Submission Identification Number: 31459

Improving the Export Performance of U.S. Wineries

Richard Castaldi
Department of Management
San Francisco State University
1600 Holloway Avenue
San Francisco, CA 94132
Phone: 415 338-2829
Fax: 415 338-0501
Email: castaldi@sfsu.edu

Sanjit Sengupta
Department of Marketing
San Francisco State University
1600 Holloway Avenue
San Francisco, CA 94132
Phone: 415 338-1016
Fax: 415 338-1016
Email: sengupta@sfsu.edu

Murray Silverman
Department of Management
San Francisco State University
1600 Holloway Avenue
San Francisco, CA 94132
Phone: 415 338-7489
Fax: 415 338-0501
Email: msilver@sfsu.edu

August 2001

International Management Division, Academy of Management Conference,
Washington DC

Submitted to the Journal of World Business, July 2001.

The authors gratefully acknowledge a Business and International Education (BIE) grant from the U.S Department of Education and a matching grant from the College of Business at San Francisco State University in support of this research.

Twin forces of globalization and consolidation are confronting the wine industry. While “old world” producers like France continue to dominate, “new world” producers like Australia are rapidly making inroads into the increasingly competitive global wine market. It is timely to study US wineries’ export performance and draw implications for international competitive strategy.

We use an established three-dimensional conceptualization of export performance encompassing financial export performance, strategic export performance and satisfaction with exports. Drawing from management and marketing literature, we identify product-market and corporate strategy variables that should be related to export performance. Mail survey data from a sample of 133 wineries in the western US is analyzed using OLS regression.

Contrary to expectations, we find that three of the product-market strategy variables, brand image, product quality, and price-value, have no impact on export performance. However, intermediary performance and export commitment have a strong, positive relationship with export performance.

We further investigate the moderating effects of two moderator variables, price-quality segment and the type of intermediary relationship, on the relationships between intermediary performance, export commitment and export performance. We find that in the economy price-quality segment, export commitment has a greater impact on export performance than intermediary performance. However, in the luxury segment, intermediary performance has a greater impact on export performance than export commitment. With regard to type of intermediary, export commitment has a greater impact on export performance than intermediary performance when a wholesaler is used. Contrarily, intermediary performance has a greater impact on export performance than export commitment when an agent is used.

Keywords: International; Export; Performance.

The US domestic wine market has grown from a retail value of $12.2 billion in 1995 to $18.2 billion in 1999. US wine exports have grown consistently from a base of $137 million in 1990 to $548 million in 1999 (Wine Institute 2000). While this export growth is impressive, US wineries also face increasing threats to their domestic market share due to globalization in the wine industry.

Until a few years ago the US wine market was largely a domestic industry, with some imports from France, Italy and Spain competing with US wineries. Recently, however, imports have risen to 20% of the US market share. This has been fueled by the tremendous inroads made by Australian, Argentinian and Chilean wines, in particular, into the US market. For example, during 1995-1999, Argentina increased the value of their exports to the US by 243% and Chile by 152% (Wine Institute 2000). Since 1995, the unfavorable balance of trade for wine in the US has increased by 78% (Wine Institute 2000).

The US wine industry is in the midst of a radical transformation, which will dramatically affect the landscape of the industry. A mature and saturated domestic market has led to globalization, industry consolidation (e.g. Foster's, an Australian beverage company recently acquired Beringer, a California winery), and more professional (versus family) management for many US wineries. The February 2000 naming of Lewis Platt as CEO of Kendall-Jackson winery (former Chairman and CEO of Hewlett Packard) provides an appropriate example.

In terms of international markets, wine is produced commercially in over 60 countries. While each wine producing country's domestic market consumes much of the wine they produce, 23% (by volume) is exported to international markets. The leading wine producers include the 'old world' wineries in France, Italy and Spain. These three countries are also the leading exporters. 'New world' producers, such as the US, Australia, Chile, Argentina and South Africa have been making both production and export inroads globally over the past few decades.

