

31.70, SD = 9.35; female=37%) via Amazon’s Mechanical

Turk using an announcement that oﬀered to pay participants

$0.36 and required that they be located in the United States.

Workers on Amazon’s Mechanical Turk were excluded from

participating in the study if they participated in a related

previous study. Based on pilot testing we aimed to recruit

260 participants so as to have suﬃcient statistical power. We

analyzed this data after all participants had completed the

study, and we report all measures and manipulations.

Design. Participants were randomly assigned to be a

buyer or a seller in a hypothetical car negotiation. Partic-

ipants in the Palterer condition (N=130) were told that they

listed their car on eBay.com, that their car has two small

dents, that it drives smoothly on the highway, and that it

needed a mechanic twice in the last four months because the

car would not start (See supplemental materials for full study

materials). These participants were then told that a potential

buyer emailed them and asked if the car had any dents or en-

gine or performance problems. Participants were then asked,

“If you were unwilling to tell the potential buyer something

that you knew to be untrue—but you wanted to sell the car

for the most money possible—which of the following would

you reply to the potential buyer?” (1=The car has two small

dents (the size of a dime) on the back bumper. The car drives

very smoothly on the highway and is very responsive; just last

week it started up with no problems when the temperature

was freezing. 2=The car has two small dents (the size of a

dime) on the back bumper. Twice in the last 4 months the

car would not start and both times I had to have a mechanic

ﬁx it.) The ﬁrst response option involves a palter because it

aims to lead the buyer to believe that there are no mechanical

problems. The second response is truthful because it trans-

parently discloses that there are mechanical problems. 78%

of participants (N=102) indicated that the paltering response

would yield them the greater monetary amount. The 22%

of participants (N=28) who chose the more truthful response

were asked the same question again. Nearly 80 percent of

these participants (N=22) switched their answer and chose

to palter in the negotiation after being prompted to read the

question again. The remaining six participants who indicated

that the truthful response would be in their best monetary

interest were then asked to imagine the scenario in which

they had responded with the palter. All participants in the

Palterer condition were told that the potential buyer replied

to the participant’s palter by stating that he/she will buy

the car at the listed price and that the buyer left the nego-

tiation with no knowledge of the car’s mechanical problems.

Participants then indicated how honest (1=Not at all hon-

est, 7=Very honest), trustworthy (1=Not at all trustworthy,

7=Very trustworthy), and deceptive (1=Not at all deceptive,

7=Very deceptive) they, the seller of the car, were.

Participants in the Target condition (N=128) were told

they are interested in a car listed on ebay.com and that the

car’s dents and engine performance are important to them.

Participants in the Target condition were then asked which of

the following emails to the seller would be more useful, given

their interests (1=I am interested in your used car. How many

dents does the car have? And, does the car have any engine or

performance problems? 2=I am interested in your used car;

How many dents does the car have?). Only one participant

failed this comprehension test, choosing not to ask about the

car’s engine performance. This participant corrected his/her

answer when asked a second time. All participants in the

Target condition were then told that they had made a deal

with the buyer when the buyer responded to their email by

stating: “The car has two small dents (the size of a dime) on

the back bumper. The car drives very smoothly on the high-

way and is very responsive. Just last week it started up with

no problems when the temperature was freezing.” This is the

paltering response that 95% of participants in the Palterer

condition thought would yield the most amount of money for

the car. Participants in the Target condition were told that,

after they made the transaction, they became aware that the

seller knowingly withheld the relevant facts about the car’s

engine performance while actively creating the opposite im-

pression. Similar to participants in the Palterer condition,

participants in the Target condition were asked how honest

(1=Not at all honest, 7=Very honest), trustworthy (1=Not at

all trustworthy, 7=Very trustworthy), and deceptive (1=Not

at all deceptive, 7=Very deceptive) the seller was. An analy-

sis plan was pre-registered before data analysis began. It can

be found at https://osf.io/8k4t5/.

Results

We tested whether participants in each condition evaluated

the ethicality of the seller diﬀerently, given their assigned per-

spectives. Measures of honesty, trustworthiness, and decep-

tiveness (reverse-coded) were highly correlated in both the

Palterer (α = 0.87, ps < 0.001) and Target (α = .92, ps <

.001) conditions so we took the average of these measures to

form an ethicality index. We ﬁnd that participants assigned

to be targets judged the paltering seller as being less ethi-

cal (M =2.49, SE = 0.13) than did participants assigned to

be the paltering seller (M =3.42, SE = 0.12), t(256)=5.14,

p < 0.001, Cohen’s d=0.64.

Study 4B

In Study 4B, we extend our investigation of the broken

ARTFUL PALTERING IN NEGOTIATION ROGERS, ZECKHAUSER, GINO, NORTON, AND SCHWEITZER

mental model to a computer mediated negotiation. In this

study, we measure both the deceivers’ and the targets’ per-

ceptions of the ethicality of paltering.

Method

Participants. We recruited participants via Amazon’s

Mechanical Turk using an announcement that oﬀered to pay

participants between $3 and $5 and required that they be

located in the United States. We collected data from 200

valid participants (M

age

= 31.69, SD = 9.74; 51% female).

Design and procedure. This study linked negotiators in

an online chat session embedded in the Qualtrics survey plat-

form. The technical back end was an expanded version of the

software initially developed by Brooks and Schweitzer (2011).

