Working with dealers in India.

U.S. participation in the Caribbean Basin Initiative, the European Lome Convention and ongoing support of the North American Free Trade Agreement (NAFTA) is solid evidence that greater trade with developing nations is a key objective of U.S. trade policy. With the enormous trade deficit of

the United States, opportunities to cultivate healthy trade relations with developing nations are among the most attractive possible actions that U.S. firms can pursue. A major obstacle in the cultivation of market opportunities in developing countries is a shortage of established wholesale and retail businesses. Firms that want to enter these markets are often forced to invest heavily in the development of local entrepreneurs to create an infrastructure through which trade can flow (Czinkota and Ronkainen, 1993; Moore, 1993). Heavy investments in facilities and training of newly recruited foreign distributors are common.

The Indian market is particularly attractive to multinational marketers. There are more than 150 million middle-class consumers who belong to well educated families with salaries in excess of $4,000 a year in local purchasing power. A sustained taste for foreign consumer durables begins at an annual income of only $2,000 in India (The Economist, 1994). This makes India's market potential greater than that of many countries in Western Europe. The potential for rapid market growth in India is staggering. Coca Cola, for example, need only raise the Indian per capita consumption of its product to 30 percent of its U.S. level to create a company as large as Coke USA is today (Moore, 1993).

The Indian government in recent years has dramatically altered key policies and is now allowing foreign companies to be majority stockholders in Indian ventures. During the 1970s, foreign equity ownership in Indian ventures was restricted by law to a maximum of 49 percent. IBM and Coca Cola were among many foreign corporations doing business in India that elected to leave that market rather than give up full ownership of their operations. In a series of moves during the 1980s, the restrictions were gradually eased and it was made easier for foreign firms to repatriate assets and profits earned in India. Recent financial reforms allow full convertibility of the Indian currency for international trade and many import license requirements for industrial products have been scrapped. IBM, Coke and numerous other western firms have returned to India. Many other American firms (e.g., PepsiCo, McDonalds, Ford, AT&T, and General Electric) have invested heavily in the Indian market recently. Foreign direct investment in India now approaches $4 billion (Bensky et al., 1993) and continues to grow rapidly.

The growth of the Indian economy and the entry by foreign marketers is likely to result in the recruitment of many new dealers and distributors, many of whom will be small-scale entrepreneurs. Entering western firms will have to build and support those entrepreneurs. Many western firms have entered growing foreign markets with a business orientation and philosophy which treated foreign distributors as independents and not partners. Like Coke (Moore, 1993), some found themselves at the mercy of local distributors who did not share common goals and philosophies. The development of effective communication strategies to manage relationships with local distributors is one of the greater problems faced by Western firms that are new entrants to the Indian market.

In this study we examined how supplier communication patterns with newly recruited dealers differ from those with relatively more established dealers. While the recruitment of distributors will be vital to foreign market expansion, Indian distributors will probably have to be managed with somewhat different communication strategies than are appropriate in the west. This paper also details how supplier-dealer communication patterns and dealer perceptions of the strength of their relationship (relationalism) with a supplier affects dealer satisfaction. Relationalism is a key ingredient for building strong economic and social ties with business partners. The existence of relationalism leads to the practice of relationship marketing which, according to Kotler and Armstrong (1994), focuses on building long-term, value-laden relationships and marketing networks. A model of the relationships among the key variables which are examined here is shown in Figure I.

[Figure I ILLUSTRATION OMITTED]

The results of our study should be of interest to multinational marketers and marketing academicians alike. There has been some previous research on interfirm communication in Indian distribution channels (cf. Kale, 1986, 1993; Frazier et al., 1989). However, differences between new and established dealers have not been reported before.

LITERATURE REVIEW AND STUDY HYPOTHESES

To determine if suppliers communicate differently with newly recruited dealers than they do with more established dealers, five facets of interfirm communication (as suggested by Mohr and Nevin, 1990) are examined: direction, frequency, formality, method, and content of supplier-dealer communication. Direction refers to whether the communication is initiated mostly by the supplier, or the dealer, or about equally by both. Frequency indicates how often the supplier and the dealer communicate with each other, leading to the total amount of communication between them. Formality refers to whether information is transmitted mostly through regularized and structured (formal) or spontaneous and unstructured (informal) modes. Method refers to the use of personal and impersonal (such as, facsimile, mail) communications. Content of communication measures the nature of influence strategies that suppliers use to persuade dealers to behave in prescribed manners. Following Frazier and Summers (1984), influence strategies are classified as indirect (information exchange and recommendations which seek to alter dealer perceptions) or direct (requests, promises, threats, and legal pleas which attempt to influence dealer behavior while giving minimal attention to altering perceptions).

Four major hypotheses are proposed and tested. These are summarized in Table 1.

Table 1
Summary of Study Hypotheses

Hypothesis                New Dealers           Established Dealers

[H.sub.1] Communication
between dealers and
suppliers:
   a. Directionality      More unidirectional   More bidirectional
   b. Frequency           Greater               Lesser
   c. Formality           More informal         More formal
   d. Method              More personal         More impersonal
   e. Content             More direct           More indirect

[H.sub.2] Strength of
relationship              Lower                 Higher

[H.sub.3] Relationship    Positively            Positively
and communications        correlated            correlated
pattern

[H.sub.4] Dealer
satisfaction with the
supplier is affected
by:
   a. communication       Positively            Positively
      pattern             correlated            correlated
   b. relationalism       Positively            Positively
                          correlated            correlated

The first hypothesis is:

[H.sub.1]: Suppliers use different patterns of

communication with new and established dealers.

Support for this hypothesis follows in the form of five subhypotheses.

Direction. In doing business with new dealers who are relatively inexperienced, suppliers need to properly educate them in the various aspects of business and direct them in most of their business activities. Therefore, communication with new dealers is likely to be initiated mostly by the supplier. If the supplier does not communicate enough, the dealers will tend to initiate communication with their inquiries. In either case, the communication is likely to be unidirectional, from supplier to dealer, or vice versa. On the other hand, established dealers will be more familiar with the business and probably favor a more balanced flow of information from both sides, thus making the communication more bi-directional. This leads to our first sub-hypothesis:

[H.sub.1a]: Communication between suppliers

and new dealers is more unidirectional

and that between suppliers and

established dealers is more bi-directional.

