A Comparison of Domestic and International Joint Ventures in Television Program and Feature Film Production

Colin Hoskins (University of Alberta)

Stuart McFadyen (University of Alberta)

Adam Finn (University of Alberta)

Abstract: International co-production /co-venture is increasingly the mode of production used for television programs and feature films. Evidence suggests the principal benefit of this mode is that, through financial pooling, it permits the accumulation of substantial budgets. However, there are concerns that cultural specificity, the Canadian nature of the resulting program or film, is often compromised. Domestic co-production /co-venture may be an alternative mode that offers the advantages of the international joint venture mode without compromising the Canadian nature of the final cultural product. In this paper we examine this issue by comparing participant experiences with domestic joint ventures to those for international joint ventures. We find that most benefit and drawback items are rated similarly. Nevertheless, there is considerable evidence that for TV projects the performance of domestic joint ventures is rated better than that of international joint ventures. However, respondents also indicate that the domestic joint venture mode is not a viable alternative to the international joint venture mode for large budget projects.

Résumé: Les co-productions / affaires en participation internationales sont de plus en plus les modes de production utilisés pour les émissions de télévision et les longs-métrages. Les circonstances suggèrent que le bénéfice principal de telles approches est que, grâce au partage des fonds, elles permettent l'accumulation d'un budget significatif. Certains craignent cependant qu'on compromet la spécificité culturelle-particulièrement la nature canadienne-de programmes ou de films produits de cette manière. Les co-productions / affaires en participation domestiques peuvent présenter une alternative qui offre les avantages des collaborations internationales sans compromettre le caractère canadien du produit culturel final. Dans cet article, nous examinons cette question en comparant les expériences de co-productions domestiques avec celles de co-productions internationales. Nous découvrons que la plupart des avantages et des désavantages sont comparables. Néanmoins, il y a beaucoup d'indices que, dans le cas de projets télévisuels, les collaborations domestiques sont plus avantageuses que les internationales. Cependant, les personnes sondées indiquent que les co-productions domestiques ne sont pas préférables aux co-productions internationales dans le cas de projets à grand budget.

International joint venture (I JV) is increasingly the mode of production used for television programs and feature films. The Canadian government has encouraged this trend by negotiating co-production treaties with 24 countries. Evidence suggests the principal benefit of this mode is that it permits the accumulation of substantial budgets; however, there are concerns that cultural specificity, the Canadian nature of the resulting program or film, is often compromised (Hoskins & McFadyen, 1993). Domestic joint venture (DJV) may be an alternative mode that offers the advantages of the I JV mode without compromising the cultural specificity of the final product. In this paper we look at this issue by comparing the experience of partners in Canadian DJVs with the experience of Canadian partners in I JVs.

Specific questions to be explored include the following. Are the benefits gained from DJVs and I JVs similar? Are the drawbacks encountered similar? Is partner cheating more of a concern for one mode than the other? How do participant perceptions of project performance, in creative /artistic as well as financial terms, differ? To what extent is the DJV mode a viable alternative to the I JV mode?

Objectives

We set out to extend our knowledge regarding the advantages and disadvantages of DJVs compared to I JVs. More specifically, the objectives are:

  1. To examine how joint-venture participants rank the importance of benefits received, drawbacks encountered, and degree of success in both creative /artistic and financial dimensions.
  2. To compare the perceptions of participants in DJVs to those of participants in I JVs with respect to the level of importance of the various benefits received, drawbacks encountered, and performance achieved.
  3. To examine under what circumstances the DJV mode is a viable alternative to the I JV mode.

Methodology

A survey was the instrument chosen to address these questions. The first step in the design of the questionnaire was to identify the possible benefits and drawbacks. An initial listing was drawn from the general international business literature on the advantages and disadvantages of the international joint-venture mode (see, for example, Contractor & Lorange, 1988; Kogut, 1988; Hennart, 1988; and Beamish & Banks, 1987). After an examination of the Canadian broadcasting and film trade literature, these listings were adapted, where necessary, to apply to TV program and feature film production. A draft questionnaire was then tested in interviews with Canadian producers with experience of joint ventures and with the Director of Telefilm Canada's London Office.