We randomly assigned participants to the role of buyer or

seller. For participants in the buyer role, we randomly as-

signed them to one of two experimental conditions: Palter or

Honest.

Participants ﬁrst answered demographic questions and then

read background information for a modiﬁed version of the

Hamilton Real Estate negotiation (Malhotra, 2010), similar

to Studies 1 and 2. In this negotiation, a seller oﬀers to sell a

property to a buyer. The zone of agreement ranges from $38

million to $60 million. The buyer knows that zoning laws will

soon change, and that the property could then be developed

for commercial rather than residential purposes. This zoning

change would make the property much more valuable. The

seller lacks this information. Buyers read a description of the

“public knowledge” to which sellers had access; it did not

explicitly state that sellers may lack this information, but the

description did imply it.

In this negotiation, participants negotiate over the single

issue of price for the property. That is, the negotiation is a

single-issue, distributive negotiation.

In addition to their background information, we gave buy-

ers and sellers additional instructions. We instructed sellers

to ask their buyer two speciﬁc questions during the negoti-

ation: “Are you going to use the property for commercial

development?” and “Are you negotiating with another party

as well?”

In addition, we instructed buyers to employ one of two

strategies in response to an anticipated question about their

intended use of the property. In the paltering condition, we

instructed buyers to palter; speciﬁcally, we instructed them

to avoid answering a property development question directly,

but still to provide a factually correct answer (e.g., by an-

swering “As you know, we have only ever done residential

development”). In the honesty condition, we instructed buy-

ers to “give an accurate answer to this question by answering

it directly and without lying.”

We informed all participants that they would have eight

minutes to complete the negotiation and that they would earn

a bonus based on how well they performed in the negotiation;

sellers would earn $0.25 for every $1 million they received

above $38 million; buyers would earn a bonus of $0.25 for ev-

ery $1 million they paid under $60 million. In a pilot study,

we found that eight minutes was suﬃcient for most negotia-

tors to reach an agreement.

After reading the materials, participants answered atten-

tion and comprehension check questions. All participants

then read all the correct answers, whether or not they an-

swered the comprehension check questions correctly. We then

directed participants to the chat session, in which they nego-

tiated with their counterpart. Neither buyers nor sellers were

told what instructions were given to their counterparts. We

gave participants a two-minute warning before the allotted

eight minutes of negotiation was complete. At the end of the

eight minutes, we told participants to complete the negotia-

tion immediately. In reality, we gave participants who were

still negotiating after eight minutes an additional two minutes

before the session conclusively terminated.

At the conclusion of the negotiation, participants indicated

whether or not they had reached a deal and reported their

ﬁnal negotiated price. As a manipulation check, we asked

sellers what they believed their buyer’s intentions were for

developing the property. We also asked buyers how honest

they were in revealing their intended use of the property on

a seven-point scale that ranged from very dishonest to very

honest.

Finally, we informed sellers that buyers anticipated that

the property would be rezoned for commercial development,

and we asked them to rate, on a seven-point scale, how honest

their buyer had been.

Data exclusions. We recruited 300 participants. Due to

timing and programming constraints and errors, we were un-

able to pair 14 participants and to record data for 24 dyads.

We excluded an additional 11 dyads for the following rea-

sons: comprehension (e.g., at least one of the pair members

failed 40% or more of the comprehension-check questions),

disagreement (e.g., the parties did not agree when reporting

the agreement they had reached), participation in a prior pi-

lot (e.g., one member had participated in the pilot study of

this experiment). We report results from 98 dyads. Of these,

90 dyads reached agreement, and eight reached an impasse.

These inclusion/exclusion decisions do not aﬀect the nature

and signiﬁcance results of the analyses reported below (see

supplemental materials). Based on pilot testing we aimed to

recruit 100 successful dyads so as to have suﬃcient statisti-

cal power. We analyzed this data after all participants had

completed the study, and we report all measures and manip-

ulations.

Results

Manipulation checks. Our manipulation inﬂuenced sell-

ers’ perceptions of buyers’ intentions. Compared to sell-

ers in the Honest condition, sellers in the Palter condition

were more likely to believe, mistakenly, that their counter-

part planned to develop a residential property (83% v. 38%,

(1) = 20.39, p < 0.001). In addition, buyers in the Palter

condition rated themselves as less honest (M =5.0, SE = .23)

than buyers in the Honest condition (M =6.1, SE = .15),

t(95)=4.10, p < 0.001, Cohen’s d=0.85.

Two research assistants coded all the negotiation tran-

scripts, including those excluded from the studies. Coders

were blind to the condition as they coded for the following:

the buyer lied about his/her intentions for the property when

asked; the buyer paltered (told a true statement, but know-

ingly led—or tried to lead—the seller to a false conclusion)

about his/her intentions for the property when asked; the

buyer did not give an answer about his/her intentions for the

property, though the buyer was asked by the seller; the buyer

was not asked by the seller and did not state his/her inten-

tions for the property; and the buyer told the truth about

his/her intentions for the property. The coders rated 50 of

the same transcripts and had nearly perfect reliability, Cron-

bach’s Alpha=0.99.