Frequency. New dealers usually need to have more frequent communication with their suppliers as they are still in the "learning stage" and are still working out many aspects of their relationship with their suppliers. Established dealers do not need to have such frequent communication because they are likely to have their business procedures more firmly established. Anderson, Lodish and Weitz (1987) suggest that communication between two channel members leads to trust, and trust is an important prerequisite for building relationship commitment (Morgan and Hunt, 1994). Therefore, suppliers should communicate very frequently with new dealers to build trust and relationship commitment. The need for such a high level of communication with more established dealers is less after trust and commitment are established. Therefore, we hypothesize:

[H.sub.1b]: The frequency of communication

between suppliers and new dealers is

greater than that between suppliers

and established dealers.

Formality. New dealers, who are mostly beginning entrepreneurs, are more likely to be virtually consumed by the demands of their business. They are typically short of a very valuable asset--time. As they have few others in the organization to depend on, they must do virtually everything themselves. New dealers are thus more likely to rely on very personal and informal modes of communication with their suppliers. Rather than write a letter to find out about upcoming promotions, a new dealer is likely to call a sales representative or even to quiz the warehouse clerks to learn what s/he can.

The social hierarchy of the Indian culture also suggests that a dyadic business relationship is likely to be very rigid when the two parties are not very well known to each other. Ironically, the rigidity in the social system causes many Indians to avoid formal communications. Rather than write letters documenting their problems and complaints to their suppliers, they tend to communicate through informal modes and even through third parties. There is a great concern that their lack of knowledge of what their business partners want might lead them to commit embarrassing mistakes. In a U.S. study, Mohr (1989) found that trading partners resisted the use of formal communication modes until their trust and commitment to the exchange relationship blossomed. Thus, the preference for informal communication modes during the early stages of a relationship may not be limited to traditional societies. As time passes and trading partners begin to feel more comfortable with each other, the use of formal, more standardized communication modes is likely to increase. Hence, we offer:

[H.sub.1c]: Communication between suppliers

and new dealers is more informal than

that between suppliers and established

dealers.

Method. The actual method of communication used is also likely to depend on the length of a supplier's relationship with a dealer. This follows from Mohr (1989) and the argument presented for the previous hypothesis (1c). In the early stages of the development of an international exchange relationship, personal face-to-face contact will probably be preferred for the proper "grooming" of the relationship. Jaworski (1988) suggests that informal controls are likely and appropriate in uncertain environments. New relationships are likely to be more uncertain, on average, than are established relationships. Thus, personal contacts, an informal control mechanism, should be used more. For channel partners who have known each other for some time, communication by telephone or mail may be considered to be more convenient and cost-effective. These considerations lead to our next hypothesis.

[H.sub.1d]: Communication takes place more

through personal contact (both absolutely

and relative to other nonpersonal

methods of communication)

between suppliers and new dealers

than between suppliers and established

dealers.

Content. Direct influence strategies are more likely under asymmetrical power situations in a channel (Frazier and Summers, 1984; Kale, 1986). New dealers, mostly "start-up" entrepreneurs in India, possess little channel or market power in comparison to dealers who have established themselves and cultivated considerable patronage loyalty in their markets. Further, it is more convenient and effective for suppliers simply to tell relatively new and inexperienced dealers what to do rather than invest time and effort to try to change their perception of what behavior is most appropriate or in their best interest. Established and more experienced dealers are likely to be more effectively persuaded if the supplier first tries to change their perception regarding what actions could be mutually beneficial rather than dictate the dealer behavior that is desired by the supplier. This leads to the next hypothesis:

[H.sub.1e]: Suppliers use direct influence strategies

more frequently (both absolutely

and relative to the use of indirect

influence strategies) when communicating

with new dealers than when

communicating with established dealers.

Relationalism. Interfirm relationalism and its impact on international marketing is an important theme in contemporary marketing thought (Bello and Donthu, 1993; Kale, 1993). Arndt (1979) described domesticated markets as those that emphasize interdependencies, interactions, reciprocities and long-term commitments. Achrol (1991) wrote of "quasicorporations" which involve complex exchange relationships among partner firms that are governed by sophisticated sociopolitical systems. Dwyer, Schurr and Oh (1987), drawing upon the work of Macneil (1980) and Scanzoni (1979), described relational exchanges as alternatives to discrete exchanges. Kaufmann and Dant (1992) and Boyle et al. (1992) have offered complementary measures of relationalism which include, among others, expectations that (1) an exchange relationship will endure over time, (2) benefits and burdens that emerge will be shared, (3) the partners will share mutual trust, and (4) planning for future transactions will take place. These are facets of relationalism which are expected to evolve over time. High levels of trust, sharing and commitment, for example, are likely to emerge only after trading partners learn through experience what each is like and expects from the relationship.

India is on the path of industrialization but remains somewhat of a traditional society (Terpstra and David, 1991). Persons in traditional societies are, in comparison to these in modern societies, somewhat more formal, emotional and resistant to change (Harris and Moran, 1987; Terpstra and David, 1991). Relationship cultivation may take many years before high levels of trust and commitment are established. Further, new dealers in India as well as other traditional societies are more likely to expect to be directed precisely in their business activities. Thus, they are less likely to expect to be involved in joint planning and decision making for mutual advantage which are key facets of interfirm relationalism. It is likely that after many years of doing business, a well established Indian dealer may become more receptive to a supplier's expectations of such relational behavior. However, new dealers in India are much less likely to share these feelings and expectations. Thus, we propose this hypothesis:

[H.sub.2]: Supplier-dealer relationalism

is lower among new dealers than among

established dealers in India.

Relationalism and Communications. The link between cultural context and communications in India has been demonstrated (Frazier et al., 1989). Mohr and Nevin (1990) proposed that channel relationalism is likely to make interfirm communications more frequent, bidirectional, indirect and informal. Based on a study of U.S. automotive passenger tire dealers, Boyle et al. (1992) provide support for this thesis. Mohr (1989) found, however, that relational channel structures led to more rather than less formal communication.

