Postponement Practices in the Wine Industry - San Francisco State ...

Download PDF
10 downloads 21006 Views 112KB Size Report
Comment
automotive and high technology sectors (Yang et al, 2005). .... California wineries account for 45% of all U.S. Wineries (Wine Institute, 2008a). However, the ...
Postponement Practices in the Wine Industry: Adoption and Attitudes of California Wineries Dr. Susan Cholette Department of Decision Sciences, College of Business San Francisco State University 1600 Holloway Ave. San Francisco, CA 94132 Phone: 415-405-2173 Fax: 415-405-0364 [email protected] Abstract (116 words) With the global growth of both export programs and private label brands, wineries must allocate production across an increasing variety of sales channels before demand is known. Misallocation may result in surpluses in some channels and lost sales opportunities in others. Although postponement would aid in solving this problem, this strategy has not been widely adopted. This research involved surveying California wineries (N = 142) on their current exporting, private labeling and postponement practices and on their predictions for future usage. A majority of the respondents export, but many fewer have private label channels or engage in postponement. Additionally, wineries anticipate increasing adoption of export and private label practices but show less enthusiasm for utilizing postponement. Keywords 1. Postponement 2. Private Labels 3. Exporting 4. Channel Allocation 5. Wine Industry 6. Survey

Biographical Sketch Susan Cholette is an Assistant Professor of Decision Sciences in the College of Business at San Francisco State University. Her research interests include supply chain management and operations issues within the wine industry. Her publications include articles in Interfaces, International Journal of Production Research, International Journal of Wine Business, International Journal of Wine Marketing, and International Journal of Revenue Management. She earned a Ph.D. in Operations Research at Stanford University and a B.S.E. in Electrical Engineering at Princeton University.

Text body: 6156 words

1

1 Introduction 1.1 The Proliferation of Sales Channels The traditional image of a winery is that it sells wines under its own brand in a local market. However, wineries may support additional sales channels, often requiring special packaging and labeling formats. For instance, wineries can sell wine in bulk to negociants or other wineries. In addition to having multiple brands, such as second labels, wineries can also produce private label or co-labeled products. Private label wines are defined as those that are branded by a retail channel, such as a supermarket or restaurant, and then sold exclusively through that channel. Co-labeled wines identify both the winery and retail channel and are sold only through that retail channel. While not as prestigious or profitable as the winery’s own brand, these types of wine serve as an additional sales channel for product that may otherwise go unsold. Private label wines represented less than 1% of the total U.S. wine market in the late 1990s (Ward et al, 2002), but by 2004 commanded a 20% share (Popp, 2003). Private label wines have achieved more presence internationally, such as within the European Union. For instance, store-branded wines account for over half of sales in the three largest U.K. supermarkets (Chaney, 2004). Private label wines exist across a wide range of prices and sales volumes, and this channel is potentially open to all wineries, not just those producing inexpensive, mass-volume wines. When a winery exports products under its own brand, it usually needs to customize labels or packaging, such as translating the ticket to the local language or meeting other requirements specific to the target market. Wine exports as share of global production increased from 15% in 1990 to 25% percent in 2000, with little sign of abating (Anderson et al, 2001). This growing internationalization of the wine industry is occurring in all regions, even ones that have not traditionally supported exporting programs. Wineries everywhere are facing a growing need to manage multiple sales channels. 1.2 Postponement as a Channel Management Tool With these additional market opportunities comes the challenge of determining product allocation to meet each channel’s needs. Many other industries have year-round sourcing, but wine production adheres to a rigid yearly cycle. Grapes are harvested or purchased once a year. Bottling follows within a few months, before demands are fully known. Products with different packages or labels may contain the same wine but are not substitutable. Product misallocation may simultaneously result in undesirable surpluses in some channels and lost sales opportunities in others. This research considers postponement as a potential channel management tool. Also referred to as delayed differentiation, postponement can be used to mitigate the effects of demand uncertainty, and many types of postponement strategies are possible. Van Hoek et al (1999) discuss when logistics postponement, a combination of time and place postponement, may be more appropriate and when form postponement may be. Van Hoek et al (1999) provide examples for different industries, and their one example of a food industry company utilizes deferred packaging at a regional warehouse near the destination market. A review of published research finds very few papers specifically investigating postponement practices within the wine industry. Two studies consider allocating wine