ARTFUL PALTERING IN NEGOTIATION ROGERS, ZECKHAUSER, GINO, NORTON, AND SCHWEITZER

Table 4

Actual Behavior by Condition, Study 4b

Condition Palter Honest Lie Not Asked Other

Palter

Condition

81% (38) 4% (2) 0% (0) 6% (3) 9% (4)

Honest

Condition

20% (10) 51% (26) 6% (4) 16% (8) 6% (3)

Table 4 reveals that 81% of buyers in the Palter condition

paltered and 51% of buyers in the Honest condition were hon-

est.

Impasse. Negotiation dyads in the two conditions dif-

fered in how often they reached an impasse. Participants in

the Palter condition were signiﬁcantly more likely to reach an

impasse than those in the Honest condition (15% vs. 2%),

(1) = 5.46, p = 0.019. This analysis compares the negoti-

ation outcomes based on what buyers were instructed to do.

An alternative analysis would involve comparing rates of im-

passe based on buyers’ “actual” responses. This entails com-

paring the rate of impasse among all dyads in which the buyer

is coded as having paltered (N=48) to all dyads in which the

buyer is coded as having been honest (N=28), thus looking

at actual responses (see Table 5).

This alternative analysis reveals a similar directional dif-

ference in the likelihood of impasse, though it is not statis-

tically signiﬁcant, χ

(1)=0.539, p=0.46. We note that this

actual response analysis is biased because buyers who were in-

structed to be honest and chose to palter are a diﬀerent kind

of negotiator than buyers who were instructed to palter and

did so. Conversely, buyers who were instructed to palter but

were honest are diﬀerent than buyers who were instructed to

be honest and complied.

Proﬁt. Of buyers who reached an agreement, those in the

Palter condition earned $1.6 million more in proﬁt (16% more)

than those in the Honest condition, t(88)=1.99, p = 0.05.

This result is consistent with Hypothesis 4. When we include

dyads that reached an impasse, we ﬁnd no diﬀerence in proﬁt

earned between the Palter and Honest conditions ($10.6M vs.

$10.5M, t(96)=0.090, p = 0.93).

Evaluation of buyers. After the negotiation concluded,

sellers learned that the buyer anticipated that the property

would be zoned for commercial development. Sellers in the

Palter condition rated their buyer counterpart to be less hon-

est (M =2.8, SE = 0.22) than did sellers in the Honest con-

dition (M=4.6, SE = 0.25), t(95)=-5.34, p < 0.001, Cohen’s

d=1.09.

To examine diﬀerences in perceptions of honesty, we con-

ducted an OLS regression with an interaction term, 2 (role:

buyer vs. seller) χ

(condition: Palter vs. Honest), clus-

tering by dyad. This analysis revealed a marginally signif-

icant interaction, β = −0.69, SE = 0.39, p = 0.081. Buy-

ers in the Palter condition rated their actions as less Honest

(M =5.0, SE = 0.23) than did buyers in the Honest condition

(M=6.1, SE = 0.15; t(95)=-4.1, p < 0.001, Cohen’s d=0.85.

Sellers’ evaluations of buyers’ honesty, however, were much

harsher than buyers’ self-evaluations. Sellers in the Palter

condition rated their buyer’s integrity much lower than the

buyers rated their own integrity (M =2.8 v. M =5.0, t(45)=-

7.01, p < 0.001, Cohen’s d=1.41). This is consistent with our

hypothesis for a broken mental model, as buyers perceived

their own paltering to be somewhat dishonest, while sellers

perceived buyers’ paltering as very dishonest.

Discussion

In both Studies 4A and 4B, we identify paltering as a risky

strategy. Although paltering may increase a negotiator’s sur-

plus, it also increases the risk of impasse and, if discovered,

causes signiﬁcant reputational harm.

Reputation and trustworthiness are critical elements to ef-

fective negotiations (Lewicki, McAllister, & Bies, 1998; Valley

et al., 1998), and paltering places both at risk. Interestingly,

in Study 4B, even when buyers were explicitly instructed to

be honest, 20% paltered anyway. This ﬁnding suggests that

people are both tempted to deceive by paltering and that po-

tential deceivers may perceive paltering as similar to honesty.

Notably, 49% of participants assigned to the Honest condition

engaged in opportunistic and dishonest behavior. This ﬁnd-

ing speaks to the temptation of deception and suggests that

our ﬁndings oﬀer a conservative test of the value-claiming

consequences of paltering versus honesty.

Finally, results from both Studies 4A and 4B highlight the

broken mental model negotiators have for paltering. When

it was made clear to targets that the actor used deception in

answering questions, we found that palterers perceive their

actions to be far more moral than targets do.

Study 5: Paltering When Prompted is More

Unethical than Paltering When Unprompted,

Partly Due to Perceived Sense of Obligation

In our ﬁnal study, we reﬁne our understanding of paltering

by extending our investigation of the broken mental model

we identify in Study 4. In this study, we consider diﬀerences

between palters that were and were not prompted by a direct

question.

Method

Participants. We recruited 255 participants (M

age

35.80, SD = 11.81; female=60%) via Amazon’s Mechanical

Turk using an announcement that oﬀered to pay participants

$0.36 and required that they be located in the United States.

This sample size was chosen to provide adequate power based

on the results of pilot studies. We analyzed this data after

all participants had completed the study, and we report all

measures and manipulations.