If channel relationalism does take time to develop and is less intense in dyads with new dealers than it is with more established dealers, then there should be a positive and significant correlation between relationalism and communications patterns. In other words, the level of relationalism that is cultivated between new dealers and their suppliers should moderate the patterns of communication between the parties. The third hypothesis is:

[H.sub.3]: There is a positive and significant

correlation between relationalism and

communications patterns (frequency,

bidirectionality, formality and indirectness)

among both new dealers and established

dealers in India.

Impact on Satisfaction. Channel member satisfaction with vendor performance is closely related to interfirm relationalism. Anderson and Narus (1990) found that interfirm cooperation and communication correlated positively with channel member satisfaction. Relationalism defines characteristics of a dyadic channel relationship in terms of perceived solidarity, trust, commitment and so on. High levels of these relationship characteristics should correlate positively with a channel member's satisfaction with a trading partner. Thus, the final hypotheses we offer as an extension of the test by Anderson and Narus (1990) are:

[H.sub.4a]: New and established dealers' satisfaction

with their primary vendor is positively

and significantly correlated

with supplier-dealer communication

(viz., frequency, bidirectionality, formality

and indirectness).

[H.sub.4b]: New and established dealers' satisfaction

with their primary vendor is positively

and significantly correlated

with relationalism (viz., frequency,

bidirectionality, formality and indirectness).

RESEARCH DESIGN

Dealers of electric lamps and lighting equipment in urban markets in India were surveyed about their relationship and communications with their major (largest) supplier. Questionnaires were personally delivered to and collected from dealers who agreed to participate in the study by a group of graduate student interns in India. Of 252 lighting dealers in India who were contacted (all members of at least one of several light dealers' associations in that country), 92 returned usable responses which resulted in a response rate of 36.5 percent.

Our goal for this paper is to compare and contrast new dealers with more established dealers. This requires that we define what a new dealer is. Indian dealers who were in business for five years or less were classified as new and dealers who were in business for more than five years were classified as established. Of the 92 dealers who returned usable questionnaires, 38 were placed in the "new" category and 54 were placed in the "established" category.

The logic for classifying dealers as new or established on the five year criterion was not completely arbitrary. Most importantly, the second author worked for an international manufacturer in the lamp and lighting industry in India and that supplier afforded special concessions to and had separate marketing support programs for dealers with less than five years of experience. This personal experience suggested this breakdown from the outset. Further, the frequency distribution of all respondents by number of years in business revealed a peak frequency of five years (11.8 percent). Noting this, we conducted l-tests on key variables between the new and established dealer groups so defined and discovered a high incidence of significant differences. Finally, logic such as this has supported age year classifications in other's research (Ursic and Czinkota, 1984).

The data in Table 2 summarize the demographic characteristics of the two sets of dealers examined here. The "new dealer" group has been in business on average for only 3.1 years. The "established" dealers have been in business for nearly 16 years. The new dealers have much smaller businesses and employ a smaller workforce. The respondents in this group are mostly owners and general managers who are responsible for overseeing all aspects of their dealership. They also are responsible for a larger share of all communications with their major supplier. In the new dealer group, only 1.68 persons on average communicate directly with their major supplier. In the established dealer group, 2.33 persons are involved.

Table 2
Descriptive Characteristics of the Two Sample Groups

                                                    Established
                                    Dealers         Dealers
Measure                             (N = 38)        (N = 54)

Number of Years in Business          3.11 (1.49)    15.95 (11.62)
Number of People Employed            4.57 (3.24)     7.17 (4.69)
Number of People Involved in
  Communication                      1.68 (1.49)     2.33 (11.62)
Personal Contacts Made per Month     8.16 (3.24)     5.87 (4.69)
Phone Contacts made per Month        9.82 (0.81)    11.13 (1.35)
Mail Contacts Made per Month         2.32 (7.61)     1.43 (4-40)
Respondent Position
  Owner                             34    (89.5%)   52    (96.3%)
  General Manager                    4    (10.5%)    2    (3.7%)
Type of Business
  Retail Only                        1    (2.6%)    10    (18.5%)
  Wholesale Only                    10    (26.3%)   15    (27.8%)
  Mixed                             27    (71.1%)   29    (53.7%)

Standard deviations are shown in parentheses for years in business through mail contacts per month. Percentages of all responses are shown in parentheses for the two categorical variables.

CONSTRUCT MEASUREMENT

We adopted a direction of communication measure from Roberts and O'Reilly (1974) and Mohr (1989). The dealers were asked to indicate what percentage of their communication or contact with their largest supplier was initiated by the dealership. The degree of bidirectionality in the communication was computed by transforming the direction measure following the procedure developed by Mohr (1989). If the percent of communication initiated by the dealer (D) was greater than 50 percent, the bidirectional measure was calculated as: ((100 - D) X 2). If the communication direction measure was less than or equal to 50 percent, then the bidirectional measure was calculated as: D X 2. This resulted in a scale with a feasible range from 0 to 100. For example, if the dealer initiated 50 percent of the communication, then the bidirectional score is 100. If the dealer initiated only 20 percent, the score is 40. If the dealer initiated 60 percent, the score is 80. Scores near 0 indicate unidirectional communication and scores near 100 indicate bidirectional communication.

The measure of communication frequency was adapted from Mohr (1989). Respondents reported the frequency of use of different communication methods (face-to-face with salespeople, telephone, fax, sales training, written letters, trade shows, dealer meetings, and seminars) when the supplier communicated with their dealership and also when the dealer communicated with the supplier. The average scores of the two sets of items (all 5-point Likert scales) were measures of frequencies of each supplier's and the dealer's communication, respectively. The frequency of all the communication was given by the sum of each supplier's and dealer's communication. Thus, scores in the total frequency measure ranged from 2 (very low frequency) to 10 (very high frequency).

To measure the formality of supplier-dealer communication, a four-item Likert scale (one item from the original five-item scale was deleted) was adapted from Mohr (1989), Anderson, Lodish, and Weitz (1987), and Ruckert and Walker (1987). The items measured the extent to which information flows through routinized and structured modes. The average score of the four items was a measure of the communication formality with 1 representing very informal communication and 5 representing very formal communication.