The US has only 4.2% (by volume) of the world export wine market, while producing 8% (by volume) of the wine produced in the world (Wine Institute 2000). Furthermore, the US wine industry exports only 13% of the wine it produces, while other countries have more intensely developed their export markets. For example, France, Italy and Spain all export more than 25% of the wine they produce, Australia exports over 40% and Chile over 80% of their production (Doering 1999). One might argue that these countries export more intensively because of the small size of their domestic market. While this may be true, US wineries run the risk of losing market share at home to those exporters who have been making inroads into the US market. A continued focus on the domestic market may place the US wine industry at a long-term disadvantage in developing the requisite skills for competing in the increasingly competitive global market place.

Tariffs and trade barriers currently play a pivotal role in obstructing US wineries' access to various country markets. As these barriers are reduced under the auspices of the World Trade Organization, greater export opportunities will open up. US wineries must be positioned competitively to exploit these new opportunities. Thus, for offensive and defensive reasons, it is timely and relevant to undertake a study of US wineries’ export performance, identify key success factors affecting export performance, and study their interrelationships to draw implications for international competitive strategy and management.

In this paper, we first review the existing literature to develop our conceptual framework and hypotheses regarding the export performance of US wineries. We then discuss the methodology of the current study, and present the results. Finally, we derive implications from our study on what the management of US wineries should do to improve their export performance in the current global environment.

Conceptual Framework and Hypotheses

The theoretical perspective adopted in this study has its roots in the structure-conduct-performance framework of industrial organization (Scherer and Ross 1990). It recognizes that organizations are dependent on their environment for resources (Pfeffer and Salancik 1978) and they manage this dependency by planning and implementing strategy. Export performance is the result of a strategic co-alignment between export strategy, organizational resources and capabilities, and the environment (Cavusgil and Zou 1994).

Export performance has been conceptualized in many different ways in the existing literature (Katsikeas, Leonidou and Morgan 2000; Zou, Taylor and Osland 1998). The most popular approach has been to focus on financial outcomes such as export sales, export sales growth, export profits or export intensity (export sales as a percentage of overall sales). A second approach has tried to capture strategic outcomes of exporting such as improved competitiveness, increased market share or stronger strategic position. A third conceptualization has included perceptual or attitudinal measures of satisfaction under the assumption that satisfaction is based on better performance. All three approaches have been combined into the three-factor Export Performance Scale (EXPERF) by Zou, Taylor and Ostland (1998) to measure the performance of a company’s export program rather than the performance in any specific product-market. In this study, we use the three dimensions of the EXPERF scale as dependent variables, financial export performance, strategic export performance and satisfaction with export program.

Export strategy has been found to play a central role in determining export performance (Cavusgil and Zou 1994). It includes all aspects of the marketing mix such as product, promotion, pricing and distribution (Aulakh, Kotabe and Teegen 2000). Preliminary interviews with US winery executives revealed that the role of promotion in wine exports is rather limited to participation in trade shows and free publicity in the trade press. Therefore, our study focuses on the product, price and distribution elements of export strategy and their impact on export performance. We investigate the impact of brand image and product differentiation, which have been identified as correlates of performance in the PIMS Project (Buzzell and Gale 1987). We also include the price-value variable (Thach and Axinn 1991) and the performance of the distribution channel intermediary (Bello and Gilliland 1997).

While many organizational characteristics have been examined in the export performance literature, a strong, recurrent theme that emerges is one of organizational commitment to exports. Many studies have concluded that in order to be successful in exports, organizations must commit adequate resources to the effort (Kirpalani and MacIntosh 1980; Cavusgil 1984; Axinn, Noordewier and Sinkula 1996). Therefore, we include export commitment to exports as an independent variable in our study.

