Beyond the Horse’S Mouth: Exploring Acquisition and Exchange Utility in Gift Evaluation

ABSTRACT - This investigation examines the contribution of acquisition utility, exchange utility, and their interaction in the evaluation of the gifts that consumers receive. Two experimental studies explore recipients’ evaluation of gifts, and the tendency to reciprocate as a function of giver’s sacrifices, the gift appropriateness, and the giver-recipient relationship. Conjoint analyses and analyses-of-variance reveal the hypothesized additive and multiplicative effects of the two utilities on gift evaluation, they express the importance of social relationships in consumers’ gift evaluations, and they indicate that behavioral and financial costs are imperfectly substitutable in gift-giving. Implications for gift-giving theory are formulated.

Citation:

Rik G.M. Pieters and Henry S.J. Robben (1998) ,"Beyond the Horse’S Mouth: Exploring Acquisition and Exchange Utility in Gift Evaluation", in NA - Advances in Consumer Research Volume 25, eds. Joseph W. Alba & J. Wesley Hutchinson, Provo, UT : Association for Consumer Research, Pages: 163-169.

Advances in Consumer Research Volume 25, 1998      Pages 163-169

BEYOND THE HORSE’S MOUTH: EXPLORING ACQUISITION AND EXCHANGE UTILITY IN GIFT EVALUATION

Rik G.M. Pieters, Tilburg University

Henry S.J. Robben, Nijenrode University

ABSTRACT -

This investigation examines the contribution of acquisition utility, exchange utility, and their interaction in the evaluation of the gifts that consumers receive. Two experimental studies explore recipients’ evaluation of gifts, and the tendency to reciprocate as a function of giver’s sacrifices, the gift appropriateness, and the giver-recipient relationship. Conjoint analyses and analyses-of-variance reveal the hypothesized additive and multiplicative effects of the two utilities on gift evaluation, they express the importance of social relationships in consumers’ gift evaluations, and they indicate that behavioral and financial costs are imperfectly substitutable in gift-giving. Implications for gift-giving theory are formulated.

Research on gift-giving behavior has emphasized stages that individuals go through in selecting, buying, and presenting a gift (cf. Belk 1979; Cheal 1987; Heeler, Francis, Okechucku and Reid 1979; Otnes, Lowrey and Kim 1993; Shrry 1983), focusing on giving instead of receiving gifts. Gift-giving research has emphasized hypothesis generation (cf. Sherry 1983; Belk and Coon 1993), placing less emphasis on hypothesis testing. The multiple hypotheses that have been offered, provide equally numerous opportunities for empirical tests. This study examines the recipients’ perspective, emphasizing factors and processes that determine the evaluation of gifts received. In particular we will explore the effects of gift evaluation on the relationship between giver and recipient, and on the tendency to reciprocate the gift, which have not been examined widely. The study builds on a recently proposed model of product evaluation in which two distinct but related types of utility are distinguished (Thaler 1985; Frenzen and Davis 1990). The model’s components are sufficiently broad to capture a large number of specific mechanisms that have been proposed in previous gift-giving research (e.g., Otnes et al. 1993; Burgoyne and Routh 1991; Sherry 1983; Webley, Lea and Portalska 1983, Webley and Wilson 1989). The models’ relevance for the evaluation of the gifts consumers receive is explored. Two initial experimental studies are offered to test hypotheses about the impact of the giver’s sacrifices, the gift’s appropriateness, and giver-recipient relationships on gift evaluation, and on the tendency to reciprocate.