In the summer of 1992, two versions, one relating to a DJV project and one to an I JV project, of a revised questionnaire were sent to each of the 188 member firms of the Canadian Film and Television Production Association. The questionnaires were translated into French and sent to the 97 members of the Association des Producteurs de Films et de Télévision du Québec. In each case a letter of support from the director of the trade association was included. Another 30 were sent out in the fall of 1992 to producers not listed as members of either association but included in Telefilm Canada's Directory of the Canadian Film, Television and Video Industry. Thus a total of 315 questionnaires were mailed. Our covering letter was addressed to the contact listed in the relevant directory but we asked that, if necessary, the survey be re-directed to the most appropriate person in the organization. Follow-up letters were sent to non-respondents.

Canadian production companies were asked to fill out the I JV questionnaire if they had completed an international co-production or co-venture between 1983 and 1991. If they had not undertaken an I JV but had undertaken a DJV they were asked to fill out the DJV questionnaire. This was done to maximize the number of responses relating to I JVs, I JVs being the focus of the larger overall program of research. However, it is a limitation of this DJV/ I JV comparison study because the DJV responses are restricted to producers who have not undertaken an I JV between 1983 and 1991. As most larger producers will have participated in an I JV, these will be excluded from the DJV sample.

As some producers were known to have done more than one joint venture, the questionnaires specifically sought information on the experience with the most recent project for which the outcome was substantially known. Eighty-two producers indicated that they had not undertaken any joint venture and 15 questionnaires were returned address unknown. Of the remainder, 22 producers provided usable responses to the DJV questionnaire and 38 to the I JV questionnaire.

In order to see if responses are related to the type of project, the questionnaire asked for categorizations of the project's industry (TV; Film; or Two-Versions, both TV and feature film versions), language (English; French; or Double, both English- and French-language versions), genre (Documentary, Drama, Animation, or Other), and the respondent's role (Majority or Minority). The breakdown of responses is shown in Table 1. As can be seen, TV, English-language, Drama, and Majority role were the most common responses.

Table 1 Frequency Distributions for DJVs and I JVs by Project Type
DJV I JV
Industry TV 20 25
Film 2 12
Two Versions 0 1
Total 22 38
Language English 19 27
French 1 9
Double 2 1
Other 0 1
Total 22 38
Genre Documentary 4 7
Drama 9 22
Animation 1 4
Other 8 0
Missing 0 5
Total 22 38
Role Majority 15 27
Minority 6 11
Missing 1 0
Total 22 38

For each benefit and drawback item, respondents were asked to indicate the level of importance of the actual benefit received or drawback encountered on a five-point Likert scale where 1 = no importance and 5 = very important. Similarly, a Likert scale was used to report the degree of success on various performance dimensions.

The responses were coded and statistical tests undertaken. First, for all joint ventures, the rank order of the level of importance of the various benefits, the various drawbacks, and the different dimensions of performance were examined. Here, attention focuses on the ranking of the specific items within each area of outcomes and whether differences between the items are significant. As this analysis compares pairs of item ratings provided by the same set of respondents (i.e., repeated measures data), paired t-tests were employed.

Next, to identify differences between the experience of DJV partners and I JV partners for each item, we compared the mean rating provided by each group of respondents. As each of these comparisons involves independent samples, an unpaired t-test was used for each comparison. The basic comparison was on an "All Projects" basis whereby the mean for all DJV projects was compared with the mean for all I JV projects. However, the comparison was also undertaken by project type where the number of observations was sufficient to make this meaningful. This is useful because it may show that the All Projects results are primarily driven by a particular type of project or may provide interesting contrasts between project types, for example, TV versus Film. Sufficient observations were collected to permit analysis for TV, Film, English-language, Drama, Non-drama (which includes Documentary, Animation, and Other), Majority, and Minority projects.

Hypotheses

In this section we examine the possible benefits and drawbacks associated with joint ventures and, where appropriate, make hypotheses with respect to differences in importance for DJVs relative to I JVs.

