3.4 Case Study 1: European Film Distribution Office (EFDO)
 
Introduction
 
Recent years have seen a gradual equalisation in the number of films produced in Hollywood and in Europe. The characteristics of these films, their target-audience and their cost of production, however, are increasingly different. In general, mainstream Hollywood films are relatively expensive and marketed to attract large and young audiences. European films, on the contrary, are made with much smaller budgets and tend to appeal to smaller, more mature audiences. Although there are many examples of cross-overs between these two generalised classifications, the gradual confinement of European productions to art-house and niche markets is already a reality in most European countries.
 
Because of their niche market position, small European productions do not enjoy the same type of distribution as mainstream films. But they serve part of the market where public intervention can be used successfully to preserve the fruition of quality cultural productions. This section explores issues involved in the subsidised marketing and distribution of low-budget European films, and to this end analyzes the fortunes of European films which received support through EFDO.
 
A number of films were examined and analyzed to investigate whether there are some systematic factors in the way European films are distributed within the separate countries that may explain why some European titles are successful in some countries but received with only tepid enthusiasm in others. Although culture-specific factors will always contribute to success stories, they cannot be the sole explanation of widely-differing performances. The existence of dynamic domestic production, strong cultural ties with neighbouring countries, or a population's ability to speak more than one European language, may all contribute to the greater success of European films in some countries. In this paper we try to isolate these factors and concentrate on the key elements in the way films are distributed and released in each country, such as the marketing expenditure associated with each film, the number of prints made available and the length of the release.

 
The data
 
The analysis is based on data provided by the European Film Distribution Office (EFDO) relating to 50 European films for which distributors across Europe applied for loans to cover distribution expenses. Eligibility for EFDO loans is currently restricted to medium and low-budget European films (defined as having a budget less than 4.5 million ECUs). Countries covered in the period examined included all EU countries, plus Switzerland and Austria. In 1992, Switzerland left the programme. Of the EU countries, Luxembourg is subsumed for distribution purposes within Belgium. Likewise, distribution of films in Ireland is included with the UK.
 
The data cover the period 1988 to 1992. The data analyze each title country by country, showing the name of the distributor, the calculated and actual distribution costs, the number of prints (as estimated beforehand by the distributor), the date and length of release, box office receipts, rentals, admissions, number of cinemas showing the film, and the number of runs. The number of prints is split between original, dubbed and subtitled copies.
 
Only the titles for which complete data were available where retained in the database, which in its final version contains 50 films and 164 releases. Also, records containing data visibly out of line with the rest of the database and where some error was assumed either in the distributor's returns or in data entry by EFDO were eliminated. For example, the analysis of average ticket price for each film in various countries (box office receipts divided by admissions) led to the removal of a number of releases for which ticket prices varied significantly from the average for the country concerned. Such possible errors are often due to the difficulty of providing final figures for admissions before EFDO's reporting deadlines, which results in admission figures being at odds with reported box office.
 
Data from other sources were used to complement the sample thus constituted and enable comparisons to be made. Average ticket price and yearly admission figures in different countries were obtained from MEDIA Salles: European Cinema Yearbook, 1993. EFDO's 99 European Films was used to identify the country of origin of all films analyzed.
 
Table 41 below shows the main characteristics of the EFDO data sample. The first column indicates the number of film releases for each country. Since for all countries the base of films is relatively small, the average values reported throughout the study do not have full statistical reliability and are to be taken as indicative. The next three columns in the table show the characteristics of film release in each country:

The last two columns show the average rentals charged by distributors expressed as a percentage of total marketing expenses and as a percentage of box office receipts.
 
The sample shows significant differences among the various countries. Differences are only partially related to the population size of the countries and tend to reflect different national marketing practices. The number of cinemas showing a given film is a significant factor: the average for all countries was 68 cinemas, but the average figures may vary significantly even between countries of similar sizes. Within the small countries, for example, Belgium uses on average 19 cinemas, but Greece has 51 and Portugal has ten. Similarly, the number of prints specified by the distributors in each country is not consistently correlated with the size of the country or the number of cinemas where the films are released.
 