TWO UTILITIES IN GIFT EVALUATION

Based on the premise that buyers and sellers in the real world may derive value from other sources than only the good or service, Thaler (1985) proposes two kinds of utility, acquisition utility and transaction utility. While acquisition utility depends on the value of the good received compared to the outlay, transaction utility depends on the perceived merits of the transaction (Thaler 1985). In line with this, Frenzen and Davis (1990) argue for a distinction between acquisition utility (U_"), which resides in the good that is transferred or in the service that is provided, and exchange utility (U_E), which resides in the social relationship between the members involved in the exchange. They suggest that the two utilities combine additively to the total utility that a buyer obtains from an exchange: U_T=U_" + U_E. This implies that "... when U_" is held constant, buyers who share strong ties with sellers are able to attain more U_E and thus greater U_T from a given transaction than buyers who share no ties or only weak ties with sellers" (Frenzen and Davis 1990, p. 3). The additive version of the two-utility model specifies that U_" and U_E are independent and that U_E is mainly determined by the preexisting relationship between the parties involved in the exchange. In the context of receiving a gift, the two-utility model would predict that a recipient is more pleased with a gift, to the extent that the value of the gift increases, and to the extent that the relationship with the giver of the gift is better or deeper. This implies that a gift’s low acquisition value can (always) be compensated by high exchange value, the social relationship. We propose to extend the two-utility model in two directions.

First, it may be worthwhile to add a multiplicative component to the model, namely U_" x U_E. Adding this component is fueled by the work of among others Mauss (1969, p.31) who argued that '.. the objects are never completely separated from the men who exchange them; the communion and alliance they establish are well-nigh indissoluble.’ The relevance of adding the multiplicative component derives from the dual functions that exchange utility may have in consumer evaluation processes. Exchange utility may function both as a buffer or rsource pool and as a 'source of norms.’ The additive formulation of the two-utility model highlights the buffer or resource pool function: a good social relationship may compensate for a gift with low transaction utility. The additive formulation of the model does not capture the normative function of social relationships. The stronger the ties between members in a social relationship are, the higher the normative expectations about each other’s behavior tend to be (for example, Duck 1992; Thibaut and Kelley 1959). The effect of the normative function is shown when, for instance, an unwanted gift is received. Then, the disappointment of the recipient is a function of the strength of the social tie with the giver: the closer or deeper the relationship is, the more disappointed the recipient is. This effect cannot be accounted for by an additive model, which would predict that higher levels of exchange utility always lead to higher positive levels of total utility. In the resource function of social relationships, higher levels of U_E lead to higher levels of U_T. In the norm function of social relationships, higher levels of U_E lead to higher positive levels of U_T if U_" is positive, and to higher negative levels of U_T if U_" is negative, all other things being equal, hence U_T=U_" + U_E + (U_" x U_E). In summary, social relationships may have direct (main) and indirect (interaction) effects on the evaluation of gifts received.

Second, it may be worthwhile to distinguish two facets of exchange utility: relationship utility (U_ER) and occasion utility (U_EO). Frenzen and Davis (1990) examined the impact of differences in exchange utility on the decision to buy and on the purchase amount in home-shopping buying. They focused on differences in relationship utility for a single occasion. Home-shopping buying constitutes a specific occasion on which consumers can engage in exchanges and on which they can express the social ties that they share. Clearly, occasions differ in the extent to which social relationships between individuals are salient. Prototypical occasions on which social ties are important include Christmas, Thanksgiving (Wallendorf and Arnould 1991), birthdays (Belk 1979), and Sunday dinners. This leads to the following formulation: U_T=U_" + U_ER + U_EO+ (U_" x U_ER) + (U_" x U_EO) + (U_" x U_ER x U_EO), in which U_T is total utility, U_" is acquisition utility, U_ER is relationship utility, U_EO is occasion utility. Evidently, not all main effects and interactions will always be statistically significant. The size and impact of components in the model is subject to various personal and situational influences. In the current study we explore direct and indirect effects of acquisition and exchange utility, and determine the empirical weights of the proposed utilities in overall gift evaluation, as will be explained below. In the two experiments offered in the sequel, occasion utility is held constant by focusing on birthday gifts. The impacts of transaction and relationship utility on the evaluation of birthday gifts, which are important and highlight the social connections between people (Belk 1979), are examined.