Benefits

(a)
Pooling of financial resources (abbreviated to "Financial" in Tables 2 and 3)

The main cost of a TV program or film is the expense of "the first copy." Extra copies for reaching additional markets can be produced at minimal incremental cost. This makes working with partners and pooling of finances to accumulate the budget for production of "the first copy" very attractive. Hoskins, McFadyen, Finn, & Jackel (1995) found this to be the most important benefit for participants in I JVs between Canada and Europe. It is no doubt possible to raise more funds in an I JV and we would thus anticipate that this item would be less important for a DJV. Therefore, we put forward the following hypothesis: #Hypothesis 1: The DJV mean will be less than the I JV mean for pooling of financial resources.

(b)
Access to partner government's incentives and subsidies (PartGov Subs)

Many foreign governments and Canadian provinces provide incentives and subsidies to television and film production. If a project is structured so that it counts as indigenous content in the market of each partner, it will be eligible for any partner government subsidies. Official co-production agreements facilitate this. The significance of this item will depend more on whether the partner government is one that provides subsidies rather than whether it is a foreign government or another Canadian provincial government. Hence, we have no reason for supposing the means would be different.

(c)
Access to partner's market (Part Mkt)

I JVs are likely to provide improved access for several reasons. First, the partner is likely to have better knowledge regarding the distribution process in his / her domestic market and better connections to key players. Second, the partner will have superior knowledge of the attributes demanded by viewers in his / her market and can help ensure the program possesses such attributes. Third, where quotas are in effect, a treaty co-production will enable the project to qualify as domestic product. These factors would not appear to be as important for DJVs where the partner also operates within Canada. #Hypothesis 2: The DJV mean will be less than the I JV mean for access to partner's market.

(d)
Access to third market (Third Mkt)

The partner may enjoy superior knowledge regarding the distribution system of the third-country market and better connections with key players in it. Viewers in the third-country market may demand similar attributes, language is one such attribute, to those in the partner's market. We would expect this to be more important for I JVs. Foreign producers are more likely to bring different skills and knowledge in this regard. #Hypothesis 3: The DJV mean will be less than the I JV mean for access to third market.

(e)
Access to a particular project initiated by partner (Part Proj)

On occasions the main motivation for a joint venture may be to access a unique project. There is no reason for suspecting the importance of this benefit will differ according to whether the joint venture is domestic or international.

(f )
Cultural benefits (Cultural)

Television programs and feature films are cultural goods and producers of them may well have non-pecuniary goals. We anticipate cultural benefits from working with a foreign partner would be greater because of the greater cultural distance between the partners. #Hypothesis 4: The DJV mean will be less than the I JV mean for cultural benefits.

(g)
Access to desired location (Location)

I JVs may facilitate access to a desired foreign location. Within Canada, we would anticipate fewer impediments to cross-border location shooting and hence would expect this factor to be less important. #Hypothesis 5: The DJV mean will be less than the I JV mean for access to desired location.

(h)
Access to cheaper inputs (Cheaper Inputs)

With Canada being a low-cost producer, attracting many U.S. runaway productions (see Gasher, 1995), we would not anticipate this being an important benefit for either DJVs or I JVs and no reason for expecting a difference in means.

(i)
Learning from partner (Learn Mktg, Learn Prod, and Learn Mgmt)

Learning opportunities may be anticipated if the partner has greater experience in program development and marketing (Learn Mktg), in the program or film production process (Learn Prod), or in general management (Learn Mgmt). The relevance of the partner's knowledge is likely to be greater when the partner operates in the same country. #Hypothesis 6: The DJV mean will be greater than the I JV mean for learning from the partner with respect to development and marketing, the production process, and general management.

Drawbacks

(a)
Increased co-ordination costs (Co-ord Costs)

These are the costs of negotiating the original deal, co-ordinating the production and distribution of the international co-production /co-venture project, and associated administrative burdens. It would seem likely that co-ordination would be more difficult and costly for projects involving a foreign partner. #Hypothesis 7: The DJV mean will be less than the I JV mean for increased co-ordination costs.

(b)
Increased shooting costs (Shooting Costs)

If both partners are involved in the shooting, perhaps with shooting in both partner locations, this will likely increase costs. Again we would expect this to be a more important drawback in I JVs. #Hypothesis 8: The DJV mean will be less than the I JV mean for increased shooting costs.