Design. Participants were assigned to be the seller in a

hypothetical car negotiation. The scenario was modeled after

the one we used in Studies 2, and 4a. We informed partici-

pants that they listed their car on ebay.com, that their car has

two small dents, that it drives smoothly on the highway, that

it started last week in very cold weather, and that it needed a

mechanic twice in the last four months because the car would

not start (See supplemental materials for full study materi-

als). We then informed these participants that a potential

buyer emailed them and asked if the car had any dents or en-

gine or performance problems. Participants were then asked,

“If you were unwilling to tell the potential buyer something

that you knew to be untrue—but you wanted to sell the car

for the most money possible—which of the following would

you reply to the potential buyer?” (1= The car has two small

dents (the size of a dime) on the back bumper. The car drives

very smoothly on the highway and is very responsive; just last

week it started up with no problems when the temperature was

ARTFUL PALTERING IN NEGOTIATION ROGERS, ZECKHAUSER, GINO, NORTON, AND SCHWEITZER

Table 5

Impasse Frequency, Instructed Versus Actual Responses

(Study 4B)

Instructed to:

Palter Honest Eligible Dyads*

Actual response:

Palter 6 0 48

Honest 1 1 28

Total Dyads 47 51

* “Eligible Dyads” denotes dyads in which the seller asked the

buyer about his/her intentions for the property. “Total Dyads”

denotes all dyads assigned to a condition

freezing. 2= The car has two small dents (the size of a dime)

on the back bumper. Twice in the last 4 months the car would

not start and both times I had to have a mechanic ﬁx it. It

needs a lot of engine work, which will be expensive). The ﬁrst

response option involves a palter because it aims to lead the

buyer to believe that there are no mechanical problems. The

second response is truthful because it transparently discloses

that there are mechanical problems. 73% of all participants

indicated that the paltering response would yield them the

most money possible. The 28% of participants who chose the

more truthful response were asked the same question again.

56% of these participants switched their answer and chose to

palter in the negotiation after being prompted to read the

question again. The remaining participants who indicated

that the truthful response would be in their best monetary

interest were then asked to imagine the scenario in which

they had responded with the palter.

Participants in the Unprompted Palter condition were told

that the potential buyer said “I assume the engine has prob-

lems” and that the potential buyer asked only “How many

dents does the car have?” These participants were also told

that they had paltered about the condition of the engine by

saying “The car drives very smoothly on the highway and is

very responsive. Just last week it started up with no problems

when the temperature was freezing” and told the truth about

the number of dents the car has. Participants in the Un-

prompted Palter condition were told that the potential buyer

left the negotiation believing that the car had no recent me-

chanical problems.

Participants in the Prompted Palter condition were told

that the potential buyer said “I assume the engine has en-

gine problems. Is that correct? How many dents does the

car have?” The rest of the information participants in the

Prompted Palter condition were told was identical to that

told to participants in the Unprompted Palter condition.

Participants then indicated how honest (1=Not at all hon-

est, 7=Very honest), trustworthy (1=Not at all trustworthy,

7=Very trustworthy), and deceptive (1=Not at all deceptive,

7=Very deceptive) they, the seller of the car, were during the

exchange. Participants were then asked to indicate how ob-

ligated they believe they were to say something about the

engine’s performance to the potential buyer (1=Not at all

obligated, 7=Extremely obligated) and how strongly they

thought the potential buyer expected them to say something

about the engine’s performance in the reply email (1=Not at

all strongly, 7=Extremely strongly).

Results

We examined if participants’ evaluations of their own eth-

icality and disclosure-obligation diﬀered based on whether

their palters were prompted or unprompted. Measures of

honesty, trustworthiness, and deceptiveness (reverse-coded)

were highly correlated in both the Unprompted Palter (α =

0.91, ps < 0.001) and Prompted Palter (α = 0.8, ps < 0.001)

conditions so we take the average of these measures to form an

ethicality index. Similarly, we created a disclosure-obligation

index because measures of perceived obligation and expec-

tation to disclose were highly correlated (α = 0.70, p <

0.001). Participants in the Prompted Palter condition judged

themselves as less ethical (M =2.74, SE=0.12) than did

participants in the Unprompted Palter condition (M =3.41,

SE = 0.16), t(253)=3.66, p = .001, Cohen’s d=0.42. Fur-

ther, participants in the Unprompted Palter felt less obliga-

tion to disclose (M =4.98, SE = 0.14) compared to those

in the Prompted Palter condition (M =5.41, SE = 0.11),

t(253)=3.66, p = 0.004, Cohen’s d=0.37.

We then examined whether participants’ disclosure-

obligation index mediated the eﬀect of condition on ethical-

ity. Including participants’ disclosure-obligation index in the

model decreases the strength of the eﬀect of condition on eth-

icality (β

= −0.39, SE = 0.20, p = 0.029). The indirect ef-

fect of participants’ disclosure-obligation index remained sig-

niﬁcant (β

= −.55, SE = 0.063, p < 0.001). To test the

signiﬁcance of the indirect eﬀect of participants’ disclosure-

obligation index on ethicality, we used the Sobel-Goodman

bootstrapping method (Preacher, K. J. & Hayes, A. F. 2004).

The resulting 95% conﬁdence interval for the bias-corrected

indirect eﬀect did not contain zero (lower bound=-0.497, up-

per bound=-0.094; conﬁdence interval without bias correc-

tion: lower bound=-0.483, upper bound=-0.083) suggesting

that participants’ disclosure-obligation index partially medi-

ates the eﬀect of condition on ethicality.