FIGURE 1

UTILITY IN EVALUATING GIFTS RECEIVED

In Figure 1 associations between components of recipient’s gift evaluation are presented. Acquisition utility is assumed to be a function of all sacrifices made by a consumer to acquire a gift. Consumers allocate financial and non-financial budgets which vary in size with the importance of the focal goal (Verhallen en Pieters 1984), in this case selecting an appropriate gift. The actual giver’s financial and behavioral sacrifices made are the input part of acquisition utility. The giver’s sacrifices are aimed at buying an appropriate gift for the recipient, which is the output part of acquisition utility.

Reciprocity in Gift-Giving

Gift-giving is frequently reciprocal (for example, Sherry 1983), at least in a longitudinal sense, and somtimes even simultaneously (for instance, at Christmas or at wedding anniversaries). Reciprocation is so widespread that it permeates exchanges of every kind (Gouldner 1960), and that a developed system of indebtedness which flows from rules of reciprocation is a unique property of human nature (Cialdini 1985). Reciprocity means that individuals engaged in social relationships expect to receive gifts and give gifts in return. Also, gifts should appropriately reflect the intimacy of the relationship and the gift-giving occasion (Robben and Verhallen 1994). Individuals expect to be given gifts, and these gifts should be appropriate to maintain a balanced set of cognitions between recipient, giver, and gift (cf. Belk 1979). It is obvious that the recipient’s reaction to the gift has feedback value both for the recipient and for the donor. The more important it is to please the recipient, the larger the impact of this feedback will be (Camerer 1988). Figure 1 presents this feedback as an effect of consumer evaluation of gifts on the tendency to reciprocate the gift (e.g., Cialdini 1985), and as an effect of consumer evaluation of gifts on relationship utility. Figure 1 displays that acquisition utility and exchange utility combine to influence gift evaluation, subsequent reciprocation tendencies, and future relationship utility.

Belk (1979) described several conditions that may alleviate a recipient from the norm to reciprocate, such as being (very) old or young, and having insufficient means. In addition, the norm of reciprocity may be reduced when the gift is received as a partial repayment of services or as a reward for performance (Belk 1979). Heeler et al. (1979) argue that the reciprocity principle is stronger in close than in superficial relationships.

This study explores the effects of acquisition utility by distinguishing input and output aspects as represented by the giver’s sacrifices versus the gift’s appropriateness. Within the input aspects the role of financial and behavioral costs is examined. The implications of exchange utility (specifically the giver-recipient relationship) and acquisition utility for reciprocation tendencies are analyzed. The results of two experimental studies are offered. Study 1 addresses U_" by manipulating the giver’s effort to present the gift, and Study 2 by manipulating effort, financial price, and appropriateness of the gift. Study 1 investigates U_E by addressing gender and age of the giver, and Study 2 by addressing the effect of giver-recipient relationship on gift evaluation.

STUDY 1: RELATIONSHIP AND SACRIFICES

In Study 1, age and gender of giver and effort to acquire the gift are manipulated in a 2³ factorial within-subjects design. Previous research has found that relationship with the giver as represented by age and gender of the giver affects gift evaluation. Age is an important factor in gift evaluation because of its implicit status connotations (see Burgoyne and Routh 1991). We expect higher age of a giver to imply a higher status of the giver, and gifts of higher status givers are evaluated more positively. In gift exchanges, gender is important because gifts given to same gender friends indicate (expected or non-threatening) liking or respect, while gifts given to opposite gender friends may suggest (unexpected or undesired) romantic motives (see Belk and Coon 1993). Effort invested into gift acquisition, i.e., the input component of acquisition utility, indicates the relative importance of recipient or gift exchange to the giver (Robben and Verhallen 1994). Extending the reasoning of Camerer (1988, S195) we hypothesize that "Any net cost of time, energy, or imagination is part of the signaling cost of a gift: the thought does count."