2

production under demand uncertainty by utilizing multi-stage, multi-facility networks, serving to decentralize final product differentiation. This setup theoretically allows wineries to effectively meet varying regional demands (Yu and Li, 2000) and to delay final bottling until transport to the regional market has occurred (Van Hoek, 1997). The latter is an especially relevant concern in the wine industry as glass comprises nearly 50% of the weight of a 750ml bottle of wine. However, actual industry practice suggests that decentralizing the bottling and distribution functions is impractical for most wineries. Both Orr (1999) and Twede et al (2000) emphasize that packaging beverage products is a high-speed, automated process requiring expensive equipment, thus favoring centralization. Despite the advantages that multiple localized warehouses may have for reaching markets, wineries typically centralize warehousing and outbound logistics functions. Before purchase by Constellation, Mondavi relied on a single U.S. warehouse facility in Lodi, located in California’s Central Valley, for handling all domestically produced wines. As logistics postponement is less feasible for most wineries, they may instead consider postponing the finishing processes that differentiate products between sales channels. Zinn and Bowersox (1988) define levels of form postponement, from deepest (earliest in the production process) to most shallow (latest) as follows: manufacturing, assembly, packaging and labeling. No specific mention of wine production and distribution is made, but the authors state the circumstances that may make different levels of postponement appealing to winemakers. The harvest cycle for grapes precludes manufacturing postponement from being a feasible option. However, assembly postponement may be appropriate for channel-specific blend recipes; perhaps a winery may change the percentages of component varietals to meet the tastes of consumers in different export markets. Packaging postponement is appropriate when the product is sold in multiple sizes or formats, such as bottles versus Tetrapaks™. Labeling postponement should be considered when the product is marketed under several brands utilizing the same bottling format. Van Hoek (1999) emphasizes that delayed packaging is likely to be a natural level of postponement within the food industry. Cholette (2009) provides a mathematical model of postponement at the packaging and labeling stages, as shown in Figure 1. Four distinct sales channels, Brand, Private Label A (PL-A), Private Label B (PL-B) and Bulk, are sourced from a common base product. If demand is uncertain within the channels, a winery can opt to hold inventory either in tanks or at a later finishing point, in blanks, which are sealed but unlabelled bottles. Blanks can be differentiated into only two finished products: Brand and PL-A. The PL-B product utilizes a different packaging format, such as a special bottle shape, size, color or closure. Figure 1 represents wine after it has been blended, and the option of tailoring the final blend to the different sales channels is considered. Results from Cholette (2009) indicate that wineries would be advised to hold part of their production as undifferentiated product at one or both intermediate stages, as this practice would improve profitability when the winery faces uncertain channel demands.

3

Figure 1: Postponement and Product Differentiation Branded channel

Tanked wine

Bottling process

Blank wine

Labeling process

Private Label A channel Private Label B channel

Bulk wine channel

Unlabeled Pl-B brand

Source: Cholette (2009)

1.3 Challenges to Implementation of Postponement Although the concept of postponement was introduced into academic literature in 1950, it received little attention for many years (Yang et al, 2004). Industry acceptance has also been slower than expected, with predominantly more implementations in the automotive and high technology sectors (Yang et al, 2005). In particular, agribusiness has lagged behind other industries (Van Hoek, 1999). One might ask “if postponement is such a good idea, why is it not more widespread?” One of the fundamental barriers to postponement is a firm’s ability to change processes or even to adjust corporate mindset to consider use of such techniques. An extensive survey by Yang et al (2005) finds that change management is the third most significant barrier to implementing postponement. Cholette (2009) provides anecdotal evidence from a few wineries that shows resistance to the idea. Although the wine industry is known for its adherence to tradition, it may stand to benefit from implementing postponement techniques. This research investigates how widespread postponement practices are within California wineries and attempts to determine whether some wineries are more likely to utilize it than others. The rest of this paper is organized as follows. Several propositions concerning California wineries’ current and anticipated practices of exporting, private labeling and postponement are presented. The survey design and implementation are discussed. Survey results are then presented, leading to discussion of whether the aforementioned propositions are supported by the survey data. These findings are then used to draw some tentative conclusions and predictions about practices in the California wine industry. Lastly, plans and recommendations for future research into postponement practices within the wine industry are presented. 2. Research Propositions While postponement practices offer real cost savings and increased responsiveness for handling multiple sales channels, prior conversations with personnel 4

from a few wineries in California would suggest that many operations managers do not yet consider postponement as a potential strategy. The authors suspect this reluctance has to do with a historical lack of need for product differentiation and that these perceptions may shift with time, especially if the number of wineries’ sales channels grows over time. The following propositions are considered. Proposition 1: Larger wineries are more likely to have export (1a) and private label (1b) channels than smaller wineries. Most very small wineries (defined as those selling less than 5000 cases annually) may serve limited markets with a single brand. As wineries grow, they may develop the economies of scale to support and the marketing need to provide multiple brands across several markets. It should be noted that both private and co-labeled wines are denoted in this research as “private label.” Proposition 2: Larger wineries are more likely to engage in postponement overall (2a) and to postpone earlier (2b) in the process than smaller wineries. Smaller wineries may not have the need or domain expertise necessary to engage in postponement. As wineries grow, they may also develop the sophistication necessary to implement postponement practices. Given that deeper postponement is typically more challenging, larger wineries would seem to be more likely to practice it. Proposition 3: Wineries that support either export or private label programs are more likely to engage in postponement practices than wineries of similar size that engage in neither practice. To mitigate any confounding effect of winery size, wineries are categorized by size and then noted by category as to whether they engage in either exporting or private labeling. Postponement usage from these wineries is compared to that of wineries with neither an export nor a private label program. Wineries supporting multiple channels may need more assistance in responding to channel demand variations. It is anticipated that such wineries would be more disposed to utilize postponement. Proposition 4: More wineries anticipate exporting (4a), private labeling (4b) and practicing postponement (4c) in the next 5 years. While wineries may not currently engage in some or all of these practices, they may anticipate adopting one or more of them within the next few years. For example, if more non-practicing wineries adopt postponement than practicing wineries discontinue postponement, this would represent an overall growth in postponement usage among established wineries. Proposition 5: Wineries predict that more California wineries will export (4a), have private labels (4b) and make use of postponement (4c) in the next five years. A winery may plan on staying small or otherwise serving a limited market. However, even niche players may foresee overall growth in these practices for their peers. Therefore, it seems reasonable to expect that wineries predict the adoption of each of these practices to increase in the next few years.