(c)
Loss of control and cultural specificity (Control Loss)

A joint venture inevitably involves compromises concerning the character of the program and the creative talent employed. Some producers (and regulators) may be concerned that the cultural integrity of the program produced is undermined. This concern is likely to have both non-monetary and monetary aspects. It may be that the compromised program, far from appealing to viewers in both markets, actually appeals to neither. This would seem likely to be more of a problem for I JVs because of the greater cultural distance between the partners. #Hypothesis 9: The DJV mean will be less than the I JV mean for loss of control and cultural specificity.

(d)
Increased costs of dealing with government (Gov Red Tape)

This is likely to be more important for I JVs as these involve partners and governments in different countries. #Hypothesis 10: The DJV mean will be less than the I JV mean for increased costs of dealing with government.

(e)
Opportunistic behaviour by partner (Cheating)

There is the possibility that the partner may "cheat" by underallocating resources to the joint venture in the hope of a "free ride," or by providing misleading information regarding the level of costs or revenues to be shared. We suspect the risk of cheating, or at least misunderstandings and perceptions of cheating, are greater between partners in different countries where often industry practices will differ. #Hypothesis 11: The DJV mean will be less than the I JV mean for opportunistic behaviour by partner.

(f )
Creating a more formidable competitor (Create Competitor)

If the partner learns from collaboration, the expertise gained may make him / her a more formidable competitor in the future. We would anticipate this is a greater potential problem for DJVs where the partner is also a domestic competitor. #Hypothesis 12: The DJV mean will be more than the I JV mean for creating a more formidable competitor.

In summary, we hypothesize that many of the benefits of a joint-venture mode will be less for DJVs than I JVs (learning being an exception) but many of the drawbacks will also be less (creating a more formidable competitor being an exception).

Table 2 Ranking of Benefits, Drawbacks, and Performance of Co-venture Mode for TV and Film (Paired t-test)
Item mean greater than those
Item Mean listed below at 5% At 10%
Benefits
Financial 4.44 PartGov Subs, Part Mkt,
Cultural, Third Mkt, Part Proj,
Location, Cheap Inputs, Learn
Mktg, Learn Prod, Learn Mgmt
PartGovSubs 3.49 Cultural, Third Mkt, Part
Proj, Location, Cheap Inputs,
Learn Mktg, Learn Prod,
Learn Mgmt
Part Mkt 3.29 Cultural, Third Mkt, Part
Proj, Location, Cheap Inputs,
Learn Mktg, Learn Prod,
Learn Mgmt
Cultural 2.45 Learn Prod, Learn Mgmt
Third Mkt 2.41 Learn Prod, Learn Mgmt
Part Proj 2.41 Learn Prod, Learn Mgmt Learn Mktg
Location 2.38 Learn Prod, Learn Mgmt
Cheap Inputs 2.36 Learn Prod, Learn Mgmt
Learn Mktg 2.09 Learn Mgmt Learn Prod
Learn Prod 1.85
Learn Mgmt 1.65
Drawbacks
Co-ord Costs 3.17 Shooting Costs, Gov Red
Tape, Control Loss, Cheating,
Create Competitor
Shooting Costs 2.64 Cheating, Create Competitor Control Loss
Gov Red Tape 2.47 Cheating, Create Competitor
Control Loss 2.33 Create Competitor
Cheating 2.07 Create Competitor
Create Competitor 1.66
Performance
Artistic 4.10 Overall, Proj Recoupment,
Co Financial
Overall 3.85 Proj Recoupment, Co Financial
Co Financial 3.27 Proj Recoupment
Proj Recoupment 2.80

Ranking of benefits and drawbacks

Ranking of benefits and drawbacks was undertaken using observations for all 60 joint ventures; the means reported in this section thus combine the observations for I JVs and DJVs. Financial pooling (Financial) is the most important benefit of joint ventures, its mean of 4.44 on the five-point scale being greater than all others at the 5% level of confidence (see Table 2). The next most important benefits are access to partner market subsidies (PartGov Subs = 3.49) and access to partner's market (Part Mkt = 3.29) which are greater than the means of the remaining items at the 5% level of confidence. Cultural benefits (2.45), access to a third market (Third Mkt = 2.41), access to a partner's project (Part Proj = 2.41), location (2.38), cheaper inputs (2.36), and learning from the partner re development and marketing (Learn Mktg = 2.09) are all more important than learning about the production process (Learn Prod = 1.85) and learning general management (Learn Mgmt = 1.65).