Table 41 
Characteristics of the Sample Used of EFDO-Supported Films
Country
Number of releases
Average cinemas
Average prints
Average weeks
Rentals as % of box office
Rentals as % of total marketing
Austria
15
8
2
8
36%
65%
Belgium
13
19
2
15
42%
72%
Denmark
10
14
2
18
35%
39%
France
19
139
15
25
50%
57%
Germany
26
186
17
43
43%
77%
Greece
3
51
2
n.a.
61%
67%
Italy
15
99
16
31
38%
49%
Netherlands
19
20
3
17
43%
51%
Spain
9
36
6
19
35%
71%
Portugal
13
10
2
9
39%
35%
Switzerland
11
34
7
34
35%
108%
UK
11
34
3
13
29%
56%
TOTAL
164
68
8
23
40%
62%
Source: London Economics' analysis of EFDO data/MEDIA Salles
 
Distributors apply to EFDO for loans to cover up to 50 per cent of "pre-costs": these include prints, subtitling, dubbing, art work, advertising and publicity materials, screenings and other promotional expenditure. The loans are issued against receipts and other information provided by the distributor and independently audited. Once the distributor has recouped his share of the pre-costs from rentals, he or she can deduct a further 30 per cent of the gross receipts to cover his or her overheads. 70 per cent of the remaining receipts is then applied to repay the loan. Therefore the distributor could be said to have broken even when rentals equal 50 per cent of total marketing pre-costs, although this includes no allowance for overheads.
 
Since rentals, on average, corresponded to 62 per cent of marketing costs, we can see that, taken as a whole, the distribution of these films was only viable thanks to the EFDO loans and a relatively small proportion of those loans are ever repaid. Our sample covers the early years of EFDO's activities as well as the more recent period and omits some of the most notable successes achieved by EFDO supported-films, such as Il ladro di bambini and The Crying Game. In the case of these and other successful films, rentals averaged nearly 180 per cent of marketing costs.
  
 
Methodology
 
The first step was to analyze the average figures country by country and gain an overview of the characteristics of each national market, in terms of average number of cinemas, number of prints, preference for subtitled or dubbed copies and so on. The second step involved the identification of successful films and the analysis of their marketing in the different countries. There are, however, several ways to define successful films. Should one look at the top earners in each country separately, or try to identify films that do well everywhere? Is a film "doing well" if it earns a lot of money, or if it attracts a lot of viewers?
 
The most obvious measure of success - net box office revenue - was rejected because the differences in ticket price and marketing costs in the various countries make it difficult to calculate "net" box office revenue and even more difficult to make comparisons. We chose to use admissions as the basic indicator, but needed to create some form of index that would serve two main purposes: rank the various films in order of success and make the different countries comparable.
 
The success index we used is given by the share of all admissions to EFDO films in the sample, obtained by each film, in those countries in which that film was released. This is calculated as the ratio of all admissions for a given film in the countries showing it, to the total admissions for EFDO films in those same countries. For example, let us suppose that Film A is shown in countries X, Y and Z and achieves admissions of 2,000 in country X, 1,000 in country Y and 100 in country Z. Admissions for all EFDO-supported films in our sample shown in country X is 4,000; in country Y, 6,000; and in country Z, 10,000. Therefore the film obtained total admissions of 3,100 out of a possible 20,000. The success index for Film A is 3,100 divided by 20,000, making 0.155. Film B is shown in countries X and Y only, and has total admissions of 2,500. So Film B has an index of 0.25 (2,500 divided by 10,000 total admissions in countries X and Y), better than Film A even though it achieved lower admissions. This method of measuring success overcomes the problem which would arise if we used a classic measure of success, such as total admissions (or even EFDO's own method, based on repayments of loans plus loans not claimed), namely that a French film may perform very well in France but not in other countries but, given the size of the French market (the largest in Europe), it would score highly both in overall admissions and in loan repayments (assuming that it repaid the loan for France, which was large, and did not repay the much smaller loans towards the releases in the other countries).
 
In order to make comparisons between different countries, we complemented the analysis with another indicator which highlights the relative performance of a given film in a particular country.
 
This "relative country performance" index is a ratio between the proportion of total admissions achieved in each country for a given film and the country's overall market share in Europe. For example, if Switzerland accounted for 10 per cent of Film A's total European admissions, and represents 10 per cent of the box office for all films in the countries where the film was shown, Film A in Switzerland would have an index of 1. If the Swiss box office for Film B accounted for 20 per cent, Film B would have an index of 2.
 
If the index for a film is greater than 1 in a given country, it will necessarily be below one for at least one of the other countries where the film has been shown. This is because the ratio only records how total admissions are split between the films in a given country and is not sensitive to the absolute level of admissions. Since a film will almost invariably do better than average in its home country, we eliminated the home-country records for each film in these calculations. Thus, the indicator we built allows us to examine the relative performance of the film in all the different countries where it is distributed, but is not a measure of the film's overall success.
 
 
The results
 
Using the first indicator of relative share among EFDO-supported films, then, we were able to highlight a number of films that have been successful in three or more European countries (Table 42).
 