The frequency of communication through personal contact was measured by averaging the frequencies of communication through face-to-face contact with salespeople, sales training, trade shows, dealer meetings, and seminars for both suppliers communicating with dealers and dealers communicating with suppliers. All of these methods represent some form of personal face-to-face communication. The average score was an absolute measure of communication through personal contact, with 1 representing a very low frequency and 5 representing a very high frequency. The measure of relative use of personal contact was obtained by dividing the absolute measure score by the score representing the frequencies of use of all methods of communication, personal as well as telephone and mail.

To measure communication content, the multi-item influence strategy scale developed by Boyle, Dwyer, and Oh (1990) was used. The respondents were asked on a 5-point Likert scale (1 = never, 5 = always) how often the supplier's representative uses each of the six influence strategies in their interaction with the dealer. The influence strategies measured were information exchange (average of 4 items), recommendations (average of 4 items), requests (average of 4 items of which one was later deleted), promises (average of 6 items of which two were deleted), threats (average of 5 items), and legal pleas (average of 5 items). The amount of absolute direct influence strategy use was computed as the average of the request, promise, threat, and legal plea scales. The amount of absolute indirect influence strategy use was computed as the average of the information exchange and recommendation scales. The relative use of direct influence strategy use was computed as a percentage of all (direct and indirect) influence strategy use.

To measure the extent of relationalism present in the dealer-supplier relationship, respondents were asked to indicate on a five point Likert scale how much they agreed (1 = strongly agree, 5 = strongly disagree) with each item statement describing a dimension of their relationship with their primary supplier. Six dimensions of relationalism were evaluated.

(1) Solidarity--a measure of the relative importance of arms-length bargaining and enforcement versus trust in the preservation of the relationship (3 items).

(2) Mutuality--extent to which the parties monitor each individual transaction or the relationship as a whole to 'assess performance (3 items).

(3) Role Integrity--contrast between the simple roles of parties in discrete exchanges versus the complex roles in relational exchanges (3 items).

(4) Flexibility--extent to which governing rules may be adjusted to accommodate each party's interests (3 items).

(5) Duration of Relationship--extent to which parties believe that their relationship will continue into the future (3 items).

(6) Trust--degree of mutual trust between the parties (18 items).

The first five scales (solidarity through duration of relationship) were adapted from Kaufmann and Stern (1988) and Boyle et al. (1992). The trust scale was adapted from Rempel, Holmes and Zanna (1985). The number of items that were retained for the analyses after scale purification are reported in Table 3.

Table 3
Psychometric Properties of Multi-Item Measurement Scales

                                      Reliability
                          Number of   (Coefficient
Measure                   Items(a)       Alpha)

Formality

Communication Formality      4          .7087

Influence Strategy Use

Indirect Strategies          8          .7729
Direct Strategies           17          .9214

Relationalism

Solidarity                   3          .6317
Mutuality                    3          .6624
Flexibility                  2          .5599
Role Integrity               2          .5713
Duration                     3          .8266
Trust                       11          .8743

Satisfaction

Satisfaction with
Major Supplier              10          .8744

                          Range of
Measure                   Values

Formality

                          1 = most informal
Communication Formality   5 = most formal

Influence Strategy Use

Indirect Strategies       1 = minimal use
Direct Strategies         5 = maximal use

Relationalism

Solidarity                1 = minimum
Mutuality                 5 = maximum
Flexibility
Role Integrity
Duration
Trust

Satisfaction

Satisfaction with         1 = dissatisfied
Major Supplier            5 = satisfied

(a) Number of items retained after measure purification procedures.

Dealer satisfaction was measured by thirteen 5-point Likert scales for various aspects of the supplier's performance. Items dealing with satisfaction with the personal dealings with the supplier's representatives, inventory management assistance, cooperative advertising, promotions, sales training, product quality and so on were included. After scale purification procedures, ten satisfaction items were retained.

A measure purification process based upon factor analyses and calculations of Cronbach's alpha was employed to maximize reliability for all multi-item scales. The psychometric properties and number of scale items retained for the various construct measures are summarized in Table 3.

TESTS OF HYPOTHESES

Communications. Hypotheses 1a through 1e were tested by a multivariate analysis of variance (MANOVA) test. The dependent variables were examined for significant differences between the two sample groups of dealers. Group means and standard deviations for each dependent variable and levels of statistical significance are reported in Table 4. The Hotelling statistic was significant at the 0.05 level indicating that differences between the new and established dealers were significant for the model consisting of the various aspects of supplier-dealer communication. This finding supports the first hypothesis. Supplier-dealer communications patterns differ between new and established dealers.

Table 4
Communication Patterns for New and Established Dealers: Mean Values

                      New             Established
Measure               Dealers         Dealers
                      (N = 38)        (N = 54)

Bidirectionality of
Communication         44.32 (27.89)   52.78 (28.11)

Frequency of
Communication          4.75 (1.07)     4.28 (1.00)

Formality of
Communication          3.89 (0.76)     3.77 (0.90)

Absolute Use of
Personal Contact       2.31 (0.67)     2.02 (0.58)

Relative Use of
Personal Contact       0.27 (0.05)     0.26 (0.07)

Absolute Use of
Direct Influence       2.45 (0.77)     2.10 (0.65)

Relative Use of
Direct Influence       0.42 (0.09)     0.38 (0.07)

Measure               Univariate    Significance
                      F-Statistic       of F

Bidirectionality of
Communication            2.03         0.157

Frequency of
Communication            4.50         0.037

Formality of
Communication            0.44         0.507

Absolute Use of
Personal Contact         4.82         0.031

Relative Use of
Personal Contact         0.08         0.780

Absolute Use of
Direct Influence         5.40         0.022

Relative Use of
Direct Influence         5.82         0.018

For the entire MANOVA model, Hotelling's Exact F = 2.45, Significance p = 0.02.

Standard deviations are shown in parentheses. Statistical differences at p [is less than] .05 are shown in bold print.

Individual effects are assessed with the univariate ANOVA test results. Hypothesis 1a was tested by comparing the level of communication bidirectionality in the two samples. As shown in Table 4, the bidirectional communication scores for new and established dealers are 44.32 and 52.78, respectively. Though the mean level of bidirectional communication is higher for established dealers, the difference between the two scores is not statistically significant. Thus, this part of the first hypothesis is not supported.