The greatest drawback is increased co-ordination costs with a mean of 3.17, greater than all other drawbacks at the 5% level of confidence. Next come increased shooting costs (2.64), increased costs of dealing with government (Gov Red Tape = 2.47), and then loss of control and cultural specificity (Control Loss = 2.33) and opportunistic behaviour by the partner (Cheating =2.07). Creating a more formidable competitor (Create Competitor = 1.66) is least important. Loss of control, cultural specificity, and opportunistic behaviour by the partner are less important than might be expected in light of the prominence given these items in the trade literature (see Hoskins & McFadyen, 1993).

Comparison of DJV and I JV level of benefits and drawbacks

As explained in the Methodology section, the basic comparison is on an All-Projects basis. The overall DJV and I JV means are reported for each benefit and drawback item in Table 3 and significant differences are noted. Sufficient observations also permit a comparison of DJV and I JV means for TV, Film, English-language, Drama, Non-drama, Majority, and Minority projects; these means for project types are only reported where there are significant differences. We examine these results in the context of the hypotheses made.

Table 3 Actual Benefits, Drawbacks, and Performance: DJV and I JV Participant Responses Compared (Unpaired t-test)
All Projects Industry Language Genre Role
English Drama Non-drama Majority Minority
Item Meansa TV Meansa Film Meansa Meansa Meansa Meansa Meansa Meansa
Benefits
Financial 4.52 : 4.40
PartGov Subs 3.71 : 3.37
Part Mkt 3.24 : 3.32
Cultural 2.70 : 2.32 2.78 > 1.82 3.17 > 1.91
Third Mkt 2.48 : 2.38
Part Proj 2.59 : 2.30
Location 2.82 > 2.13 2.85 > 2.08 3.00 >1.89
Cheap Inputs 2.64 : 2.19
Learn Mktg 2.57 > 1.82 2.58 > 1.84 2.56 > 1.67
Learn Prod 2.36 > 1.54 2.40 > 1.72 2.00 > 1.18 2.32 > 1.54 2.44 > 1.43 2.33 > 1.48
Learn Mgmt 2.14 > 1.37 2.15 > 1.52 2.00 > 1.08 2.05 > 1.37 2.44 > 1.32 2.00 > 1.22
Drawbacks
Co-ord Costs 3.09 : 3.21
Shooting Costs 2.62 : 2.66
Gov Red Tape 2.38 : 2.51
Control Loss 2.23 : 2.40 2.10 > 2.76
Cheating 2.00 : 2.11
Create Competitor 1.57 : 1.71
Performance
Artistic 4.36 : 3.95 4.45 > 3.76 4.39 > 3.73
Overall 4.18 > 3.65 4.30 > 3.60 4.21 > 3.67 4.15 > 3.55 4.33 > 3.67
Co Financial 3.77 > 2.94 3.95 > 2.96 3.95 > 3.12 4.00 > 2.90 3.62 > 2.73 3.73 > 2.92 4.33 > 3.00
Proj Recoupment 2.96 : 2.70
Experienceb 2.41 : 2.26 1.50 < 2.17
a
Except for All Projects, means are only reported (DJV first) when the difference is significant at the 5% level of confidence () or 10% level of confidence ().
b
Note that Experience is measured on a three-point scale whereas all other items are measured on a five-point scale.

Results re benefit hypotheses

(a)
Pooling of financial resources (abbreviated to "Financial" in Table 3)

#Hypothesis 1: The DJV mean will be less than the I JV mean for pooling of financial resources. This is not supported at the 5% or 10% level of confidence. Although the difference in means is statistically insignificant, it is actually the DJV mean that is greater.

(b)
Access to partner government's incentives and subsidies (PartGov Subs)

As expected, no significant difference in means is found.

(c)
Access to partner's market (Part Mkt)

#Hypothesis 2: The DJV mean will be less than the I JV mean for access to partner's market. No significant difference is found.