TABLE 1

GIFT RECEIVING SITUATIONS AND PREFERENCES: STUDY 1

Age, gender, and effort will have significant main effects on gift evaluation. In addition, the two-utility model predicts significant interaction effects between the componentsof U_" and U_E. An Age by Gender interaction is expected, as this represents status differences between giver and recipient (cf. Burgoyne and Routh 1991), and an Age by Effort interaction, as this indicates the qualifying effects of behavioral costs (Robben and Verhallen 1994).

Method

Subjects and Measures. Fifty-five undergraduates in Marketing participated as a course requirement. Mean age of the 23 women and 32 men was 22 years. Subjects received a ten-page booklet containing descriptions of eight gift situations, questions concerning gift evaluation, and several background questions. The gift situations represented all conditions in a 2 (older versus younger) by 2 (same gender versus opposite gender) by 2 (not much effort versus much effort) within-subjects design. The order of the eight situations was randomized. Subjects were asked: 'Please read each description carefully. Try to imagine how you would feel and how you would react in that situation. Next, complete the questions below each description.’ The situations read: 'Imagine that you receive a present in this situation. Read the description carefully. From an 'X’ person of 'Y’ you receive a birthday present. You have the impression that the donor has invested 'Z’ to choose a nice present.’ In the original descriptions, the letters between quotations were replaced by: X='younger’ versus 'older’ (factor: Age of giver), Y='the same gender’ versus 'the opposite gender’ (factor: Gender of giver), and Z='much effort’ versus 'not much effort’ (factor: Effort by giver).

As a manipulation check, subjects indicated for each situation whether they felt their own status was 'higher,’ 'the same,’ or 'lower’ than that of the donor, and they indicated on a 4-point scale whether they thought the donor had put 'a lot of effort-no effort at all’ in each gift-giving situation. In addition, subjects indicated their preference for each of the eight situations by ordering them from most (1) to least (8) liked.

Manipulation checks. As expected, subjects perceived their own status to be significantly lower when they received a gift from an older person (X=2.10) than when they received a gift from a younger person (X=2.29; F (1,54)=20.57, p<.001). Also, subjects felt that an other-gender giver (X=5.20) was significantly nicer than a same-gender giver (X=4.82; F (1,54)=9.68, p<.003), although both were 'liked.’ Finally, subjects recognized that givers who invested much effort in finding a nice gift had gone through more trouble (X=3.15) than givers who had invested less effort (X=1.47; F (1,54)=352.41, p<.001).

Results

Table 1 offers the design of the gift situations. The final column displays the average preference rank score for the eight gift receiving situations, with '1’ representing the highest preference and '8’ the lowest preference.

Conjoint analyses were performed on subjects’ preferences for the eight gift receiving situations. Conjoint analysis estimates the structure of preferences given overall evaluations of a set of alternatives that are pre-specified in terms of levels of different attributes (Green and Srinivasan 1990). Part-worths indicate the influence of a factor level on preference for a particular gift receiving situation. For each factor, an 'importance’ score is calculated by taking the utility range for the factor and dividing it by the sum of all utility ranges. The higher the score, the more important a factor is in the overall solution. Finally, Kendall’s tau statistic, a rank-order correlation coefficient, is calculated as an overall measure of fit between observed and estimated preference orderings.

TABLE 2

CONJOINT ANALYSIS ON PREFERRED GIFT RECEIVING SITUATIONS: STUDY 1

Three models are fitted. In model A, only main effects of Age, Gender, and Effort are estimated. In model B, both main efects and the three first-order interaction effects are estimated. In model C, main effects, first-order interaction effects, and the second-order interaction effect are estimated. If the proposed two-utility model holds in the experimental situation, models B and C outperform model A, the additive model. Table 2 presents the results of the estimation procedure.

The results show that model A insufficiently accounts for the observed preference orderings: the overall fit between observed and estimated preference orderings is low and statistically insignificant (Kendall’s t=.327, p=.131). Model B fits the observed data well as indicated by a highly significant correlation coefficient (t=.852, p=.002). Including the second-order interaction in model C increases the fit (t=1.00, p<.001), but Model B appears more parsimonious.