Wine is a packaged consumer product marketed at a number of price points targeted at different quality segments. Preliminary interviews with winery executives revealed that five price-quality segments are commonly followed in the industry: economy (under $3 per bottle), popular ($3 - $7 per bottle), super premium ($7 - $14 per bottle), ultra premium ($14 - $25 per bottle), and luxury (over $25 per bottle). Clearly, the economics of wine production and marketing vary across these segments. It would seem logical, then, that the relationships between export strategy, organizational characteristics and export performance should also vary across price-quality segments. Since we found no studies in the existing literature looking at the impact of segmentation on export performance, we include this as a moderator variable in our framework.

The role of distribution channel performance in export markets cannot be underestimated. Cavusgil and Zou (1994) point out that export performance is enhanced when management provides strong support to foreign distributors. Bello and Gilliland (1997) state that firms choose foreign distributors carefully, ensuring that they have the necessary skills, are trustworthy, and have a cooperative attitude. US wineries have little choice with regard to export distribution channels. Such choice is constrained by availability based on historical first-mover advantage. Distributors are specialized by geographical markets or end-user segments such as retail chains, restaurants or cruise ships. Some are based in the US and some in the foreign country market. Export channels for wine seem to have developed idiosyncratically rather than following any evolutionary patterns.

The one area of overlap with export channels in other industries is that distribution intermediaries in the wine industry include agents and wholesalers. Agents do not take title to the goods. Their role is limited to promoting products in export markets. Wholesalers, on the other hand, do take title to the goods and resell product to other intermediaries such as retailers, or to end-user segments such as restaurants, hotels or cruise ships. Since the economics of wine distribution vary depending on the role of the distributor, we expect the relationships between export strategy, organizational characteristics and export performance should also vary by type of intermediary. Therefore, in this study, we treat the type of intermediary as a moderator variable in our framework.

Our conceptual framework is presented in Figure 1. Brand image is the set of positive perceptions that customers have about a specific winery’s offering. Clearly, the stronger the brand image, the greater the demand for the winery’s product, leading to greater export sales. Product differentiation is what sets apart a specific winery’s offering from its competitors. Management and marketing scholars have long recognized the role of differentiation in creating competitive advantage for firms (Porter 1980; Day and Wensley 1988). Price is a variable all firms use to exchange value with customers. Greater value in relation to price creates significant demand for a product. Cavusgil (1983) found that appropriate pricing can influence the success of export marketing programs. From this discussion, we offer our first three hypotheses:

Hypothesis 1

Brand image is positively related to winery export performance.

Hypothesis 2

Product differentiation is positively related to winery export performance.

Hypothesis 3

Value in relation to price is positively related to winery export performance.

Channel performance or the effectiveness of channel intermediaries has also been linked to export performance (Rosson and Ford 1982; Cavusgil and Zou 1994; Johnson and Raven 1996). The evidence is clear that the success of an export program in an overseas market rests, in large part, on the ability of the exporter to develop and maintain effective relationships with agents, wholesalers and retailers who operate in that market. This leads to our next hypothesis:

Hypothesis 4

Intermediary performance is positively related to winery export performance.

Existing literature has documented the impact of export commitment on export performance. Beamish et al (1999) found that firms that make a commitment to exports through the formation of a separate export unit significantly outperform firms that treat exports as just a part of their existing business. Axinn, Noordweier and Sinkula (1996) document five studies in which export commitment has been linked to export performance. Therefore, we expect to find a similar relationship in the wine industry.

Hypothesis 5

Export commitment is positively related to winery export performance.

As previously discussed, the wine consumer market is divided into many price-quality segments, which could potentially moderate the relationship between intermediary performance and export performance. In the lower price-quality segments, price is likely to be the main driver of demand, little else is required in terms of marketing or sales support. However, success in the higher price-quality segments requires intense selling efforts especially in relation to the “old world” European wines. Thus, the role of the intermediary is far more important in the higher price-quality segments. This leads to our first moderator hypothesis:

Hypothesis 6

The targeted price segment moderates the relationship between intermediary performance and winery export performance. The positive relationship is stronger for the luxury price segment than the economy price segment.