5

3. Research Methodology 3.1 Survey Design The survey instrument is comprised of 13 questions followed by an optional identification section and shown in Appendix A. To increase the response rate the survey was kept brief on purpose. The survey was pre-tested on a small audience of winery personnel and other wine industry professionals, resulting in some adjustments to questionnaire wording. Both the survey itself and the introductory letter were worded as concisely and neutrally as possible to minimize response bias. The ordering of the questions is deliberate, as inquiries about exporting, a more common practice among wineries, precede those on private labeling and postponement. Participant opinions about each of these practices were solicited after asking the winery to identify their own actual practices. Except for the free-form identification question, all questions conform to a radiobox format, a multiple choice list allowing a single selection. The majority of questions offer the respondent either a yes/no/NA or a yes/no/”no opinion” selection to keep survey taking as simple and quick as possible and to reduce the possibility of respondent error. While most yes/no questions may seem inappropriate to also have an N/A option, this response option was purposely included to cover respondents who may have been unsure about their winery’s status or were confused by the question. The survey is implemented through Zoomerang, a leading internet survey provider. Each respondent’s answers are saved individually and anonymously, except when identification is voluntarily provided. While a couple respondents provide contradictory answers, such as claiming not to export and then stating that their exports require a different labeling or packaging format, a detailed examination of individual respondents’ records shows very few seeming inconsistencies. All such discrepancies, and any re-interpretation of responses, are noted in the appropriate results section. The earlier (deeper) into the production process that a company engages in postponement, the more sophisticated the support processes must be. Van Hoek (1999) finds not only that food companies lag behind other sectors in adopting postponement, but also that their level of postponement is often the most shallow, occurring at the very end of the production process. The survey lists stages of postponement in order from the earliest (blending) to the least involved (packing). Although not formally considered a level of form postponement by Zinn and Bowersox (1988), the packing process, the boxing and/or palletizing of labeled wines, may be purposefully delayed. Wineries are asked to select the earliest stage of postponement they practice. For example, if a winery indicates that they engage in postponement at the packaging stage, it is not possible to determine whether or not that winery also employs pure labeling postponement for another set of products or as a second decoupling point within the production process. But it would be certain that the winery does not postpone the blending process for any of their products. It would have been more informative to know whether a winery engages in each of these postponement stages, and it is possible some respondents may have misinterpreted what is meant by “earliest stage.” However, the question format was designed to support the perceived need for survey brevity.

6

3.2 Survey Methodology Limited resources and the need to reach a large number of participants restricted the survey to implementation through the internet. The sampling frame consists of the 1480 California wineries in a recent industry contact database (Wines and Vines, 2006) that provide email addresses. This figure represents approximately 60% of the state’s bonded wineries (Wine Institute, 2008a). In April 2008 a letter was emailed to each of the wineries, requesting that the winery complete the survey by following an embedded link and guaranteeing confidentiality of individual responses. While a contact email was included, and the survey administrator 1 provided some clarifications by email, all communications attempted to maintain as neutral tone as possible to minimize any response bias. No follow-up contact occurred, and the survey was left open for 3 weeks, with the majority of responses collected within the first week. Of the 1480 emails sent, 124 survey invitation emails were returned due to outdated or incorrect contact information. The online survey was visited just under 200 times, resulting in 142 completed individual surveys. No surveys were only partially completed. While the survey responses are recorded anonymously, each email recipient received an individually keyed survey, preventing double counting of participants. The response rate of 10.5% is small but not atypical for an email-administered internet survey. In fact, a recent internet survey administered by Wines and Vines (Gordon, 2008) claims a “healthy” response rate of 6%. An N of 142 is comparable in size to that of four recent industry-sponsored surveys; Fisher (2008a and 2008b), Gordon (2008), and Pregler (2008) have N values of 372, 242, 288 and 172, respectively. It should also be noted that each of those four surveys target all U.S. wineries, not just ones within California. This broadening more than doubles the potential sample frame, given California wineries account for 45% of all U.S. Wineries (Wine Institute, 2008a). However, the sample frame is restricted to California to limit responses to include only producers from a more established wine region. For instance, California wineries are responsible for 95% of total U.S. wine exports (Wine Institute, 2008a). It may be informative to sample other survey-based investigations concerning the prevalence of and attitudes towards emerging supply chain practices from relatively recent operations and supply chain management (SCM) journals. While Van Hoek (1999) administered a detailed questionnaire on postponement practices and implementation barriers to 16 respondents across multiple countries and industries, most other researchers target either a specific country or industry and opt to use a less detailed survey instrument in order to collect a larger sample. Basnet et al (2003) query New Zealand firms on the state of several SCM practices, obtaining 69 responses and achieving an 11% response rate. Simatupang and Sridharan (2004) follow up in the same region, focusing on intrafirm collaboration, and they get data from 76 New Zealand retailer and supplier companies, resulting in a 21% response rate. Olhager and Selldin (2004) survey members of the Swedish Production and Inventory Management Society and gather 128 responses, a 25% response rate, concerning the adoption of and opinions about several SCM practices. Sánchez and Pérez (2005) investigate the supply chain flexibility of Spanish firms in the automotive industry, and they achieve a higher total number of responses (N=126) and response rate (35%) than many other researchers. Skjoett-Larsen 1