In model B one main effect, Age, and two interactions, Age x Gender and Age x Effort, dominate the preference orderings, as indicated by importance scores. The main effects indicate that gifts received from older persons (average rank=4.28) are preferred to gifts received from younger persons (average rank=4.72).

The Age by Gender interaction indicates that gifts from and older, same gender person (average rank=3.89) are preferred to all other reception situations, which are close in preference (average rank=4.70). Gifts from an older, higher status, person (Burgoyne and Routh 1991), communicate the most positive, and least ambiguous message (Camerer 1988).

The Age by Effort interaction adds additional insight: gifts received from an older person investing much effort are most preferred (average rank=4.08), and gifts received from a younger person investing much effort are least preferred (average rank=5.08). Gifts in other situations are ranked in between the two extremes. Investment of effort in a gift is better when it is received from the right (high status) giver, than when it is received from the wrong (low status) giver. These results support the relevance of the multiplicative component in the two-utility model of gift evaluation, and they indicate that relationship utility interacts with transaction utility.

The within-subjects design in study 1 may stimulate demand effects. In addition, the transparent manipulations and straightforward scenario could sensitize our student subjects to make distinctions that they do not make in the real world. Study 2 extends study 1 by employing a between-subjects design, a more realistic scenario and a non-student sample of subjects. Also, additional aspects of transaction and relationship utility are examined.

STUDY 2: SACRIFICE, APPROPRIATENESS, AND RELATIONSHIP

The gift’s appropriateness as well as the giver’s financial and behavioral costs are added in a 2⁴ between-subjects factorial design. In addition, the recipient’s tendency to reciprocate the gift is explored. It is hypothesized that compared to recipients of gifts in superficial relationships, recipients in close relationships expect, value and intend to reciprocate appropriate gifts, and gifts associated with much effort to a larger extent as these instances emphasize and enhance exchange utility.

Method

Design. Four factors were manipulated in a 2⁴ full-factorial between-subjects design: closeness of relationship (low: same gender, superficial acquaintance; high: same gender, good friend), effort invested by giver (low: little effort invested and little consideration; high: very much effort and very much consideration), relative price for giver (low: inexpensive; high: very expensive), and appropriateness of the gift (low: inappropriate or already owned by recipient; high: appropriate and wanted by recipient).

Subjects and Measures. Subject were selected by randomly drawing 400 numbers from national telephone directories in The Netherlands. Of 264 persons contacted, 160 (61%) agreed to participate in a telephone survey, 95 females (59%) and 65 males (41%). Subjects were contacted and interviewed by one of four trained female telephone operators of a market research company. To avoid bias, interviewers worked with varying gift-giving scenarios. The experimental factors were incorporated into gift-giving scenarios that were read to the participants. An example of such a scenario is: 'Imagine yourself in the following situation: It is your birthday and you receive a gift from a (same gender) superficial acquaintance. The giver has invested very little effort in finding an appropriate gift for you, and has given the gift little thought. The gift is inexpensive for the giver; it only costs a few guilders. When you receive the gift, you notice that you already own one or that it is not appropriate.’

Subjects indicated whether they expected a birthday gift from this person (1='Completely unexpected,’ 7='Completely expected’), how happy they felt with the gift (1='Very disappointed,’ 7='Very happy’), how they evaluated the relationship with the giver (1='Very superficial,’ 11='Very close’), the probability of reciprocating on the giver’s next birthday (1='No chance,’ 7='Certain’), the effort they intended to exert for this reciprocation (1='No,’ 7='Everything possible’), the minimum and maximum amounts of money they would be willing to spend on that gift (in US$ from: 1='Less than 1.25 US$,’ to 9='More than 10 US$’, and from 1='Less than 2.50 US$,’ to 12='More than 50 US$’). Response alternatives were read to the participants.