Discussion

In Study 5, we ﬁnd that individuals judged prompted pal-

tering to be less ethical than unprompted paltering. This

study breaks new ground by distinguishing prompted from

unprompted deception. In our investigation, we contrast

prompted and unprompted paltering, but this distinction also

applies to other forms of deception, such as lies of commis-

sion. That is, the extant deception literature has conﬂated

unprompted and prompted lies, and ﬁndings from this study

reveal that this distinction is not only theoretically impor-

tant, but also practically important because the ethical con-

sequences of the two forms of deception may be very diﬀerent.

General Discussion

Deception is pervasive in social life: individuals mislead

their relational partners, their family members, their friends,

and their work colleagues. A large body of literature has

examined two types of deception — lying by omission and

lying by commission — and the resulting stream of research

has informed interventions to decrease the likelihood of be-

ing deceived, such as the important prescription to ask direct

questions to curtail the risk of being deceived by lying by

omission. We identify a third type of deception, paltering,

that we as well as our participants distinguish from both ly-

ing by commission and lying by omission. We deﬁne paltering

ARTFUL PALTERING IN NEGOTIATION ROGERS, ZECKHAUSER, GINO, NORTON, AND SCHWEITZER

as an active form of deception that involves the use of truth-

ful statements to convey a mistaken impression. We demon-

strate that laypeople can readily distinguish paltering from

other forms of deception (Pilot Study 1) and that paltering is

commonly used by experienced negotiators (Pilot Study 2).

Though paltering operates similarly to lying by commission

insofar as it enables individuals to claim value (Studies 2 and

4b) at the risk of incurring reputational damage (Studies 2, 3,

and 4b), we ﬁnd that people view paltering to be less aversive

than lying by commission (Study 1). We also ﬁnd that the am-

biguity inherent in paltering enables self-serving evaluations:

targets perceive paltering as the ethical equivalent of making

false statements, but palterers judge paltering less critically

(Studies 3, 4a and 4b). Finally, some conditions moderate

how ethical palters are perceived to be, notably whether or

not a palter is oﬀered in response to a direct question (Study

5).

Taken together, our studies identify paltering as a distinct

and frequently employed form of deception. Paltering is a

common negotiation tactic. Negotiators who palter claim

value but also increase the likelihood of impasse and, if dis-

covered, risk harm to their reputations. This latter ﬁnding

suggests that those who might view paltering as a (decep-

tive) strategy for claiming more value in a negotiation must

be cautious. It may be eﬀective in the short-term but harm-

ful to relationships if discovered. Paltering is less aversive

to negotiators than lying by commission and just as likely

to be eﬀective. Indeed, these results inform the literature on

self-concept maintenance (Mazar et al, 2008): By using truth-

ful statements, paltering may provide a particularly palatable

form of deception for deceivers.

Our ﬁndings expand our conceptual understanding of de-

ception. By identifying paltering—the active use of truthful

statements to mislead others—as a form of deception, we dis-

entangle active deception from the use of false statements.

This work introduces a novel framework for studying decep-

tion and creates the foundation for a substantial stream of

future work.

Our ﬁndings have particular application to negotiations,

where deception poses a unique challenge. Deception is preva-

lent in negotiations, inﬂuencing the negotiation process, ne-

gotiated outcomes, and negotiator reputations. Prior work

has shown that detected deception harms trust (Boles et al.,

2000; Schweitzer, Hershey, & Bradlow, 2006) and increases

retribution (Boles et al., 2000; Wang, Galinsky, & Murnighan,

2009). Our studies reveal that when detected paltering may

harm reputations and trust just as much as does lying by

commission. Quite possibly, however, negotiators who palter

may misperceive their behavior to be more acceptable than it

is and thus fail to forecast the harmful relational eﬀects their

actions trigger — if their paltering is subsequently detected.

Though we investigate paltering with respect to negotia-

tions, we expect paltering to pervade every interpersonal in-

teraction characterized by information asymmetry and mis-

aligned incentives. For example, paltering is likely to play

an important role in persuasion (Petty & Cacioppo, 1996),

strategic communication (Buller et al., 1994), and impression

management (Leary & Kowalski, 1990). In addition, consis-

tent with our ﬁndings in negotiation contexts, we expect in-

dividuals in these interpersonal contexts to frequently prefer

to palter than to engage in other forms of deception.

Limitations and Directions for Future Research

Future work should extend our investigation of paltering to

further explore a number of related questions. For instance,

under what conditions is paltering most likely to arise rela-

tive to lying by commission? Pilot study 2 with executives

suggests that paltering is prevalent. However, further work,

including ﬁeld studies, should explore the frequency, eﬃcacy,

and perceptions of paltering. Similarly, future work should

explore the short-term and long-term returns on paltering. In

our experiments, we selected both the paltering language and

the contexts in which to palter. Importantly, we ultimately

revealed the deception involved. To fully assess the relative

risk of paltering, however, it is important to discern how likely

paltering and lying by commission are to be uncovered. This

important question merits future research.