Jon Carney is acknowledged for his thoughtful and thorough work in administering the survey. This research would be incomplete without the data provided by (Carney, 2008).

7

et al (2003) administer an email-delivered survey on CPFR practices to Danish firms, achieving a much lower response rate of 7.8%, with 218 responses. Of the papers reviewed, the survey implementation of this research is most similar to that of SkoettLarsen et al (2003), so a 10.5% response rate is not unexpectedly low. Three wineries explicitly refused to participate, citing privacy concerns. A couple responded that they were misidentified by the database as wineries, and thus could not respond as the survey was misdirected. No other follow up of non-respondents was feasible, preventing further accounting for any non-response bias. It might be expected that wineries completing the survey are more likely to engage in exporting, private label or postponement practices than a truly random sample of wineries would yield. However, no incomplete surveys were received, so no wineries dropped out after discovering that they did not engage in the practices studied. 4. Results The results from the survey are considered in the order the questions are asked. The relatively small sample size and response rate precludes formal hypothesis testing, so frequency counts are presented. As winery size is central to many of the propositions, it is important to consider whether the sample is representative of the population. Although no data that quantifies California wineries by production level is readily available to the public, it is possible to compare the distribution of responding wineries’ sizes to that found by other surveys. Table 1 displays the respondent breakdown by size. While the greatest percentage of respondents sell less than 5000 cases (where a case is 12 750ml bottles) annually, this is not an anomalous result. Table 1 shows that a recent WineBusiness.com survey by Fisher (2008b) on closures also finds that 47% of respondents sell fewer than 5000 cases annually. Two WineBusiness.com surveys (Fisher 2008b, Pregler 2008) have a larger representation of wineries with annual sales over 50,000 cases, at 19% and 20% respectively, than does this survey, at 6%, suggesting the sample may under-represent medium to large wineries. Such variation between surveys is not uncommon. Gordon (2008) presents a tasting room survey where 64% of their respondents identified annual sales as being under 5000 cases. Yet only 40% of respondents are below the 5000 case level in a comparableWineBusiness.com tasting room survey (Fisher 2008a). Table 1: Question 1 Results, Comparison to Similar Survey Postponement Survey Winebusiness.com Counts Percentage Closure Survey very small Less than 5,000 cases 67 47% 47% 5,000 to 9,999 cases 34 24% 13% small 10,000 to 24,999 cases 23 16% 11% 25,000 to 49,999 cases 8 6% 10% medium to 50,000 to 99,999 cases 6 4% 9% large 100,000 to 499,999 cases 2 1% 6% 500,000+ cases 2 1% 4% Total 142

8

The demarcation of what makes a winery “small” does vary in both the academic and trade literature. Undoubtedly, a winery selling 30,000 cases annually has more in common with “small” wineries than a winery selling hundreds of thousands or even millions of cases a year. Furthermore, such a winery would still be classified as small by Fisher (2008b). Three size categories are defined: very small, small, and medium-tolarge. The 67 very small wineries, 57 small wineries, and 18 medium-to-large wineries comprise a reasonable number of respondents for each size. It should be noted that sales volumes, rather than production levels are considered; this inherently assume that sales volumes are close to production levels, although a winery often may not be able to sell all its production in reality. 4.1 Export Practices Wineries’ current practices and predictions concerning exporting are examined in questions 2 through 5. The 5 respondents who indicate “N/A” are counted as not exporting. There are 79 exporters, representing a slight majority (56%) of respondents. Table 2 shows the affirmative responses, categorizing wineries by size. A quick inspection of Table 2 reveals that the medium-to-large wineries have a greater proportion of exporters, with all but one exporting. Conversely, only 25 of the 67 very small wineries, 37%, currently export. Size appears to be an important factor in the likelihood of a winery exporting, as expected by proposition 1a. Table 2: Affirmative Responses to Exporting Practices (Questions 2 and 3) Question #3- Do your exports Question #2: Does your winery require special export? labelling/packaging? very small small medium-large total

counts (N=142) % affirmative counts (N = 79) % affirmative 25 37% 9 36% 37 65% 23 62% 17 94% 16 94% 79 56% 48 61%