To test Hypothesis 1b, the total communication frequency measures from the two samples were compared. The mean value for the new dealers (4.75) is significantly higher than that for the established dealers (4.28). The univariate F-statistic (4.50) is significant at the 0.05 level. Thus, Hypothesis 1b is supported.

Hypothesis 1c was tested by comparing the measures of communication formality from the two samples. Lower values indicate the use of more informal modes. We expected but did not find greater informality in communications among the new dealers. The mean formality score for new dealers is slightly higher than it is for established dealers (3.89 and 3.77). However, the difference in the mean scores in Table 4 for communication formality between new and established dealers is not statistically significant and Hypothesis 1c is not supported by the data.

To test the use of personal contact in communication (hypothesis 1d), the two variables measuring the absolute and relative use of personal contact were compared across the new and established dealer groups. The absolute use of personal contacts is higher for the new dealers (mean score 2.31 versus 2.02 for the established dealers, with a univariate F-statistic of 4.82, significant at the 0.05 level), as hypothesized. However, there is no significant difference between the two dealer groups for the relative use of personal contacts, that is, the percentage of all communications that were personal. Hypothesis 1d is only partially supported.

Hypothesis 1e was tested by comparing the variables measuring the absolute use and relative use of direct influence strategies for the two sample groups. Direct influence strategy use was significantly higher, both absolutely and relatively, for the new dealers. The mean scores for absolute and relative use of direct influence strategies are 2.45 and 42.1 percent, respectively, for the new dealers, and 2.10 and 38.1 percent, respectively, for the established dealers. The univariate F-statistics for the two variables are 5.44 and 5.82, respectively, both being significant at the 0.05 level. Hypothesis 1e is supported by the data.

In summary, mixed support was found for the hypothesis that suppliers' communication patterns with new dealers differ significantly from those with older, more established dealers. The global MANOVA test supports the hypothesis that communication patterns differ between the two groups. Communication with new dealers is more frequent and influence attempts are more direct. Modest support was found for the thesis that communication with new dealers is more personal. Communications were found to be neither more unidirectional nor informal among new dealers in contrast to more established dealers as hypothesized.

Relationalism. The second hypothesis is that channel relationalism is greater for established dealers than it is for new dealers. Results of a MANOVA with the six facets of relationalism as dependent variables to test this are reported in Table 5. The Hotelling statistic was not significant indicating that there is no overall difference in relationalism between new and more established dealer in India. A separate t-test for differences between a composite measure of relationalism (average of the six facets) also failed to reveal any significant difference. (The second t-test was conducted separately because inclusion of the composite relationalism score in the MANOVA would have corrupted that test.)

Table 5
Relationalism for New and Established Dealers

                     Mean Values
                 New             Established
Measure          Dealers         Dealers
                 (N = 38)        (N = 54)

Solidarity       4.27 (0.64)     4.47 (0.55)

Mutuality        3.91 (0.72)     4.07 (0.80)

Flexibility      3.59 (0.88)     3.53 (0.93)

Role Integrity   3.24 (1.02)     3.14 (1.04)

Duration         4.15 (0.78)     4.35 (0.61)

Trust            3.40 (0.84)     3.40 (0.78)

Composite
Relationalism    0.76 (0.51)     3.83 (0.45)

                 ANOVA Test Results
Measure          Univariate    Significance
                 F-Statistic      of F

Solidarity        2.42           0.124

Mutuality         0.88           0.350

Flexibility       0.09           0.760

Role Integrity    0.22           0.641

Duration          1.65           0.202

Trust             0.00           0.978

Composite
Relationalism    -0.61           0.550

For the entire MANOVA model, Hotelling's Exact F = 0.729, Significance = 0.627.

Standard deviations are enclosed in parentheses.

All of the F statistics are not significant. The Composite Relationalism variable was not included in the MANOVA test. Results of a separate ANOVA test are included here. Since Levine's Test for Equality of Variances showed F = .722, unequal variances in the two groups were assumed for the latter analysis.

The sample sizes are slightly smaller than those used in the tests of hypotheses 1a-1e due to missing data for some of the relationalism scales.

As these tests failed to support the hypothesized relationships, a final analysis was conducted. Rather than divide the sample into two groups, simple correlations of the six facets of relationalism and the composite measure with the number of years the dealers had been in business were calculated (see Table 6). Only the correlation coefficient for role integrity (r = -.201, p = .026) is significant. That this is negatively correlated with years in business suggests that the dealers perceive that the integrity of the roles they and their supplier play actually diminishes over time. In other words, role ambiguity and conflict typically grows over time. This is due, perhaps, to changing dealer and/or supplier expectations after both become more experienced with each other.

Table 6
Correlation of Years in Business with Relationalism

                          Pearson
Relationalism Measures    Correlation   Significance
                          Coefficient   (One Tailed Test)

Solidarity                -.005         .479

Mutuality                 -.013         .452

Flexibility                .092         .191

Role Integrity            -.201         .026

Duration                   .152         .075

Trust                      .104         .168

Relationalism Composite   -.033         .383

Coefficient in bold print is statistically significant.

Relationalism and Communication. The third hypothesized relationship is that supplier communication is affected directly by relationalism. Our logic here is that relationalism neutralizes the natural differences in suppliers' communication patterns between new and more established dealers. To test this, Pearson correlation coefficients were computed and are reported in Table 7. There is some support for this hypothesis. Among the new dealers, relationalism is positively and significantly correlated with communication bidirectionality (r = .349), frequency (r = .362), formality (r = .419) and the relative incidence of indirect communication strategies by the supplier (r = .795). Relationalism is not correlated with the absolute use of indirect communication strategies, however. Among the more established dealers, bidirectionality and absolute use of indirect influence of communications are not correlated with relationalism. Frequency (r = .367), formality (r = .279) and relative incidence of indirect communication (r = .520) are positively associated with the composite relationalism measure.