US wineries are motivated by return on investment in selecting which price-quality segments to serve. Anecdotal evidence from preliminary interviews with winery executives indicates that export markets are highly price-quality sensitive, so the payoffs are likely to be more from the lower price-quality segments. For example, Gallo, the largest US wine exporter, has achieved great success by exporting to the lower price-quality segments. Thus, we expect the relationship between export commitment and export performance should be stronger in the lower price-quality segments.

Hypothesis 7

The targeted price-quality segment moderates the relationship between export commitment and winery export performance. The positive relationship is stronger for the economy price segment than the luxury price segment.

In differentiating the roles of the agent versus the wholesaler as channel intermediaries, Aulakh and Kotabe (1997) note that wholesalers are independent distributors who do not share any marketing activities in the export market with the manufacturer. On the other hand, agents do share marketing responsibilities with manufacturers in export markets. In wine exports, winery executives do visit export markets in conjunction with their agents. Once product is sold to wholesalers the winery has little knowledge of the wholesaler’s activities or customers in the export market. Thus, wineries develop closer relationships with agents than with wholesalers. And the performance of agents becomes more critical to the winery export performance than wholesalers. This leads to our next moderator hypothesis:

Hypothesis 8

The type of intermediary moderates the relationship between intermediary performance and winery export performance. The positive relationship is stronger for the agent than for the wholesaler.

While performance of agents requires developing and maintaining close relationships, wholesalers respond better to direct financial incentives or resource allocations. Therefore, we expect the positive relationship between export commitment and winery export performance, will be stronger for wholesalers than for agents. This leads to our final moderating hypothesis:

Hypothesis 9

The type of intermediary moderates the relationship between export commitment and winery export performance. The positive relationship is stronger for the wholesaler than for the agent.

The US wine industry is comprised of approximately 1500 wineries. The industry, however, is highly concentrated with the top 10 wineries accounting for 70% (by volume) of US production. Wine is produced in every state except Alaska. California dominates the US wine industry in many ways. California has over 900 wineries and accounts for more than 90% of the wine produced and exported by US wineries. Northwest wineries (Washington, Oregon and Idaho) comprise approximately 200 wineries and are developing an export presence as well as excellent reputation for quality wines (Goldman 2000).

We began our study with extensive, personal interviews of senior executives in about 15 Northern California wineries, the Wine Institute, journalists and industry analysts that cover the wine industry. The questions were open-ended and exploratory, with interviews lasting one to four hours. The purpose of these interviews were (a) to find out what critical issues wineries face in developing and implementing successful export programs, (b) to hone in on the important variables that might impact export performance and their inter-relationships, and (c) to build industry support for the next, structured survey phase of our study.

Sampling Frame

An annual directory of US wineries (Wines & Vines 1999 Directory and Buyers Guide 1998) was used to generate our sampling frame. A list of all the wineries in California, Oregon, Washington and Idaho was compiled. These 1012 wineries accounted for well over 95% of the wines produced and exported by US wineries, making this a comprehensive sampling frame.

Based on field interviews with winery and industry association executives, and a review of the relevant literature, we designed, developed and pre‑tested two survey questionnaires, one version for exporters and another one for non-exporters. In March of 1999 both surveys were mailed to all 1012 wineries with a cover letter explaining that they should complete the exporter's version if they had exported in the last three years. The surveys were sent to a senior executive identified in the directory (Wines & Vines 1999 Directory and Buyers Guide 1998), typically a Vice President or higher, with a request to answer it if they were familiar with the winery’s export program, or to pass it on to an executive who had the necessary familiarity. The quality of responses in the returned questionnaires assured us that these were qualified key informants for the survey.