Results

Table 3 contains the means of the dependent variables as a function of Closeness of the relationship, Effort, Price, and Appropriateness. Table 4 summarizes the main and interaction effects in the analyses-of-variance.

Gift Expectations and Evaluation. The expectation of receiving a gift is higher in a close relationship than in a superficial relationship. Consumers expect to receive a gift which contains much effort more than a gift which contains little effort, expect an inexpensive gift more than an expensive gift, and expect an appropriate gift more than an inappropriate gift. These results are indicative of the norm function of relationship utility. Two first-order interactions emerge: for close relationships, the expectation of receiving a gift increases with the appropriateness of the gift (X=3.79; Relationship x Appropriateness) and the effort invested in the gift (X=3.94; Relationship x Effort). These effects were supported by a second-order interaction between Effort, Price and Appropriateness, also indicative of normative functions of social relationships. As expected, subjects evaluated a high effort gift more positively than a low effort gift, and an appropriate gift more than an inappropriate one, although the effect is only marginally significant.

Relationship and Reciprocation. The gift’s appropriateness appeared to have a feedback effect on the evaluation of the giver-recipient relationship. The relationship was deemed closer when an appropriate gift was received (X=7.74) than when an inappropriate gift was received (X=6.81; F (1,135)=10.30, p<.005). The probability of reciprocation was higher for gifts in a close relationship than for gifts in a superficial relationship, and higher for a high effort gift than for a low effort one. Interestingly, no significant main effects of the gift’s financial price and of the gift’s appropriateness on the tendency to reciprocate emerged. A first-order interaction approaching significance indicated that for superficial relationships, the probability of reciprocation increased with the gift’s financial price (X=5.78; Relationship x Price). A second-order effect approached significance (Effort x Price x Appropriateness), suggesting that te probability of reciprocation is lowest for an inappropriate, low effort and inexpensive gift, with little differentiation among the other conditions. However, for appropriate gifts, the probability of reciprocation increased for low effort, inexpensive gifts, and for high effort, expensive gifts. It was lower for low effort, expensive gifts and for high effort, inexpensive ones. The interaction suggests that appropriate gifts should also exhibit a balanced mix of behavioral and financial costs. Low effort, expensive gifts, and high effort, inexpensive ones may signal to the recipient that something unexpected has occurred in the gift buying process. These results indicate that behavioral costs may be more important than the gift’s appropriateness and financial price, and that the giver-recipient relationship plays a dominant role in gift evaluation.

Behavioral and Financial Costs. The expected effort invested in a return gift was higher for a high effort than for a low effort gift, and higher when an appropriate gift was received than when an inappropriate gift was received. This indicates a matching principle in behavioral investments in gifts. The significant Relationship by Effort interaction showed that exchange utility qualified these intentions, since the expected effort to be invested was higher in a close than in a superficial relationship.

Relationship utility had a significant effect on the amounts recipients are willing to spend on the gift they would return: the minimum and maximum amounts of money were higher in a close than in a superficial relationship. Interestingly, while the price of the gift received had no effect on the tendency to reciprocate, it did affect the minimum and maximum amounts of the gift that one would reciprocate. Minimum and maximum amounts are higher when the received gift is expensive instead of inexpensive. Several first-order interactions reach significance. The minimum amount for the return gift is highest for a high effort gift in a close relationship (X=7.48; Relationship x Effort), lowest for a low-priced and inappropriate gift (X=4.85; Price x Appropriateness), and it rose for a high effort appropriate gift (X=7.15; Effort x Appropriateness). The maximum amount was highest for an appropriate gift in a close relationship (X=8.21; Relationship x Appropriateness), and lowest for an inappropriate, inexpensive gift (X=4.71; Price x Appropriateness).