(d)
Access to third market (Third Mkt)

#Hypothesis 3: The DJV mean will be less than the I JV mean for access to third market. No statistical difference is found but the DJV mean is actually larger.

(e)
Access to a particular project initiated by partner (Part Proj)

As expected, no statistically significant difference is found.

(f )
Cultural goals (Cultural)

#Hypothesis 4: The DJV mean will be less than the I JV mean for cultural benefits. This hypothesis is not supported. For All Projects there is no statistically significant difference while the DJV mean is actually greater than the I JV mean for English-language projects (means of 2.78 and 1.82 respectively) at the 5% level of confidence and for Minority projects (means of 3.17 and 1.91 respectively) at the 10% level of confidence.

(g)
Access to desired location (Location)

#Hypothesis 5: The DJV mean will be less than the I JV mean for access to desired location. Not only is this hypothesis not supported but for All Projects access to desired location is more important for DJVs (mean 2.82) than I JVs (mean 2.13) at the 10% level of confidence. The same difference is also significant for English-language projects at the 5% level of confidence and TV projects at the 10% level of confidence.

(h)
Access to cheaper inputs (Cheap Inputs)

As expected, no statistical difference is found.

(i)
Learning from partner (Learn Mktg, Learn Prod, and Learn Mgmt)

#Hypothesis 6: The DJV mean will be greater than the I JV mean for learning from the partner with respect to development and marketing, the production process, and general management. This is supported for each learning item for All Projects at the 5% level of confidence. There is also significant support for many of the project breakdowns.

In summary, hypotheses that DJV means would be less than I JV means for pooling of financial resources, access to partner's market, access to third market, cultural benefits, and access to desired location are not supported. In fact for the latter two benefit items there is some evidence that the DJV mean is greater. The hypothesis that the DJV mean is greater for the three learning benefits is supported.

Results re drawback hypotheses

(a)
Increased co-ordination costs (Co-ord Costs)

#Hypothesis 7: The DJV mean will be less than the I JV mean for increased co-ordination costs. The direction of inequality is as expected but the difference is not statistically significant.

(b)
Increased shooting costs (Shooting Costs)

#Hypothesis 8: The DJV mean will be less than the I JV mean for increased shooting costs. No statistically significant difference is found.

(c)
Loss of control and cultural specificity (Control Loss)

#Hypothesis 9: The DJV mean will be less than the I JV mean for loss of control and cultural specificity. There is no statistically significant difference in means for All Projects although the direction of the inequality is as expected. At the 10% level of confidence there is support for the hypothesis for TV projects.

(d)
Increased costs of dealing with government (Gov Red Tape)

#Hypothesis 10: The DJV mean will be less than the I JV mean for loss of control and cultural specificity. This hypothesis is not supported although the direction of inequality is as expected.

(e)
Opportunistic behaviour by partner (Cheating)

#Hypothesis 11: The DJV mean will be less than the I JV mean for opportunistic behaviour by partner. This hypothesis is not supported although the direction of inequality is as expected.

(f )
Creating a more formidable competitor (Create Competitor)

#Hypothesis 12: The DJV mean will be more than the I JV mean for creating a more formidable competitor. This hypothesis is not supported. Although not statistically significant, the All Projects DJV mean is actually less than the I JV mean.

In summary, except for loss of control and cultural specificity, which for TV projects is less for DJVs than for I JVs at the 10% level of confidence, there is no significant difference between the means for the drawback items. However, although our hypotheses that most items would be less important for DJVs is thus not statistically supported, we note the direction of inequality is as expected.

Comparison of performance

Participants were asked to assess the degree of success for five performance measures. Four of the measures used a five-point Likert scale ranging from 1 = no success to 5 = great success. These measures are creative /artistic performance (abbreviated to "Artistic" in Tables 2 and 3), financial performance purely from the firm's perspective (Co Financial), the financial recoupment performance for the project as a whole (Project Recoupment), and the firm's perception of overall performance (Overall). A further measure, on a three-point scale, is also an indicator of overall success in terms of whether the experience with the project decreases, leaves unchanged, or increases the likelihood of the participant becoming involved in a future joint venture. This measure is labeled "Experience" in Table 3.