The cover letter mentioned that the study was funded, in part, by a grant from the US Department of Education. It also stated that the authors' university, and WineVision, supported the study. WineVision is an industry-wide initiative involving most of the major US wineries to improve their export effectiveness.

Thirty-four surveys were returned as undeliverable, resulting in a reduced sampling frame of 978 wineries. Wineries that did not return their survey by the specified date were sent a follow-up fax under the signature of the CEO of Fetzer Vineyards, who is one of the WineVision founders, asking them to complete the survey. The number of usable returned surveys was 236, representing 103 non-exporters and 133 exporters. Overall this represents a 24% response rate. The results presented in this paper relate only to the 133 exporters.

Operationalization of Variables

In order to address the issue of how to measure export performance, a study by Zou, Taylor and Osland (1998) developed a three-factor measurement model of export performance: the EXPERF scale. The three factors were Financial Export Performance (profitability, sales volume and growth), Strategic Export Performance (global competitiveness, strategic position and global market share) and Satisfaction with Export Venture (export venture success, satisfaction with export performance and meeting export venture expectations). The three-factor EXPERF scale had both convergent and discriminant validity in their study. We adopted the same 9 measures of export performance as Zou, Taylor and Oslund (1998).

After data collection, exploratory factor analysis of these 9 items showed they loaded on the same three factors comprising the EXPERF scale: financial export performance, strategic export performance, satisfaction with export program. We then tested a measurement model using confirmatory factor analysis with the same 9 items loading on the same 3 latent factors allowing the three factors to covary. We got an excellent fit for the measurement model with an average absolute standardized residual of 0.05, a Bentler-Bonnett Normed Fit Index of 0.91 and a Comparative Fit Index of 0.93. Each item loaded significantly onto its latent construct. We thus have evidence of convergent validity. Then we ran 3 more measurement models constraining each of the inter-factor covariances to 1, one at a time. In each case, the constrained measurement model had a worse fit than the original measurement model. This provides evidence of discriminant validity. Thus, financial export performance, strategic export performance, and satisfaction with exports can be used as separate dependent variables measuring different aspects of export performance (Table 1).

Two items tapping brand reputation,and packaging and labeling, respectively, were used to measure Brand Image (Table 1). Preliminary interviews with wine industry experts revealed that in the wine industry major product differentiation comes from the following three attributes: (1) varietal (name of the grape), (2) appellation (geographical region where the grapes are grown), and (3) whether the grapes have been grown in inhouse vineyards. Therefore, 3 items were used to tap these dimensions (Table 1). Two items measuring the attractiveness of price and the popularity of the wine were used to measure Price-Value (Table 1).

Organizational commitment to exports has been referred to in many different ways in the literature, as export commitment or top management commitment or managerial commitment (Stump, Athaide and Axinn 1998). However, a consensus has emerged that any such conceptualization must include both attitudinal and behavioral components. Attitudinal components measure how serious the organization is about the export program while behavioral components measure actual resource allocations to exports. Our measure of export commitment includes both attitudinal and behavioral components (Table 1).

We asked exporting wineries to evaluate the intermediary in their primary export country market on three items. These items, when factor analyzed, resulted in a single factor labeled Intermediary Effectiveness (Table 1).

Any cross-sectional study of individual firm performance must include control variables such as age of the firm and firm size. We use the number of years in business as a measure of the winery’s age. Firm size is often related to export success (Reid 1982; Axinn 1988; Moini 1995). Therefore, we include it as a control variable in our study. The number of cases a winery sold last year was used as the measure of winery (Table 1).

All of the multi-tem variables in Table 1 have reasonably high Cronbach’s alpha except brand image, product differentiation and price-value. A composite score on each multi-item variable was obtained by a simple average of the individual item scores.