Table 7 Correlation of the Composite Relationalism Measure with Communication Variables for

New and Established Dealers

                                            New Dealers

Communication Variables                    r      p [is less
                                                  than or
                                                  equal to]

Bidirectionality of Communication          .349   .050
Total Communication Frequency              .362   .010
Degree of Communication Formality          .419   .010
Indirect Influence Strategy Use            .309   N.S.
Relative Indirect Influence Strategy Use   .795   .001

                                                  Established
                                                    Dealers

Communication Variables                    r      p [is less
                                                  than or
                                                  equal to]

Bidirectionality of Communication          .119      N.S.
Total Communication Frequency              .367      .010
Degree of Communication Formality          .279      .050
Indirect Influence Strategy Use            .261      N.S.
Relative Indirect Influence Strategy Use   .520      .001

Coefficient shown in bold print are significant at p [is less than] .05. Actual significant levels are

NS = Not Significant.

In summary, the third hypothesis that dyed relationalism would affect the patterns of interfirm communications among new and more established dealers received mixed support. The hypothesis held up somewhat better among the new dealers but there is some support for this among the established dealers also. The only difference between the two dealer groups is that relationalism is positively associated with the bidirectionality of communications among the new but not among the established dealers. Though the statistical support is mixed for this hypothesis, we feel that this evidence is important. Further investigation of the phenomenon is warranted to clarify this issue.

Satisfaction. The fourth and final hypothesis is that dealer satisfaction is positively correlated with both communications patterns and relationalism. At-test for mean difference on composite satisfaction scores for the new and the established dealers indicated that the new dealers are significantly more satisfied with their major supplier. The mean new dealer satisfaction score is 4.13 and the established dealer score is 3.80 on five point scales from 1 = not satisfied to 5 = very satisfied. (t-statistic = 2.09, p = .04). Though not hypothesized, this suggests an interesting phenomenon. Does the satisfaction of all dealers with their major supplier decline over time? Is this phenomenon generalizable across cultures?

The fourth hypothesis was tested using correlation coefficients which are reported in Table 8. Since the various communication variables and relationalism measures are intercorrelated within each set, partial correlations rather than simple zero-order correlation coefficients were calculated. Pearson correlation coefficients are reported for the composite relationalism measures.

Table 8 Correlation of Communication Patterns and Relationalism with Satisfaction for New and Established Dealers

Communication and         New Dealers   Established Dealers
Relationalism Variables     (n=38)           (n=54)

1. Communication

Bidirectionality          -.428(b)           .320(b)

Frequency                  .232              .151

Formality                  .342              .470(a)

Direct Influence
  Strategy Use            -.052             -.398(a)

Indirect Influence
  Strategy Use            -.019              .577(a)

2. Relationalism

Solidarity                -.045              .389(a)
Mutuality                  .341              .315(b)
Flexibility                .770(a)           .082
Role Integrity            -.278             -.363(b)
Duration                  -.313             -.081
Trust                     -.079              .278
Relationalism Composite    .244              .496(a)

Communication and         All Dealers
Relationalism Variables     (n=92)

1. Communication

Bidirectionality            .419(a)

Frequency                   .289(b)

Formality                   .340(a)

Direct Influence
  Strategy Use              .169

Indirect Influence
  Strategy Use              .348(a)

2. Relationalism

Solidarity                  .156
Mutuality                   .260(b)
Flexibility                 .317(a)
Role Integrity              .160
Duration                   -.189
Trust                       .172
Relationalism Composite     .372(a)

All coefficients except those for the "Relationalism Composite" measure are partial correlation coefficients with satisfaction as the dependent variable. For the composite relationalism measure, Pearson's correlation coefficients are shown.

The t-values for the partial correlation and Pearson's correlation coefficients shown in bold print are significant at (a) p [is less than] .01, (b) p [is less than] .05.

The results of these tests are among the most intriguing of this study. First, among all dealers (n = 92), communication frequency, formality, and indirect influence strategy use are positively and significantly associated with dealer satisfaction. The use of direct influence attempts (requests, promises, threats and legal please is not significantly correlated with satisfaction (r = -.169 but p [is greater than] .05). The negative association between satisfaction and bidirectional communication from the supplier to the dealer (r = -.419, p [is less than] .01) indicates that Indian dealers prefer to receive lots of communication from their suppliers but they prefer not to be expected to respond or provide unsolicited information. This indicates strong support for hypothesis 4a.

There are significant differences, between new and established dealers. Among the new dealers, only communication bidirectionality is correlated with dealer satisfaction and it is inversely related (r = -.428). Total frequency of communication, communication formality and influence strategies used are not associated with dealer satisfaction. This means that suppliers who communicate frequently and expect or demand little communication back from new dealers are likely to have more satisfied dealers.

Among the established dealers, satisfaction is positively correlated with both formality (r = .470) and the level of indirect communications (r = .577), but is negatively correlated with bidirectionality (r = -.320) and direct influence strategy use (r = -.398). New dealers seem unaffected by the use of either direct or indirect influence strategies. However, the negative associations with bidirectionality and direct communication patterns suggest that more established Indian businesspeople may come to resent a supplier's use of requests, promises, threats, and legal pleas more than do the new dealers. Established dealers are more satisfied when their suppliers communicate more formally and rely more upon indirect communication strategies--simple exchanges of information and recommendations. Like the new dealers, however, even the established Indian dealers are more satisfied with suppliers who communicate often to them but expect or require little response.

Differences were found also between new and established dealers when we examined the correlations between dealer satisfaction and channel relationalism (hypothesis 4b). First, the analyses of association for all respondents (n = 92) indicate that the composite relationalism measure (r = .372) and two of that measure's components, mutuality (r = .260) and flexibility (r = .317), are positively associated with dealer satisfaction. However, separate analyses for the two sets of dealers, new and established, reveals additional insights. This indicates moderate support for hypothesis.

Among new dealers, the composite relationalism measure is not associated with dealer satisfaction. Only one of the six facets of relationalism, flexibility (r = .770), correlated with new dealer satisfaction. Among established dealers, three of the relationalism facets and the composite score of relationalism are positively and significantly related to dealer satisfaction. Flexibility is not associated with established dealer satisfaction (r = .082 and p [is greater than] .05). Neither are expectations that the relationship will endure (r = -.081, p = -.189) or trust (r = .278, p = .172). However, solidarity (r = .389) and mutuality (r = .315) are positively correlated. For established dealers, role integrity is negatively associated with their satisfaction. In other words, the established dealers seem happier with more simple and well defined role assignments and performance. They do not associate complex relational expectations with satisfaction. Recall that the earlier discussion of Hypothesis 2 revealed a negative correlation between role integrity and years in business. The evidence here suggests further that among established dealers, role integrity is not associated with satisfaction.