An issue that arises in self-reported, retrospective assessments of performance is hindsight bias. Bettman & Weitz (1983), in a management context, have shown that people have a tendency to attribute success to their actions and failure to environmental factors beyond their control. Whether this issue is relevant to this study is open to debate as respondents are not directly asked to account for success or failure. In any case, we see no reason to suppose that our comparison of DJV and I JV performance is affected by hindsight bias.

From Table 2, we see that of the five-point scale measures, creative /artistic performance (Artistic mean = 4.10) is ranked highest, followed by overall performance (3.85), and then financial performance from the company's perspective (Co Financial = 3.27). Financial recoupment performance for the project as a whole (Project Recoupment = 2.80) is ranked last.

We made no hypotheses regarding performance of DJVs and I JVs; however, the results reported in Table 3 indicate that performance is often rated better by DJV participants. For creative /artistic performance, the DJV mean is greater than the I JV mean for TV projects at the 5% level of confidence, while the same applies for non-drama projects at the 10% level of confidence. For All Projects, the creative /artistic mean is 4.36 for DJVs compared to 3.95 for I JVs but this difference is not statistically significant.

Financial performance from the company's perspective (Co Financial) is greater for DJVs at the 5% level of confidence for All Projects. This is a very strong result being repeated at the 5% level of confidence for TV projects, English-language projects, drama projects, Majority and Minority projects, and at the 10% level of confidence for non-drama.

On the other hand, the difference in means for financial recoupment performance for the project as a whole (Proj Recoupment) is not statistically significant although the All Project mean is greater for DJVs.

The overall performance is rated better for DJVs at the 5% level of confidence for All Projects, TV projects, and Majority projects, and at the 10% level of confidence for English-language projects and non-drama. This result is not repeated for the experience performance measure; the All Projects DJV mean is greater than that for I JVs but the difference is not statistically significant. Surprisingly, the only statistical difference for the experience performance measure is for film at the 10% level of confidence, where the DJV mean is less than the I JV mean.

In summary, there is considerable evidence that DJV performance is rated superior to I JV performance. However, these results seem to be driven by the predominant TV projects. For TV projects, the DJV means for overall performance, artistic performance, and company financial performance are all greater than the corresponding I JV means at the 5% level of confidence. In contrast, the only statistically significant performance result for Film relates to experience where I JV performance is superior.

When is the DJV mode a viable alternative to the I JV mode?

A distinguishing feature of I JVs is their greater budget. The average total budget for all I JVs in our survey is $3.9 million compared to $1.4 million for DJVs (differences in means significant at the 5% level of confidence using an unpaired t-test) while the budget-per-hour is $1.5 million compared to $0.8 million (differences in means significant at the 10% level of confidence). Thus I JVs typically have a total budget two to three times greater than DJVs while their budget-per-hour is almost twice as large.

Participants in I JVs were asked whether it would have been feasible to undertake the project as a DJV instead. If they answered no, they were asked for reasons. Twenty-nine respondents said the project could not have gone ahead as a DJV compared to only eight that indicated this would have been possible. The overwhelming reason given for a negative response was that a DJV would not have permitted the accumulation of a budget (financial pooling) large enough to undertake the project. This suggests that Canadian producers do not regard accumulating the budget from Canadian capital markets as a practical alternative. It seems likely that capital markets are less efficient than usual for an industry such as this which relies heavily on government funding and operates largely on a project by project basis.

Conclusion

Hypotheses that DJV means would be less than I JV means for pooling of financial resources, access to partner's market, access to third markets, cultural benefits, and access to desired location were not supported. For the latter two benefit items there is actually some evidence that the DJV mean is greater. The hypothesis that the DJV mean is greater for the three learning benefits was supported. Thus, contrary to our hypotheses for a number of the items, the evidence suggests that generally the level of importance attached to benefit items is similar for the two modes.

One possible explanation of the benefit item results could be the self-selection process which produced our samples of DJV versus I JV respondents. Larger producers will usually have undertaken an I JV during this period and hence will be excluded from our DJV sample. These larger firms may well have evaluated certain benefit categories for DJVs lower than for I JVs (as hypothesized) but the smaller firms, who may not have even been able to consider an I JV, may well have viewed the situation differently. For the smaller firms, market access, marketing skills, and even shooting locations provided by another Canadian partner may be just as important as objectively greater advantages achievable internationally by their larger brethren.