Those who palter in everyday life can determine how and

when to palter. Presumably they choose contexts where pal-

tering readily creates false impressions and where detection

of paltering is unlikely. For both reasons, the self-selected

use of paltering may be particularly eﬀective. This may be

especially true if people engage in paltering instead of lying

by commission in domains where the costs of a detected pal-

ter may be less than the costs of a detected lie by commis-

sion (e.g., when a lie by commission would have legal conse-

quences). Future experiments could explore this proposition

and examine how targets of paltering and lying by commission

apportion blame for being misled (e.g., do they blame them-

selves for not asking further questions when they are misled

by paltering).

Our experiments focused on negotiation contexts that are

predominantly distributive. We believe that this is the right

place to begin an investigation, and our ﬁndings demonstrate

that paltering can enable negotiators to claim greater value.

Quite possibly, these ﬁndings would be moderated in ne-

gotiations that are strongly integrative. In these domains,

paltering—like other deceptive tactics—may impair informa-

tion exchange and limit the ability of negotiators to discover

mutually beneﬁcial exchanges (Weingart, Bennett, & Brett,

1993).

Our studies were characterized by single-shot interactions,

though we did inquire after the fact about trust, distrust, and

willingness to negotiate in the future. Future work should

explore the use of paltering in long-term relationships both

within and explore how explicit ex-post justiﬁcations for pal-

tering (e.g., “My statements were true.”) moderate reactions

to it.

Finally, we also call for future work to explore paltering

in contexts outside of negotiations. In some domains, pal-

tering may be even more common than we observe in our

studies. For example, DePaulo and Kashy (1998) found that

people tell fewer self-serving lies to friends than they do to

acquaintances. We postulate that the aversion to lying by

commission to a close friend may prompt the use of paltering,

and as a consequence people may palter more to close friends

than they do to acquaintances.

In addition, we suspect that paltering is likely to be par-

ticularly common in political discourse, as the quote at the

beginning of this paper by President Clinton illustrates. Pal-

tering may enable politicians to mislead others, but retain the

ability to claim that their statements were true. Recent ﬁnd-

ings suggest that politicians are very careful about the claims

they make. Clementson and Eveland (2016) found presidents

tend to refuse to answer questions in press conferences when

speaking to the public, whereas in debates they frequently

change the topic—a tactic that often goes unnoticed (Rogers

ARTFUL PALTERING IN NEGOTIATION ROGERS, ZECKHAUSER, GINO, NORTON, AND SCHWEITZER

and Norton, 2011). We suspect that many politicians who are

keen to avoid lying by commission ﬁnd paltering a tempting

alternative, and we expect studies of political communication

to be a particularly ripe area for future investigation.

Conclusion

We identify paltering as a distinct form of deception. Un-

like lies by omission, paltering involves the active use of state-

ments to create a false impression. Unlike lies by commission,

paltering involves the use of truthful statements to mislead

others. Importantly, paltering readily enables self-serving

assessments of morality. By contrast, if discovered, targets

harshly judge palterers who actively misled them. This con-

trast identiﬁes a broken mental model. How greatly this bro-

ken model matters depends on how likely paltering is likely

to be discovered relative to other deception tactics. Most im-

portantly, we identify paltering as not only a distinct form of

deception, but also a widely employed tactic in negotiations.

REFERENCES

Adler, A. (1930). The education of children. Oxford, Eng-

land: Greenberg.

Allport, G. W. (1955). Becoming; basic considerations for

a psychology of personality (Vol. 20). Yale University

Press.

Anton, R. J. (1990). Drawing the line: An exploratory test

of ethical behavior in negotiation, International Journal

of Conﬂict Management, 1, 265-280.

Aquino, K. (1998). The eﬀects of ethical climate and the

availability of alternatives on the use of deception dur-

ing negotiation. The International Journal of Conﬂict

Management, 9(3), 195-217.

Aquino, K. & Becker, T. (2005). Lying in negotiations:

How individual and situational factors inﬂuence the use

of neutralization strategies. Journal of Organization Be-

havior, 58, 661- 679.

Aquino, K., & Reed, A., II. (2002). The self-importance

of moral identity. Journal of Personality and Social Psy-

chology, 83, 1423-1440.

Bandura, A. (1991). Social cognitive theory of self-

regulation. Organizational Behavior and Human Deci-

sion Processes, 50, 248-287.

Bazerman, M. H., Curhan, J. R., Moore, D. A., & Valley,

K. L. (2000). Negotiation. Annual Review of Psychology,

51, 279-314. Bok, S. (1978). Lying: Moral choices in

public and private life. New York: Pantheon.

Boles, T. L., Croson, R. T., & Murnighan, J. K. (2000).

Deception and retribution in repeated ultimatum bar-

gaining. Organizational Behavior and Human Decision

Processes, 83(2), 235-259.

Bond, C. F., & DePaulo, B. M. (2006). Accuracy of decep-

tion judgments. Personality and social psychology Re-

view, 10(3), 214-234.

Brooks, A. Wood, & Schweitzer, M. E. (2011). Can nervous

nelly negotiate? How anxiety causes negotiators to make

low ﬁrst oﬀers, exit early, and earn less proﬁt. Organiza-

tional Behavior and Human Decision Processes, 115(2),

43-54.

Burgoon, J. K., & Buller, D. B. (1994). Interpersonal de-

ception: III. Eﬀects of deceit on perceived communica-

tion and nonverbal behavior dynamics. Journal of Non-

verbal Behavior, 18(2), 155-184.