DISCUSSION

The results indicate that additive and multiplicative effects of U_" and U_E drive the recipient’s evaluation of gifts. Consumers determine the value of a gift they receive by balancing the acquisition utility of the gift, both the sacrifices of the giver and the gift’s appropriateness, with the exchange utility of the gift, both characteristics of the giver, such as age and gender, and the depth of the relationship with the giver. The two-utility model reflects previous efforts in gift evaluation research by subsuming factors studied earlier (for instance, social status) as represented by U_E (social relationship), by a combination of U_" and U_E (Age x Effort), or by U_" (for instance, money amounts involved). The observed direct effects of the giver-recipient relationship on gift evaluation, reciprocation tendencies and the amounts recipients were willing to spend in their return gifts, and the indirect effects of the relationship in combination with gift attributes underline the importance of the social imbeddedness of utility in gift-giving.

Specifically, the two studies support findings reported in the literature, such as the effects of giver’s age and gender on gift evaluation (Burgoyne and Routh 1991), but extended them as well by incorporating the U_" and U_E interaction terms. In study 1, age of the giver plays a dominant role. Both its main effect, and its interactive effects with gender and effort of the giver affect the preference orderings of the recipients significantly. The statistically significant Age by Gender interaction in study 1 specifies two highly valued gift receiving situations. A gift from a same gender, older donor implies a high difference in relative status but probably reflects a non-intimate gift situation. In contrast, a gift from an opposite gender, younger donor reflects similar social status between giver and recipient but probably portrays them in a more intimate gift situation.

TABLE 3

MEANS AS A FUNCTION OF CLOSENESS OF RELATIONSHIP, EFFORT, PRICE, AND APPROPRIATENESS: STUDY 2

TABLE 4

RESULTS OF ANALYSES OF VARIANCE: STUDY 2

The analyses of study 2 revealed that the effort invested in the gift by the giver had a significant effect on the tendency of the recipient to reciprocate the gift, and on the effort s/he would invest in reciprocating, but not on the amount of money invested in the gift. On the other hand, the amount of money invested in the gift by the giver had no effect on the tendency of the recipient to reciprocate the gift, and on the effort invested in reciprocating, but that it had significant effects on the amounts of money invested in the returned gift. These results are new and intriguing. Although the rule of reciprocity is powerful, the results indicate that it is not invoked homogeneously in gift-giving. For birthday presents, the financial value of the gift has less effect on reciprocation tendencies than the effort or "behavioral" value has. But when reciprocating a gift, a high financial value stimulates gifts of high financial value, while a high "behavioral" value stimulates gifts of high "behavioral" value. Taken together these results support Foa's (1971) findings that in interpersonal exchanges, people prefer to exchange like for like, i.e., money for money, and effort for effort.

The finding in study 2 that appropriateness of the gift affects the experienced closenss of the relationship supports the feedback mechanism in our conceptual framework. Appropriateness of the gift signals the uniqueness of the recipient, and the special relationship between giver and recipient (Camerer 1988). In close relationships appropriate gifts are given, and when appropriate gifts are given, the relationship tends to become closer; hence acquisition and transaction utility mutually influence each other in developing relationships.

Limitations

The two studies are subject to several limitations. First, the two-utility model that formed the basis of the studies is general, and does not allow to formulate very specific questions about the impact of factors on gift evaluation. This is a clear disadvantage. Moreover, each of the two utility facets is itself multi-faceted. While rather general, the two-utility model helps to structure the relevant forces in gift evaluation, and their resultant effects. It explicitly distinguishes the resource and norm functions of social relationships. Still, much more work is needed to more specifically define the various indicators of the utility components proposed.

Even with all possible model specification, mathematical descriptions only reveal a limited part of the process of gift evaluation. Due to the social nature of gift giving, and the complexity of gift, gift transfer and context, only a combination of quantitative and qualitative research can reveal the experiential qualities that make gift giving so impactful to both recipient and giver.