DISCUSSION AND MANAGERIAL CONCLUSIONS

This empirical study illustrates notable differences between supplier communication patterns with new and more established dealers in India. Communication between suppliers and new dealers is more frequent than that between suppliers and established dealers. Personal face-to-face contacts are used more often in communication between suppliers and new dealers than between suppliers and old dealers. However, there is no difference in the relative use of personal contacts (with respect to the use of phone and mall) for the two groups of dealers. Therefore, the higher frequency of communication by face-to-face contact for new dealers may be only a reflection of the overall higher frequency of communication for this group. Suppliers in India use direct influence strategies more often with new dealers than with old, more established dealers (both absolutely and relative to indirect influence strategies).

The unexpected finding that relationalism is not greater among established dealers than among new ones might be based upon characteristics of the Indian culture. Specifically, relationalism may not develop over time as one expects in the west. We hypothesized that initial business dyads would exhibit modest levels of relationalism as parties would initially be cautious in dealing with a new partner. We expected relationalism to build with experience. India is a high-context culture in which communication depends heavily upon nonverbal aspects. Words in such cultures are not to be taken at face value; rather, meaning must be inferred. This meaning is normally tied to the immediate context or situation. In India, a high degree of trust, solidarity and mutually may be required to initiate any viable business relationship. It may be that no business relationship can be established in India without high levels of these relational attributes. Without these, communication and trade may be impossible. Whereas in the west, relationalism initially may be low and build over time, in India it may have to be high from the outset of a business relationship. In retrospect, we must acknowledge the influence of the traditional Indian culture upon the new dealers. This finding suggests that new suppliers to the Indian market should not expect relationalism to grow over time in dealings with Indian business people. Rather, the goal should be to establish high levels of relationalism at the inception of the relationship and to strive to maintain those levels.

Though the findings are somewhat mixed, we did find moderate support for the argument that channel relationalism moderates communication patterns among new and established dealers. Because the findings are mixed, however, we acknowledge the need for and encourage further investigation of this issue. As marketers around the globe strive to practice relationship marketing, knowledge about the link between communication patterns and relationship development and maintenance will grow in importance.

Dealer satisfaction with major suppliers was found to be strongly related to the channel communications patterns. For both new and established dealers, satisfaction is inversely related to communication bidirectionaliy. New dealer satisfaction is strongly associated with only one facet of relationalism--flexibility. The satisfaction of older, more established dealers is subject to a greater variety of relational attributes.

This suggests that a supplier to new dealers should strive to be flexible and able to deal effectively with the problems that those dealers are sure to encounter as they are developing their business. This analysis reveals that all other aspects of relationalism, which tend to be associated with longer-term issues and more lofy ideals, are not correlated with new dealer satisfaction. This is not surprising upon reflection as our study has revealed that there is no link between relationalism and duration of relationships in India. As noted above, high levels of relationalism seem to be indicated as requirements for the establishment of any relationship.

Established dealers in India have a different set of expectations. Once relationships are established, the importance of flexibility to dealer satisfaction seems to diminish. Over time, trading partners likely come to know what is expected of each other and this experience diminishes the importance of supplier flexibility. High solidarity reflects a dealer's belief that the relationship with the supplier is not a simple arms-length bargaining situation. Rather, it is based upon trust and is likely to endure beyond transaction-specific breakdowns. Simply, operating problems are expected and should not threaten the relationship. Mutuality is evidenced by a dealer's focus on the overall relationship and not individual transactions to assess performance. Despite the finding that relationalism does not differ between new and established dealers, it is suggested here that established dealers' satisfaction is directly associated with perceptions of solidarity and mutually which can be tested or strained over time in a dynamic business relationship.

This paper offers new information about working with trading partners in India. Two limitations of the study should be acknowledged. Most importantly, we studied only dealers in the electric lamp and lighting industry in urban markets and focused on their relationship with their largest supplier. Thus, these findings are not necessarily generalizable to other industries, nonurban areas and smaller suppliers. Second, our sample is relatively modest, only 92 dealers responded despite a solid research effort. However, the response rate of 36.5 percent is quite good for studies such as this and exceeded our expectations given the Indian culture and the oddly of conducting survey research in that country. Our analyses of the responses of early and late respondents revealed no differences which might suggest serious nonresponse problems.

Western investors seeking to cultivate relationships with Indian distributors should infer three strategic recommendations from these findings. First, communicate directly and often with new dealers. Do not be surprised if newly recruited dealers do not engage in a high level of bidirectional communication. When told exactly what is expected, a satisfied new dealer is not likely to engage in a bidirectional dialogue. Explanation or justification for a supplier's request or directive is generally not needed. Rather, a new Indian dealer is more likely to simply implement the recommended or directed action (s) without debate or hesitation.

Second, new suppliers to the Indian market should at all times remember that rapid and traumatic change is occurring in India's economy. As foreign investors flood the area to exploit emerging market opportunities, Indian businesspeople are being inundated with a great variety of trading partners' expectations. Much learning and adjusting to new patterns of business relationships are underway. A supplier who is new in this market should communicate early and often and not interpret modest Indian communication responses negatively.

Third, anticipate the need to be very flexible with new Indian dealers. Indian new dealers' satisfaction is closely associated with their perception that their suppliers are willing to adjust to unforeseen circumstances. Newly recruited channel members are learning and adjusting to new and sometimes turbulent market conditions. Existing and newly emerging dealers and distributors are creating a new distributive infrastructure in India. They are being called upon to make many difficult adjustments. They expect their suppliers to respond to their need for flexibility.

With more experience, established dealers do not seem to require as much flexibility. They do expect, however, clear and precise role definitions and performance. Simply, they want to be told what they should do and what they should expect the supplier to do. And, they want to be able to depend on those assertions. They prefer simple, not complex, business relationships with foreign trading partners. Keep things simple! This is consistent especially with established dealers' lack of concern about supplier flexibility. If role expectations are met consistently, then flexibility is not a major issue. With experience, Indian dealers come to expect higher levels of supplierdealer relationalism in the form of solidarity and mutuality. In other words, they expect to build relationships which evidence a true partnership with their major suppliers. They seem to favor close alignments and, to some extent, want to identify with their major suppliers.