Our first moderator variable is price-quality segment. In the questionnaire, we asked each winery to give us a breakdown of 100% of their export sales by five price-quality segments: economy (under $3 per bottle), popular ($3 - $7 per bottle), super premium ($7 - $14 per bottle), ultra premium ($14 - $25 per bottle), and luxury (over $25 per bottle). If a winery allocated greater than 50% of sales to a particular segment, it was categorized as predominantly exporting to that segment. If no segment received greater than 50%, the winery was not categorized as belonging to any particular segment. The frequency distribution of wineries across the five segments were as follows: economy (under $3 per bottle) 5, popular ($3 - $7 per bottle) 8, super premium ($7 - $14 per bottle) 26, ultra premium ($14 - $25 per bottle) 45, and luxury (over $25 per bottle) 34. Since some segments had few observations, we collapsed five segments into two along the median price: $14 and under economy, and over $14 luxury. We now have a dichotomous moderator variable, price-quality segment.

Our second moderator variable is type of intermediary. In the questionnaire we asked each winery to tell us, across all country markets, what kind of an intermediary is predominantly used (a) agent (b) wholesaler (c) retailer. Only 7 wineries reported predominant use of a retailer, too few for a meaningful analysis. Therefore, type of intermediary is a dichotomous variable, agent or wholesaler.

The means, standard deviation and the correlation matrix of our variables are presented in Table 2. In order to test Hypothesis 1-5 on main effects, we performed the OLS regressions presented in Table 3. All coefficient estimates in this and subsequent tables are standardized betas so no constant term is shown in the regressions. The adjusted R²  statistics, ranging from 0.35 to 0.41 are in line with similar surveys based on self-report data from key informants.

From columns 2, 3, 4 in Table 3 we see that brand image, product differentiation and price value are not significantly related to financial export performance, strategic export performance, and satisfaction with export program. So Hypotheses 1, 2 and 3 are not supported by our data. While this may run counter to well-established theory in strategy or marketing, we think we have a measurement problem with the operationalization of these variables. As seen in Table 1, these variables have the lowest Cronbach’s alpha, and therefore have high measurement error associated with them. We think this is a possible explanation for the non-significant results.

From columns 2, 3, 4 in Table 3 we find that intermediary performance and export commitment are significant and positively related to financial export performance, strategic export performance and satisfaction with the export program, providing evidence in support of Hypothesis 5 and 6. These results are robust in columns 5, 6, 7 even after we drop the non-significant variables, brand image, product differentiation and price-value. These three non-significant variables are dropped from further analyses of the moderating effects.

Results of the test of the moderating variable, price-quality segment, as spelled out in Hypotheses 6 and 7 are presented in Table 4. Columns 2 and 3 compare the economy segment ($14 and under per bottle) with the luxury segment (over $14 per bottle) in terms of financial export performance, columns 4 and 5 in terms of strategic export performance and columns 6 and 7 in terms of satisfaction with exports. Looking at the independent variable, intermediary performance, across the columns we see a consistent pattern. The standardized beta for intermediary performance is always significant and of higher magnitude in the luxury segment than for the economy segment. This shows that intermediary performance has a stronger positive relationship with export performance in the luxury segment than in the economy segment, providing support for Hypothesis 6.

In similar fashion if we look across columns for the export commitment independent variable, we see a different but consistent pattern. The standardized beta for export commitment in the economy segment is always positive significant and higher in magnitude than in the luxury segment. This is true for all three export performance measures. This shows that the positive relationship between export commitment and export performance is stronger for the economy segment than for the luxury segment, providing support for Hypothesis 7.

To test the moderating effects spelled out in Hypotheses 8 and 9, we look at the regression results in Table 5. Here we compare standardized betas of the same independent variables across type of intermediary, wholesaler versus agent. For intermediary performance, we find the standardized beta in the agent column positive significant and always higher than the wholesaler column for all three measures of export performance. The positive relationship between intermediary performance and export performance is stronger for the agent than for the wholesaler, providing support for Hypothesis 8.