In summary, we believe that the two-utility model provides an elegant framework to formulate and test hypotheses about the way components of utility combine into overall evaluations of the gifts consumers receive. The model calls for further extension and refinement, as it is still rather crude in its current specification. For instance, the fining that behavioral costs of the gift appear to influence the tendency to reciprocate, while financial costs of the gift appear to influence the price range of the reciprocated gift may be explicitly modeled in future research.

REFERENCES

Belk, Russell W. and Gregory S. Coon (1993), "Gift Giving as Agapic Love: An Alternative to the Exchange Paradigm based on Dating Experiences," Journal of Consumer Research, 20, 393-417.

Belk, Russell W. (1979), "Gift-giving Behavior," in Research in Marketing, vol. 2 , ed. in Jagdish Sheth, Greenwich, Connecticut: JAI Press, 95-126.

Burgoyne, Carol B. and David A. Routh (1991), "Constraints on the Use of Money as a Gift at Christmas: The Role of Status and Intimacy," Journal of Economic Psychology, 12, 47-69.

Camerer, Colin (1988), "Gifts as Economic Signals and Social Symbols," American Journal of Sociology, 94 (supplement), 180-194.

Cheal, David (1987), "Showing Them You Love Them: Gift-giving and the Dialectic of Intimacy," Sociological Review, 35, 150-169.

Cialdini, Robert B. (1985), "Influence: Science and Practice," Glenview, Ill.: Scott, Foresman and Company.

Duck, Steve (1992), "Human Relationships," 2nd edition. London: SAGE Publications Ltd.

Foa, Uriel G. (1971), "Interpersonal and Economic Resource," Science, 171 (2 January 191), 345-351.

Frenzen, Jonathan K. and Harry L. Davis (1990), "Purchasing Behavior in Embedded Markets," Journal of Consumer Research, 17, 1-12.

Gouldner, A.W. (1960), "The Norm of Reciprocity: A Preliminary Statement," American Sociological Review, 25, 161-178.

Green, Paul E. and V. Srinivasan (1990), "Conjoint Analysis in Marketing: New Developments with Implications for Research and Practice," Journal of Marketing 54, 3-19.

Heeler, Roger, J. Francis, Chike Okechuku, and S. Reid (1979), "Gift versus Personal Use Brand Selection," Advances in Consumer Research, 6, 325-328.

Mauss, Marcel (1969), "The Gift," London: Routhledge and Paul Kegan.

Otnes, Cele, Tina M. Lowrey and Young Chan Kim (1993), "Gift Selection for Easy and Difficult Recipients: A Social Roles Interpretation," Journal of Consumer Research, 20 (2), 229-244.

Robben, Henry S.J. and Theo M.M. Verhallen (1994), "Behavioral Costs as Determinants of Cost Perception and Preference Formation for Gifts to Receive and Gifts to Give," Journal of Economic Psychology, 15, 333-350.

Sherry, John F. Jr. (1983), "Gift Giving in Anthropological Perspective," Journal of Consumer Research, 10, 157-168.

Thaler, Richard (1985), "Mental Accounting and Consumer Choice," Marketing Science, 4, 199-214.

Thibaut, John W. and Harold H. Kelley (1959), "The Social Psychology of Groups," New York: John Wiley and Sons, Inc.

Verhallen, Theo M.M. and Rik G.M. Pieters (1984), "Attitude Theory and Behavioral Costs," Journal of Economic Psychology, 5, 223-249.

Wallendorf, Melanie and Eric J. Arnould (1988), "My Favorite Things: A Cross-Cultural Inquiry into Object Attachment, Possessiveness, and Social Linkage," Journal of Consumer Research, 14, 531-547.

Webley, Paul, Stephen E.G. Lea, and R. Portalska (1983), "The Unacceptability of Money as a Gift," Journal of Economic Psychology, 3, 223-238.

Webley, Paul, an Richenda Wilson (1989), "Social Relationships and the Unacceptability of Money as a Gift," Journal of Social Psychology, 129, 85-91.

----------------------------------------