In conclusion, we caution that every Western firm will need to overcome a variety of obstacles to successfully expand in any foreign market. In an environment as culturally diverse as that in India, the problems associated with identifying and responding appropriately to those obstacles are more difficult than most. The development and successful implementation of a sound communication program, especially toward new, entrepreneurial dealers with high potential for growth, is an essential component of a market entry strategy in a developing country such as India. Knowledge of the different communication patterns used by current suppliers with new and established dealers in India and how those are related to dealer satisfaction should help Western firms to formulate effective interfirm communication strategies when they expand into that lucrative market.

References

Achrol, Ravi. 1991. "Evolution of the Marketing Organization: New Forms for Turbulent Environments." Journal of Marketing 55 (October): 71-93.

Anderson, James C., and James A. Narus. 1990. "A Model of Distributor Firm and Manufacturer Firm Working Partnerships." Journal of Marketing 54 January): 42-58.

Anderson, E., L. Lodish, and B. Weitz. 1987. "Resource Allocation Behavior in Conventional Channels." Journal of Marketing Research 24 (February): 85-97.

Arndt, Johan. 1979. "Toward a Concept of Domesticated Markets." Journal of Marketing 43 (Fall): 44-54.

Bello, Daniel C., and Naveen Donthu. 1993. "Exchange Relationships in Nonintegrated International Marketing." In Enhancing Knowledge Development in Marketing. Eds. David Cravens and Peter R Dickson. Proceedings of the 1993 AMA Summer Educators' Conference. Chicago: American Marketing Association.

Bensky, J.M., M. Benefiel, V. D'Souza, and K. Saha. 1993. India Fiscal Year 1994 Country Marketing Plan. New Delhi, India: U.S. and Foreign Commercial Service, American Embassy.

Boyle, Brett, Robert F. Dwyer, and Sejo Oh. 1990. "Measuring Interfirm Influence in Franchise Channels of Distribution." Society of Franchising Conference Proceedings. Scottsdale, AZ.

--, --, Robert A. Robicheaux, and James T. Simpson. 1992. "Influence Strategies in Marketing Channels: Measures and Use in Different Relationship Structures." Journal of Marketing [Research 29 (November): 462-71.

Czinkota, Michael R., and Ilkka A. Ronkainen. 1993. International Business. Fort Worth, TX: The Dryden Press.

Dwyer, Robert F., Paul H. Schurr, and Sejo Oh. 1987. "Developing Buyer-Seller Relationships." Journal of Marketing 51 (April): 11-27.

The Economist. 1994. "Blinking in the Sunlight: India's Businesses." London, United Kingdom: The Economist Newspaper Ltd. (April 9): 72-74.

Frazier, Gary L., James D. Gill, and Sudhir H. Kale. 1989. "Dealer Dependence Levels and Reciprocal Actions in a Channel of Distribution in a Developing Country." Journal of Marketing 53 January): 50-69.

--, and John O. Summers. 1984. "Interfirm Influence Strategies and Their Application in Distribution Channels. "Journal of Marketing 48 (Summer): 43-55.

Harris, Phillip R., and Robert T. Moran. 1987. Managing Cultural Differences. Houston: Gulf Publishing Co.

Jaworski, Bernard J. 1988. "Toward a Theory of Marketing Control: Environmental Context, Control Types, and Consequences." Journal of Marketing 52 (July): 23-39

Kale, Sudhir H. 1986. "Dealer Perceptions of Manufacturer Power and Influence Strategies in a Developing Country." Journal of Marketing Research 23 (November): 387-393.

--. 1993. "Dealer Dependence and Influence Strategies in a Manufacturer-Dealer Dyad." Journal of Applied Psychology 74 (3): 379-84.

Kaufmann, Patrick, and Rajiv Dant. 1992. "The Dimensions of Commercial Exchange." Marketing Letters 3 (May): 175-85.

--, and Louis Stern. 1988. "Relational Exchange Norms: Perceptions of Unfairness and Retained Hostility in Commercial Litigation." Journal of conflict Resolution 32 (September): 534-52.

Kotler, Philip, and Gary Armstrong. 1994. Principles of Marketing. Englewood Cliffs, NJ: Prentice Hall, Inc.

Macneil, Ian R. 1980. The New Social Contract: An Inquiry into Modern Contractual Relations. New Haven, CT: Yale University Press.

Mohr, J. 1989. "A Conceptual Model of Communication for Marketing Channels with an Empirical Test." Dissertation. The University of Wisconsin-Madison.

--, and J.R. Nevin. 1990. "Communication Strategies in Marketing Channels: A Theoretical Perspective." Journal of Marketing 54 (October): 36-51.

Moore, Martha T. 1993. "Fountain of Growth Found Abroad" USA Today (August 18): B1.

Morgan, Robert M., and Shelby D. Hunt. 1994. "The Commitment-Trust Theory of Relationship Marketing." Journal of Marketing 58 July): 20-38.

Rempel, J., J. Holmes, and M. Zanna. 1985. "Trust in Close Relationships."Journal of Personality, and Social Psychology' 49 (1): 95-112.

Roberts, K, and C. O'Reilly. 1974. "Failures in Upward Communications in Organizations: Three Possible Culprits." Academy of Management Journal 17 (June): 205-215

Ruekert, R., and O. Walker. 1987. "Marketing's Interaction With Other Functional Units: A Conceptual Framework and Empirical Evidence." Journal of Marketing 51 (January): 1-19.

Scanzoni, J. 1979. "Social Exchange and Behavioral Independence." In Social Exchange in Developing Relationships. Eds. R.L. Burgess and T.L. Huston. New York: Academic Press, Inc.

Terpstra, Vern, and Kenneth David. 1991. The Cultural Environment of International Business. Cincinnati, OH: South Western.

Ursic, Michael L., and Michael R. Czinkota. 1984. "An Experience Curve Explanation of Export Expansion." Journal of Business Research 12 (Spring): 159-168