In similar fashion if we look across columns for the export commitment independent variable, we see a different but consistent pattern. Here the standardized beta for the wholesaler column is positive significant and higher in magnitude than the agent for all three measures of export performance. The positive relationship between export commitment and export performance is stronger for the wholesaler than for the agent, providing support for Hypothesis 9.

We believe this study is a pioneering attempt at documenting the key success factors related to US winery exports in a large-scale, systematic manner. We replicate the results of other industry studies that find intermediary performance and export commitment to be key success factors for exports. Finding effective intermediaries for different export markets and supporting them in an appropriate manner appears to bring out the best in intermediary performance, which should also boost winery export performance. Thus, intermediaries are an external resource that wineries can benefit from when used effectively.

The other key success factor is internal, developing the winery’s internal commitment to an export program and supporting it with adequate resources. In our interviews with winery executives we found wide variance in terms of export commitment. Many small and medium wineries do not perceive any payoffs from participating in price sensitive export markets. Their myopic view does not recognize the increasingly global, competitive wine market and the huge inroads made into the US domestic market by wines from Australia, Argentina and Chile. An educational effort is needed here to get these wineries moving in the direction of exports. Industry-wide initiatives like WineVision have their work cut out for them.

A value-added contribution of this study is the examination of moderating effects of segmentation and type of intermediary on export performance. We examine two price-quality segments, economy ($14 and under per bottle) and luxury (over $14 per bottle). Within export markets wineries choose which segment they predominantly want to compete in. If a winery chooses to play in the economy segment, our findings suggest that export commitment is the more important success factor than intermediary performance. Developing this commitment within the management of the winery and then allocating adequate resources to travel, promote products, and participate in trade shows will yield greater payoffs than cultivating intermediary relationships. On the other hand, if a winery chooses to compete in the luxury segment, here the efforts of a high-performing intermediary will yield greater payoffs than throwing money at travel and promotion. The personal efforts of an effective intermediary developing export markets will yield greater results in the luxury segment while impersonal sales promotion techniques may be more effective in the economy segment.

Wineries are constrained by the availability of channel intermediaries in different export markets. They usually go with an agent or wholesaler depending on who is available. Our findings suggest that, if they go with a wholesaler, export commitment is more important than intermediary performance as a key success factor. Wholesalers are direct customers. They do not share marketing responsibilities in the export market with the winery. The best wineries can do with wholesalers is to give them promotional deals so they buy more for resale. Instead, if they go with an agent, intermediary performance is more important than export commitment for boosting export performance. With an agent, the winery has to share marketing responsibilities in the export market. They may travel together, make joint customer visits and develop common goals. The relationship with the agent, therefore, has to be closer and rely on social considerations in addition to economic ones.

Clearly, this study has all the limitations of a cross-sectional survey. The relationships uncovered should be treated as associational rather than causal. Further, the results may not be generalizable to industries other than wine. While we explain about 40% variance in export performance on the basis of two key success factors, clearly we had measurement problems with the brand image, product differentiation and price-value variables. Nevertheless, the findings of the study do have some interesting implications for wine industry practice and future academic research.

Aulakh, P. S. and M. Kotabe (1997), “Antecedents and Performance Implications of Channel Integration in Foreign Markets,” Journal of International Business Studies, First Quarter, 145-175.

Aulakh, P. S., M. Kotabe, and H. Teegen (2000), “Export Strategies and Performance of Firms from Emerging Economies: Evidence from Brazil, Chile, and Mexico,” Academy of Management Journal, 43 (3), 342-361.

Axinn, C. (1988), Export Performance: Do Managerial Perceptions Make A Difference?”, International Marketing Review, Summer, 61-71.

Axinn, C., Noordewier and Sinkula (1996), “Export Strategies and Export Performance: An Empirical Investigation of a Products/Market Typology”, Advances in International Marketing, (Ed.) S.T. Cavusgil, JAI Press Inc., 27-58.