Carr, A. Z. (1968). Is business bluﬃng ethical? Harvard

Business Review, January-February, 143-153.

Chance, Z., Norton, M. I., Gino, F., & Ariely, D.

(2011). Temporal view of the costs and beneﬁts of self-

deception. Proceedings of the National Academy of Sci-

ences, 108(Supplement 3), 15655-15659.

Chertkoﬀ, J., & Baird, S. (1971). Applicability of the big

lie technique and the last clear chance doctrine to bar-

gaining. Journal of Personality and Social Psychology,

20, 298-303.

Clementson, D & Eveland, WP (2016). When Politicians

Dodge Questions: An Analysis of Presidential Press Con-

ferences and Debates. Mass Communication and Society,

19, 411- 429

DePaulo, B.M. & Kashy, D.A. (1998). Everyday lies in

close and casual relationships. Journal of Personality and

Social Psychology, Vol 74(1), Jan 1998, 63-79.

DePaulo, B. M., Kashy, D. A., Kirkendol, S. E., Wyer,

M. M., & Epstein, J. A. (1996). Lying in everyday life.

Journal of Personality and Social Psychology, 70(5), 979-

995.

Ekman, P. (1980). Facial asymmetry. Science, 209, 833-

834.

Ekman, P., & Friesen, W. V. (1969). Nonverbal leakage

and clues to deception. Psychiatry, 32, 88-105.

Ekman, P., & Friesen, W. V. (2003). Unmasking the face:

A guide to recognizing emotions from facial cues. Cam-

bridge, Massachusetts: Malor Books.

Ekman, P., & O’Sullivan, M. (1991). Who can catch a

liar? American Psychologist, 46(9), 913. Ekman, P.,

O’Sullivan, M., & Frank, M. G. (1999). A few can catch

a liar. Psychological Science, 10(3), 263-266.

Gaspar, J. P., & Schweitzer, M. E. (2013). The aﬀect de-

ception model: A review of deception in negotiation and

the role of emotion in deception. Working paper.

Gino, F., & Ariely, D. (2012). The dark side of creativ-

ity: original thinkers can be more dishonest. Journal of

personality and social psychology, 102(3), 445.

Gino, F., Shu, L. L., & Bazerman, M. H. (2010). Nameless

+ Harmless=Blameless: When seemingly irrelevant fac-

tors inﬂuence judgment of (un)ethical behavior. Organi-

zational Behavior and Human Decision Processes, 111(2),

102-115.

Kelley, H. H., & Thibaut, J. W. (1969). Group problem

solving. The handbook of social psychology, 4, 1-101.

Knobe, J. (2003). Intentional action in folk psychology:

An experimental investigation. Philosophical Psychol-

ogy, 16(2), 309-324.

Koning, L., Dijk, E. V., Beest, I. V., & Steinel, W. (2010).

An instrumental account of deception and reactions to

deceit in bargaining. Business Ethics Quarterly, 20(01),

57-73.

ARTFUL PALTERING IN NEGOTIATION ROGERS, ZECKHAUSER, GINO, NORTON, AND SCHWEITZER

Kray, L. J., Kennedy, J. A., & Van Zant, A. B. (2014).

Not competent enough to know the diﬀerence? Gender

stereotypes about women’s ease of being misled predict

negotiator deception. Organizational Behavior and Hu-

man Decision Processes, 125(2), 61-72.

Kronzon, S., & Darley, J. (1999). Is this tactic ethical? Bi-

ased judgments of ethics in negotiation. Basic and Ap-

plied Social Psychology, 21(1), 49-60.

Kruger, J., & Dunning, D. (1999). Unskilled and unaware

of it: how diﬃculties in recognizing one’s own incompe-

tence lead to inﬂated self-assessments. Journal of per-

sonality and social psychology, 77(6), 1121.

Leary, M. R., & Kowalski, R. M. (1990). Impression

management: A literature review and two—component

model. Psychological bulletin,107(1), 34.

Leslie, A. M., Knobe, J., & Cohen, A. (2006). Acting in-

tentionally and the side-eﬀect eﬀect theory of mind and

moral judgment. Psychological Science, 17(5), 421-427.

Levine, E. E., & Schweitzer, M. E. (2014). Are liars eth-

ical? On the tension between benevolence and honesty.

Journal of Experimental Social Psychology,53, 107-117.

Levine, E. E., & Schweitzer, M. E. (2015). Prosocial lies:

When deception breeds trust. Organizational Behavior

and Human Decision Processes, 126, 88-106.

Lewicki, R. J. (1983). Lying and deception: A behavioral

model. Negotiating in organizations, 68, 90.

Lewicki, R. J., Litterer, J. A., Minton, J. W., & Saun-

ders, D. M. (1994). Negotiation (2nd ed.). Boston, MA:

Irwin/McGraw-Hill.

Lewicki, R. J., McAllister, D. J., & Bies, R. J. (1998). Trust

and distrust: New relationships and realities. Academy

of Management Review, 23, 438-458.

Lewicki, R. J., & Robinson, R. J. (1998). Ethical and un-

ethical bargaining tactics: An empirical study. Journal

of Business Ethics, 17, 665-682.