Question 3 asks whether exporting wineries require special labels for their exported wines and only considers the answers from the 79 wineries who previously indicate they export. Two “yes” responses were received from wineries who indicated they did not export, but their responses are not included as it is likely they were merely considering the question hypothetically. As can be seen in Table 2, all but one of the medium-to-large exporting wineries utilizes special labeling or packaging. Yet just over a third (36%) of the very small exporting wineries utilize special labels or packaging. Such discrepancies may occur because these very small wineries are likely to export only to Canada, a relatively easy market for U.S. companies. Other than mandating a French translation for the origin and type of wine, Canada has similar labeling requirements to the U.S. Thus wineries can either attach an additional sticker on wine bottles bound for Canada or, as is commonly done within the U.S. produce sector, label all products to conform to Canadian standards, even those predominantly destined for domestic markets. Canada is the largest single external market, accounting for 25% of U.S. export sales,

9

although the European Union as a whole accounts for half of U.S. wine exports (Wine Institute, 2008b). Wineries are also asked to anticipate future export practices. Question 4 concerns a winery’s self-predictions, and question 5 involves forecasting for the California wine industry as a whole. Table 3 shows that respondents predict the prevalence of exporting to increase, as 70% expect to be exporting in 5 years, compared to the 56% who currently export. This supports proposition 4a, which predicts that more wineries plan to export. Furthermore, this optimism towards exporting holds across all size categories. While the only medium-to-large winery that does not currently export has plans to, over half of the very small wineries plan to export, a notable increase from the 36% that currently do. Table 3: Wineries’ Predictions about Exporting Practices (Questions 4 and 5)

very small small medium-large totals

Question #5: Do you anticipate that more Question #4: Do you anticipate exporting California wineries will export in the next 5 years? in the next 5 years? counts counts no % counts counts no % yes no opinion affirmative yes no opinion affirmative 40 25 2 60% 51 1 15 76% 42 9 6 74% 40 2 15 70% 18 0 0 100% 15 1 2 83% 100 34 8 70% 106 4 32 75%

Comparing individual responses to question 2 and 4 simultaneously also yields insight. Considering the 79 wineries that currently export, only five either actively anticipate discontinuing (1 winery) the program or have not decided (4 wineries) whether to continue it. The remaining 74 wineries, 94% of the current exporters, plan to continue their export programs. The broader question of whether wineries predict if more of their peers will export follows. Three-fourths of respondents expect that more California wineries will export over time, supporting proposition 5a. Additionally, Table 5 shows comparatively little difference in opinion by size. This is not surprising, given that this question involves predictions for the overall regional industry. Wineries that are planning to remain small or target local niche markets seem to recognize that their peers may plan to expand, even if they themselves do not. 4.2 Private Label Practices Question 6 asks whether wineries have more than one brand. Table 4 suggests that the prevalence of multiple brands increases with winery size. Of the 63 wineries that offer multiple brands, the majority (63%) of these wineries also have private label products. Overall, private label wines remain a less common channel, utilized by only 28% of all respondents. Table 4 shows that half of all medium-to-large winery respondents have a private label wine, suggesting that larger wineries are more likely to use private label channels than are their smaller counterparts, as anticipated by proposition 1b. However, support is at best tentative, given the small number of respondent wineries, especially within the medium-to-large category.

10

Table 4: Affirmative Responses to Labeling Practices (Questions 6 and 7)

very small small medium-large overall

Question #7: If you Question #6: Does your bottle multiple brands, % of wineries are any private label? that engage in bottle multiple brands? counts counts private labeling (N=142) % affirmative (N=63) % affirmative 23 34% 15 65% 22% 27 47% 16 59% 28% 13 72% 9 69% 50% 63 44% 40 63% 28%

Wineries’ predictions about private labeling follow. Table 5 shows that more wineries expect to use this channel, as proposition 4b predicts. In particular, 36% of respondent wineries anticipate offering private label products in the next five years, an increase from the 28% who currently use this channel. Table 5 also shows that this predicted level remains unchanged for medium-to-large wineries, with the increase coming from small and very small wineries planning to offer private labels. Comparing individual answers to questions 7 and 8 reveals that only 5 of the 40 wineries that currently practice private labeling are either planning to stop the practice (3 wineries) or have not decided whether to continue it into the future (2 wineries). While this represents a greater percentage of less committed practitioners than is seen for exporters, seveneighths of the respondents with private label wines plan to continue servicing this channel. Table 5: Wineries’ Predictions about Private Label Practices (Questions 8 and 9)

very small small medium-large totals

Question #9: Do you anticipate that more Question #8: Do you anticipate having private California wineries will have private label or co-labeled brands in the next 5 years? label or co-labeled brands in the next 5 years? no % no % counts yes counts no opinion affirmative counts yes counts no opinion affirmative 20 39 8 30% 31 3 33 46% 22 28 7 39% 32 2 23 56% 9 9 0 50% 11 1 6 61% 51 76 15 36% 74 6 62 52%