In the case of financial pooling, importance may be a decreasing function of respondent size as the larger the respondent company the more resources typically available to it. If this is the case, our sample of DJVs will value this benefit item more highly than would an unbiased sample.

Another consideration is that interprovincial hindrances may be greater than we anticipated, in fact comparable to intercountry barriers. For example, it may be considerably easier to sell a program to a provincial educational broadcaster if a partner from that province is in place.

Except for loss of control and cultural specificity, which for TV projects is less for DJVs at the 10% level of confidence, there is no statistical evidence that the degree of importance of drawback items differs between DJVs and I JVs. Our hypotheses that most items would be less important for DJVs is thus not supported. However, the direction of inequality for these items is as expected which suggests larger samples might have revealed some statistical differences.

One note of caution is necessary in interpreting the relative performance of DJVs and I JVs. There may be an element of sample bias as smaller producers, who are likely to have responded to our DJV questionnaire, are more likely to go out of business and hence our DJV responses may tend to reflect the experiences of the more successful ones who have continued to operate. However, taken at face value, there is considerable evidence that DJV performance is rated better than I JV performance. This result, which seems to be driven by the predominant TV projects, suggests that for TV projects producers would be well advised to choose the DJV mode rather than I JV mode whenever possible. There is one big qualification, though. For approximately 80% of the I JV projects in our survey, a DJV was not regarded as a viable alternative. Reasons given stressed the difficulty of raising the necessary funds through a DJV. Other evidence from the questionnaire responses suggests that the I JV mode is the mode of choice for high budget projects, the typical I JV having a total budget two to three times greater than a typical DJV while the budget-per-hour is almost twice as large. However, a question a producer contemplating an I JV might consider is whether a less ambitious version of the project organized instead as an DJV is possible and preferable.

Note

1
An earlier version of this paper was presented at the Canadian Communication Association Annual Conference, Montreal, June 1995. The authors wish to thank Francine Lavoie of Telefilm Canada's Montreal Office for translating the questionnaire that was distributed to Quebec producers. They also acknowledge the advice received from Robert Linnell, Director of the London Office of Telefilm Canada, and Deborah Drisdell, Manager, Co-productions, at the Montreal Office of Telefilm Canada.

References

Beamish, P. W., & Banks. J. C. (1987, Summer). Equity joint ventures and the theory of the multinational enterprise. Journal of International Business Studies, 18(2), 1-16.

Bettman, J. R., & Weitz, B. A. (1983). Attributions in the boardroom: Casual reasoning in corporate annual reports. Administrative Science Quarterly, 28, 165-187.

Contractor, F. J., & Lorange, P. (1988). Competition vs. cooperation: A benefit /cost framework for choosing between fully-owned investments and co-operative relationships. Management International Review (Special Issue on Co-operative Issues in International Business), 28, 5-18.

Gasher, M. (1995). The audiovisual locations industry in Canada: Considering British Columbia as Hollywood North. Canadian Journal of Communication, 20(2), 231-254.

Hennart, J. F. (1988). A transaction cost theory of equity joint ventures. Strategic Management Journal, 9, 361-374.

Hoskins, C., & McFadyen, S. (1993). Canadian participation in international co-productions and co-ventures in television programming. Canadian Journal of Communication, 18(2), 219-236.

Hoskins, C., McFadyen, S., Finn, A., & Jackel, A. (1995). Film and television co-production: Evidence from Canadian-European experience. European Journal of Communication, 10(2), 221-243.

Kogut, B. (1988). Joint ventures: Theoretical and empirical perspectives. Strategic Management Journal, 9, 319-332.



  •  Announcements
    Atom logo
    RSS2 logo
    RSS1 logo
  •  Current Issue
    Atom logo
    RSS2 logo
    RSS1 logo
  •  Thesis Abstracts
    Atom logo
    RSS2 logo
    RSS1 logo

We wish to acknowledge the financial support of the Social Sciences and Humanities Research Council for their financial support through theAid to Scholarly Journals Program.

SSHRC LOGO