Lewicki, R. J., & Spencer, G. (1991, August). Ethical rel-

ativism and negotiating tactics: Factors aﬀecting their

perceived ethicality. Paper presented at the Academy of

Management, Miami, FL.

Malhotra, D. (2010). Hamilton Real Estate (TN). Harvard

Business School Teaching Note 910- 037.

Mazar, N., Amir, O., & Ariely, D. (2008). The dishonesty

of honest people: A theory of self- concept maintenance.

Journal of Marketing Research, 45(6), 633-644.

Murnighan, J. K., Babcock, L., Thompson, L. & Pillutla,

M. M. (1999). The information dilemma in negotiations:

Eﬀects of experience, incentives, and integrative poten-

tial. International Journal of Conﬂict Management, 10,

313-339.

O’Connor, K. M., & Carnevale, P. J. (1997). A nasty

but eﬀective negotiation strategy: Misrepresentation of a

common-value issue. Personality and Social Psychology

Bulletin, 23, 504-516.

Olekalns, M., & Smith, P. L. (2009). Mutually dependent:

Power, trust, aﬀect and the use of deception in negotia-

tion. Journal of Business Ethics, 85(3), 347-365.

O’Sullivan, M. (2003). The fundamental attribution er-

ror in detecting deception: The body-who- cried-wolf-

eﬀect. Personality and Social Psychology Bulletin, 29,

1316-1327.

Petty, R. E., & Cacioppo, J. T. (1996). Attitudes and per-

suasion: Classic and contemporary approaches. West-

view Press.

Pinkley, R. L., Griﬃth, T. L., & Northcraft, G. B. (1995).

“Fixed pie” a la mode: Information availability, informa-

tion processing, and the negotiation of suboptimal agree-

ments. Organizational Behavior and Human Decision

Processes, 62(1), 101-112.

Rogers, T., & Norton, M. I. (2011). The artful dodger:

Answering the wrong question the right way. Journal of

Experimental Psychology: Applied, 17(2), 139.

Schauer, F. & R. Zeckhauser (2009). Paltering, In Brooke

Harrington (ed.), Deception: From Ancient Empires to

Internet Dating (pp. 38-54), Stanford: Stanford Univer-

sity Press, 2009.

Schweitzer, M. E., Brodt, & Croson, R. (2002). Seeing

and believing: Visual access and the strategic use of de-

ception. International Journal of Conﬂict Management,

13(3), 258- 375.

Schweitzer, M. E., & Croson, R. (1999). Curtailing de-

ception: The impact of direct questions on lies and omis-

sions. International Journal of Conﬂict Management, 10,

225-248.

Schweitzer, M. E., DeChurch, L. A., & Gibson, D. E.

(2005). Conﬂict frames and the use of deception: Are

competitive negotiators less ethical? Journal of Applied

Social Psychology, 35(10), 2123-2149.

Schweitzer, M. E., & Hsee, C. K. (2002). Stretching the

truth: Elastic justiﬁcation and motivated communica-

tion of uncertain information. Journal of Risk and Un-

certainty, 25, 185-201.

Schweitzer M. E., Hershey, J. C., & Bradlow, E. (2006).

Promises and lies: Restoring violated trust. Organiza-

tional Behavior and Human Decision Processes, 101, 1-

19.

Shalvi, S., Gino, F., Barkan, R., & Ayal, S. (2015). Self-

Serving Justiﬁcations Doing Wrong and Feeling Moral.

Current Directions in Psychological Science,24(2), 125-

130.

Shell, G. R. (1991). When is it legal to lie in negotiation?

Sloan Management Review, 32(3), 93- 101.

Spranca, M., Minsk, E., & Baron, J. (1991). Omission and

commission in judgment and choice. Journal of Experi-

mental Social Psychology, 27, 76-106.

Tenbrunsel, A. E. (1998). Misrepresentation and expecta-

tions of misrepresentation in an ethical dilemma: The

role of incentives and temptation. Academy of Manage-

ment Journal, 41, 330-339.

Thompson, L. L. (1991). Information exchange in negotia-

tion. Journal of Experimental Social Psychology, 27(2),

161-179.Uhlmann, E. L., Pizarro, D. A., Tannenbaum,

D., & Ditto, P. H. (2009). The motivated use of moral

principles. Judgment and Decision Making, 4(6), 479-

491.

ARTFUL PALTERING IN NEGOTIATION ROGERS, ZECKHAUSER, GINO, NORTON, AND SCHWEITZER

Uhlmann, E. L., Pizarro, D. A., Tannenbaum, D., & Ditto,

P. H. (2009). The motivated use of moral principles.

Judgment and Decision Making, 4(6).

Valley, K. L., Moag, J., & Bazerman, M. H. (1998). A mat-

ter of trust: Eﬀects of communication on the eﬃciency

and distribution of outcomes. Journal of Economic Be-

havior & Organization, 34(2), 211-238.

Wang, C. S., Galinsky, A. D., & Murnighan, J. K. (2009).

Bad drives psychological reactions, but good propels be-

havior responses to honesty and deception. Psychological

Science, 20(5), 634-644.

Weingart, L., Bennett, R., & Brett, J. (1993). The impact

of consideration of issues and motivational orientation

on group negotiation process and outcome. Journal of

Applied Psychology, 78, 504-517.

Received September 15, 2014

Revision received September 13, 2016

Accepted September 18, 2016