Question 9 concerns proposition 5b, which predicts that the wineries anticipate more California wineries offering private and co-labeled brands. Responses show that a slight majority (52%) of respondents predict that these practices will increase for California wineries as a group. Six of the 142 respondents, only 4%, expect no such increase, and the remaining 44% had no opinion. 4.3 Postponement Practices Questions 10 through 13 relate to the central focus of this research, wineries’ current and anticipated postponement practices, with the responses summarized in Table 6. Three respondents answered “N/A” for whether they postpone finishing, and these responses as considered as negative, especially as their answers to the next question show that no earliest stage of postponement exists for any of these 3 wineries. These respondents may have been confused as to the exact definition of “finishing,” which for a 11

winery would entail the blending, packaging and labeling of wine, followed by the packing of bottles to be ready for sale to the next tier within the supply chain. Table 6: Affirmative Responses to Postponing Practices (Questions 10 and 11)

very small small medium-large overall

Question #10: Does your winery engage in postponing Question #11: If yes, what is the earliest the finishing of wine? stage of postponement? counts yes % affirmative blending bottling labeling packing (N=142) 19 28% 3 13 1 2 16 28% 6 7 2 1 8 44% 4 2 2 0 43 30% 13 22 5 3

While response numbers are too low to draw anything but tentative observations, it would seem that medium-to-large wineries are more likely to engage in postponement than smaller wineries, supporting proposition 2a. Figure 2 presents the 43 wineries that engage in postponement by size, showing the earliest stage utilized by percentage of wineries in that size category. The use of blending postponement increases with winery size. For instance, half (50%) of the medium-to-large wineries that practice postponement start at the blending stage, compared to 38% of the small wineries and only 16% of the very small wineries. For very small wineries, bottling postponement is the most common earliest starting point. Interestingly, the practice of starting postponement at the labeling stage increases with winery size. Figure 2: Earliest Postponement Stage by Size, by Percentage of Wineries that Practice Postponement 80% 70% 60%

blending

50%

bottling 40%

labeling

30%

packing

20% 10% 0% very small

small

medium-large

Winery Size

Given the suspicion that winery size influences postponement practices, it is appropriate to control for size when considering if adoption of postponement may be affected by the practice of exporting or of private labeling. Responses are thus grouped

12

by size. All of the medium-to-large wineries engage in either one or both of these practices, while only 41 of the 57 small wineries (72%) and 36 of the 67 very small wineries (54%) do. One can consider each of these latter two size categories to assess whether wineries that are engaged in at least one of these practices are more likely to postpone, as proposition 3 anticipates. A review of the data does not support this proposition. For instance, 31% of very small wineries with either exporting or private labeling channels postpone. Only 26% of very small wineries without such channels postpone, an unremarkable difference. Furthermore, only 27% of small wineries with either exporting or private label channels postpone, while 31% of wineries having neither export and private label channels use postponement. The survey results provide no evidence that postponement may be more likely for wineries that have export or private label channels, once winery size is taken into account. Table 7 summarizes wineries’ predictions about postponement. Contrary to the expectation inherent in proposition 5c, respondents do not foresee increasing their own postponement practices. Nor do the majority of respondents expect postponement practices to increase among their peers. It should be noted that a majority (60%) of wineries do not have an opinion about the future of postponement in California wineries. If only the subset of 57 wineries with opinions is considered, 61% expect postponement practices to increase. Table 7: Wineries’ Predictions about Postponement Practices (Questions 12 and 13)

very small small medium-large totals

Question #8: 12. Do you anticipate engaging in postponement in the next 5 years? no % counts yes counts no opinion affirmative 16 44 7 24% 16 35 6 28% 8 9 1 44% 40 88 14 28%

Question #13: Do you anticipate that more Californian wineries will engage in postponement in the next 5 years? counts no % counts no opinion affirmative yes 13 14 40 19% 15 5 37 26% 7 3 8 39% 35 22 85 25%

Comparing results from question 10 to those of questions 12 and 13 shows that that none of the small or medium-to-large wineries that currently use postponement plan to discontinue it or have definite negative opinions about its industry adoption. Only 2 of the very small wineries that current postpone plan to discontinue the process. Almost all (94%) of the respondent wineries that have postponement programs plan to continue them. 5. Conclusion 5.1 Implications and Recommendations for Wineries As previously mentioned, this survey has a relatively low response rate and small sample size, precluding formalized hypothesis testing. However, this exploratory study provides some insights into wineries’ current practices and future directions. Seventy percent of California wineries surveyed predict that they will be exporting within 5 years, an increase from the 56% that presently do. An even greater percentage (75%) expects that more of their peers will soon export. To a lesser extent, respondents also anticipate 13