Bello, D. C. and D. I. Gilliland (1997), “The Effect of Output Controls, Process Controls, and Flexibility on Export Channel Performance,” Journal of Marketing, 61 (January), 22-38.

Beamish, P., Karavis, L., Goerzen, A. and Lane, C. (1999), “The Relationship Between Organizational Structure and Export Performance”, Management International Review, 39, 37-54.

Buzzell, R. D. and B. T. Gale (1987), The PIMS Principles: Linking Strategy to Perfromance, New York: Free Press.

Cavusgil, S.T. (1983), “Success Factors in Export Marketing: An Empirical Analysis,” Journal of International Marketing and Marketing Research, 8 (2), 63-73.

Cavusgil, S.T. (1984), “Organizational Characteristics Associated with Export Activity”, Journal of Management Studies, 21(1), 3-22.

Cavusgil, S.T. and Zou, S. (1994), “Marketing Strategy Performance Relationship: An Investigation of the Empirical Link in Export Market Ventures”, Journal of Marketing, 58 (1), 1-21.

Day, G. S. and R. Wensley (1983), “Marketing Theory with a Strategic Orientation,” Journal of Marketing, 47 (Fall), 79-89.

Doering, William F. (1999) “…And Now the Numbers,” Wines & Vines, July, 44-51.

Goldman, David (2000), “Wine Shipments Buoyed By Surging Exports,” Wine Business Monthly, July, 44-47.

Johnson, J.L. and Raven, P.V. (1996), “Relationship Quality, Satisfaction and Performance in Export Market Channels”, Journal of Marketing Channels, 5, (3/4), 19-48.

Katsikeas, C. S., L. C. Leonidou and N. A. Morgan (2000), “Firm-Level Export Performance Assessment: Review, Evaluation, and Development,” Journal of the Academy of Marketing Science, 28 (4), 493-511.

Kirpalani, V. H. and N. B. MacIntosh (1980), “International Marketing Effectiveness of Technology Oriented Small Firms,” Journal of International Business Studies, 10 (Winter), 81-90.

Moini, A.H. (1995), “An Inquiry into Successful Exporting: An Empirical Investigation Using a Three Stage Model”, Journal of Small Business Mnangement, July, 9-25.

Porter, M. E. (1980), Competitive Strategy, New York: Free Press.

Pfeffer, J. and G. R. Salancik (1978), The External Control of Organizations, New York: Harper & Row.

Reid, S. D. (1982), “The Impact of Size on Export Behavior in Small Firms,” in Export Management: An International Context, M. R. Czinkota and G. Tesar, eds. New York: Praeger Publishers.

Rosson, P. J. and L. D. Ford (1982), “Manufacturer-Overseas Distributor Relations and Export Perfromance,” Journal of International Business Studies, 13 (Fall), 57-72.

Scherer, F. M. and David Ross (1990), Industrial Market Structure and Economic Performance, 3^rd edition, Boston: Houghton Mifflin Company.

Stump, R.L., Athaide, G.A. and Axinn, C. (1998), “The Contingent Effect of the Dimensions of Export Commitment on Export Financial Performance: An Empirical Examination”, Journal of Global Marketing, 12 (1), 7-25.

Thach, S. V. and C. N. Axinn (1991), “Pricing and Financing Practices of Industrial Exporting Firms,” International Marketing Review, 8, 32-46.

Wine Institute (2000), "Another Record for US Wine Exports: $548 million," www.wineinstitute.org/communications/statistics/exports99.htm, Wine Institute, 425 Market Street, Suite 1000, San Francisco, CA 94105.

Wines & Vines 1999 Directory and Buyers Guide (1998), Volume 79 (December), The Hiaring Company, San Rafael, CA.

Zou, S., Taylor, C.R. and Osland, G.O. (1998), “The EXPERF Scale: A Cross National Generalized Export Performance Measure”, Journal of International Marketing, 16 (3), 37-58.