growth in private labels. Interestingly, despite the anticipated growth in these sales channels, they do not foresee changing operations, at least with respect to using postponement to delay product differentiation. The survey results agree with the literature with respect to the increasing importance of exporting. Even among the very small wineries, over a third export, although nearly two-thirds of these very small exporters do not require special labels or packaging. The larger wineries have both a greater propensity to export and to have exported goods require differentiated labels or packaging. California wineries of all sizes seem to recognize that their region is gaining global awareness and are planning to enter international markets. As more wineries export and as these export programs expand to more markets, wineries’ operations managers will need to handle an increasing variety of labeling and packaging formats. Relatively few wineries (28%) have private label wines. While more respondents anticipate that they will engage in private labeling than presently do, there is less industry enthusiasm over this channel, as only 38% of the respondents actively plan to use it. Private label wines have lower margins and convey less prestige than do winery-branded sales, so it is understandable that wineries show less enthusiasm for adopting this practice than they do for exporting. A slim majority (52%) of respondents expect that more California wineries will engage in private labeling. The percentage difference between these questions suggests that wineries recognize private labeling is increasing in importance but still prefer to leave this channel for others. If the predictions for California wineries hold in aggregate, more wineries may eventually embrace private labeling than currently anticipate doing so. As a group these wineries are planning to expand their sales channels, both by growing their export programs and, to a lesser extent, engaging in private labeling. At present few respondent wineries (30%) practice any form of postponement. Also, although the wineries that have postponement programs in place plan to continue them, very few anticipate adopting postponement practices in the near future. This lack of connection between the expansion of sales channels and underlying operational strategies to help support them is noteworthy, but not surprising in retrospect. The publications reviewed show both that companies are often slow to implement postponement strategies and that agribusiness lags behind other sectors in the use of such techniques. In consider the interaction between winery size and postponement usage, there is no strong evidence that a winery’s size affects whether postponement is used or to what level postponement is practiced. However, graphical views of the responses would suggest that some influence may exist. For instance, assembly-level postponement (blending) is more prevalent among medium-to-large wineries than within smaller ones. Within a size category, the presence of export or private label channels does not seem to make a winery more likely to postpone. It seems reasonable to expect that before a winery adopts postponement practices it must not only have the need of managing demand uncertainty across more sales channels, but also the operational expertise to support such an implementation. Such sophistication is more likely to come with size and experience. For instance, Figure 2 shows increasing use of labeling-only postponement with size. Economies of scale may make deferred labeling more appealing when larger volumes are being held back, and a smaller or more recently established

14

winery may not be able to employ this technique, even if it exports to multiple markets, each with particular label requirements. The short survey did not investigate other operational strategies, and it is possible that wineries may utilize different tactics for supporting the increased complexity inherent in allocating production among a greater number of sales channels. For instance, building stronger relationships with supply chain partners may enable the winery to produce to a more accurate sales plan and receive commitments that all initial allocations will be sold. Yet even a winery benefiting from accurate forecasts and aggressive marketplace promotion of its brands will still experience channel misallocations. The number of respondents lacking opinions on postponement is high, suggesting that many wineries may not yet be familiar with postponement techniques. Industry attitudes towards postponement may change over time. Wineries with an agile operations strategy towards meeting demand uncertainties will be more profitable and will be better poised to expand, perhaps even inspiring neighboring wineries to follow their example. 5.2 Recommendations for Future Research This research is primarily exploratory in nature, and brief surveys inherently have limitations. In particular, the implementation process did not allow for clarification of ambiguous questions or answers nor did it enable any follow up with appropriate questions. Engaging in a more in-depth investigation of a subset of the participants will result in a better understanding of the industry perceptions of postponement and private label practices. Of the 142 respondents, 37 responded to the optional identification section, offering to be contacted, and several responded personally to the letter of introduction, expressing interest in the results and a willingness to be of further assistance. The global wine industry is in a state of change, with evolving consumption and production patterns. The Wine Institute (2008a) shows U.S. exports have increased in value by 73% in the last 5 years. Given that wineries predict more exporting both for themselves and as a group, it would be useful to administer a similar survey with the same sampling frame in a few years time to see how responses change. For instance, will the responding wineries’ predictions for low levels of postponement implementations come true or might the original suppositions about increasing postponement practices be validated? Many of the WineBussiness.com surveys on wineries’ practices and attitudes towards bottle closures, equipment investments and tasting room sales are repeated annually or bi-annually, with differences between years emphasized and the implication of these changes discussed in detail. Finally, this study’s sampling frame is restricted to California wineries. It would be useful to compare exporting, private labeling and postponement practices in different countries. For instance, France, Italy and Spain have a history of centuries, if not millennia, of wine production. Many European producers, even small family wineries, have been in business for generations and have built strong export programs. It is likely that these wineries may better utilize postponement as their greater emphasis on exporting necessitates specific labels and packaging for different markets. Additionally, European supermarkets sell more store-branded wines than their U.S. counterparts, and regional wineries would be likely to source private label products for these retailers. As both exporting and private label wine channels are expected to increase for U.S. producers, studying the practices of wineries in regions with more developed exporting

15

and private label channels may yield some insights. If postponement practices are indeed more widely utilized in these regions, it may behoove California wineries to consider the best practices of their peers abroad.

16

Appendix A: Survey Instrument 1. What size is your winery, in 9-liter case annual sales? (select one) Less than 5,000 cases 5,000 to 9,999 cases 10,000 to 24,999 cases 25,000 to 49,999 cases 50,000 to 99,999 cases 100,000 to 499,999 cases 500,000 + cases Exports 2. Does your winery export? 3. If so, do your exports require different labels/packaging? 4. Do you anticipate exporting in the next 5 years? 5. Do you anticipate that more California wineries will export in the next 5 years?

YES NO N/A YES NO N/A YES NO N/A YES NO No_Opinion

Other Brands Note: We define Private Label brands are those owned by a retailer or off-premise channel, such a supermarket’s or restaurant’s house brand. Co-labeled brands include the winery’s name.

6. Does your winery bottle multiple brands? 7. If so, are any of these brands private label or co-labeled wines? 8. Do you anticipate having private label or co-labeled brands in the next 5 years?

YES NO N/A YES NO N/A YES NO N/A

9. Do you anticipate that more California wineries will have private label or co-labeled brands in the next 5 years?

YES NO No_Opinion

Postponement Practices Note: We define postponement as holding wine at an unfinished stage to better meet market demands.

10. Does your winery engage in postponing the finishing of wine? YES NO N/A 11. What is the earliest stage of postponement? By earliest stage, we mean at what point do you first purposefully delay activities. (select one) Blending of wines Packaging (Bottling) Labeling Packing (Boxing/Palletizing) N/A- no postponement

12. Do you anticipate engaging in postponement in the next 5 years? 13. Do you anticipate that more Californian wineries will engage in postponement in the next 5 years?

YES NO N/A YES NO No_Opinion

Optional Identification Section. We realize that you may value your privacy, and may wish to return the survey anonymously. Please feel free to do so and skip the last question.

14. Would it be okay to contact you for follow-up questions?

YES

NO

If yes, please provide an email address:________________________________

17

References Anderson K., D. Norman and G. Wittwer. (2001) Globalization and the world’s wine markets: overview, Discussion Paper No. 0143, Center for International Economic Studies: Adelaide University, Australia. Basnet C., J. Corner, J. Wisner and K. Tan. (2003) Benchmarking supply chain management practice in New Zealand, Supply Chain Management: an International Journal, Vol. 8, No. 1. Carney J. (2008) Analysis of the labeling, exporting and postponement practices utilized by California winemakers, Master’s Thesis: San Francisco State University. Chaney IM. (2004) Own-label in the UK grocery market, International Journal of Wine Marketing, Vol. 16, No. 3. Cholette S. (2009) Mitigating demand uncertainty across a winery’s sales channels through postponement, International Journal of Production Research, Vol. 47, No. 13. Fisher C. (2008a) 2008 Tasting room survey report, Wine Business Monthly, 15 June. . Fisher C. (2008b) 2008 Closure survey report, Wine Business Monthly, 15 June. . Gordon J. (2008) Selling face to face…direct sales survey, Wines & Vines, . Orr S. (1999) The role of capacity management in manufacturing strategy: experiences from the Australian wine industry, Technology Analysis & Strategic Management, Vol. 11, No. 1. Olhager J. and E. Selldin. (2004) Supply chain management survey of Swedish manufacturing firms, International Journal of Production Economics, Vol. 89, No. 1. Popp J. (2003) Private label makes its mark, Beverage Industry, Vol. 94, No. 6. Pregler B. (2008) Winery equipment survey, Wine Business Monthly, 15 March. . Sánchez, A.M. and M.P. Pérez. (2005) Supply chain flexibility and firm performance: a conceptual model and empirical study in the automotive industry, International Journal of Operations and Production Management, Vol. 25, No. 7.

18

Simatupang T.M. and R. Sridharan. (2004) The collaboration index: a measure for supply chain collaboration, International Journal of Physical Distribution and Logistics Management, Vol. 35, No. 1. Skjoett-Larsen T., C. Thernoe and C. Andresen. (2003) Supply chain collaboration: theoretical perspectives and empirical evidence, International Journal of Physical Distribution and Logistics Management, Vol. 33, No.6. Twede D., R.H. Clarke and J.A. Tait. (2000) Packaging postponement: a global packaging strategy, Packaging Technology and Science, Vol. 13, No. 1. Van Hoek R.I. (1997) Postponed manufacturing: a case study in the food supply chain, Supply Chain Management, Vol. 2, No. 2. Van Hoek R.I. (1999) Postponement and the reconfiguration challenge for food supply chains, Supply Chain Management, Vol. 4, No. 1. Van Hoek R.I., B. Vos and H.R. Commandeur. (1999) Restructuring European supply chains by implementing postponement strategies, Long Range Planning, Vol. 32, No. 5. Ward M., B. Shimshack, J.P. Perloff and J.M. Harris. (2002) Effects of the private label invasion in food industries, American Journal of Agricultural Economics, Vol. 84, No. 4. Wines and Vines. (2006) 2006 Wines and Vines Directory and Buyer’s Guide, Wines and Vines, Inc.: San Rafael, California. Wine Institute (2008a), Number of California wineries, .

18

January.

Wine Institute (2008b), U.S. wine exports, 95 percent from California, reach $951 million in 2007, 29 February. . Yang B., N.D. Burns and C.J. Backhouse. (2004) Management of uncertainty through postponement, International Journal of Production Research, Vol. 42, No. 6. Yang B., N.D. Burns and C.J. Backhouse. (2005) An empirical investigation into the barriers to postponement, International Journal of Production Research, Vol. 43, No. 5. Yu C.S. and H.L. Li. (2000) A robust optimization model for stochastic logistic problems, International Journal of Production Economics, Vol. 64, No. 1. Zinn W. and D.J. Bowersox. (1988) Planning physical distribution with the principle of postponement, Journal of Business Logistics, Vol. 9, No. 2.

19