Table 42
Top films by success index
Rank 
position		 Success index  
Ratio		   

Film title
No of 
Countries showing film
1 0.236 La Vie Est un Long Fleuve Tranquille
3
2 0.157 Der Philosoph
5
3 0.114 La Belle Noiseuse
7
4 0.109 Tilai
7
5 0.107 Fuglekrigen Kanofleskoven
3
6 0.097 Babette's Feast
3
7 0.087 Distant Voices, Still Lives
4
8 0.084 Ay, Carmela!
2
Source: London Economics' analysis of EFDO data/MEDIA Salles
 
This analysis produces results which are at odds with common-sense notions of what is and is not a successful film. EFDO's own definition is based on repayments of loans by distributors, or the extent to which distributors do not have to call on loans because of a film's success. When films are successful, and distributors do not call down loan facilities, they may not provide to EFDO full data on performance. Therefore, some films may be excluded from our sample, and the sample is consequently biased towards relatively less-successful films. Having said that, we note that La Belle Noiseuse is one of the most successful films supported by EFDO. A possible flaw in the repayment/non-claim definition is that a distributor might refund a loan when the very modest release he undertakes goes into the black, even though the level of admissions for the film is very low. One very successful film for EFDO, in terms of repayments but also admissions was Riff Raff, with more than 100,000 admissions each for France, Germany and Italy and 60,000 in Switzerland, but it does not appear in Table 42 which is biased towards films which play in the smaller markets.
 

Relative performance by country
 
Using the two measures discussed above, we conducted a two-fold analysis of films. First we separated success and flops, and then we looked at how the films have performed in the different countries. Some countries do consistently better than others, for example, as do some distributors. A few titles do well in all countries where they have been distributed.
 
When looking at the top films that do consistently well, the prominent titles are Der Philosoph (Germany), La Belle Noiseuse (France), and Tilai (Switzerland, France, UK, Italy, Burkina Faso).
 
Table 43 provides a country-by-country analysis of relative performance of EFDO-supported films, showing the average, minimum and maximum rates given by our market share index. Switzerland, Denmark and Portugal exhibit the highest ranking in the study. Switzerland in particular, shows a market share for all EFDO-supported films which is larger than we would expect given its size. The countries where European films tend to do as well as expected, that is to say where our index approaches one on average, are France, Germany and Spain. Finally, last place goes to the UK, showing the lowest average values of relative market share. The remaining countries in the sample do not show a definite trend.
 
It is particularly important, therefore, to concentrate on the countries that exhibit strong trends and investigate whether they show differences that can explain these trends.
 
Switzerland constitutes the best example of a country with a fairly consistently high rate of success for EFDO-supported films. One feature that makes Switzerland a particularly good market for European films is its multi-lingual population, which means that a large section of the audience can enjoy either French or German (or Italian) films in their original version. However, there was no bias towards French or German films among the films successfully released in Switzerland.
 
This analysis also points up the effectiveness of EFDO support in the smaller countries; on the whole, EFDO-supported films pick up a relatively-high proportion of their admissions in those countries. For example, while Greece represents a little over 1 per cent of cinema admissions in Europe, it accounts for around 2 per cent of admissions to the EFDO-supported films in our sample.
 

Table 43 
Relative performance of EFDO-supported films
Rank
 Country
Average 
Maximum
Minimum
1
 Switzerland
2.61
5.52
0.91
2
 Denmark
2.22
6.95
0.26
3
 Greece
1.88
2.86
1.1
4
 Portugal
1.86
5.67
0.12
5
 Netherlands
1.80
6.59
0.05
6
 Austria
1.69
9.53
0.01
7
 Spain
1.12
2.24
0.16
8
 France
0.98
1.83
0.21
9
 Germany
0.92
2.31
0.13
10
 Italy
0.86
1.79
0.03
11
 Belgium
0.84
1.58
0.25
12
 UK
0.65
3.55
0.12
All countries
1.37
9.53
0.01
  
 
Using our index of success to classify the different titles, we concentrated on the top and tail of the ranking to highlight common patterns. The first indicator we looked at in each country is the number of prints that are associated with successful and unsuccessful films. For all countries this number is higher for the more successful titles and decreases for the least successful ones. A higher number of prints is also often found in the countries where a film records a better relative performance. This is certainly the case for Switzerland, where the best features were circulated in 16 copies on average, and the least successful had three copies. As with many of the indicators we consider, we have not yet managed to sort out the causality: do Swiss distributors have the greatest ability to gauge the success of a film? Are distributors capable of gauging the popularity of a film and deciding, on this basis, exactly how many prints they are going to make? It is certain that, all things being equal, the more prints and the more spent on marketing, the more successful a film will be. Conversely, too small a number of prints will depress revenues. The significance of the number of prints has become greater over time as the tendency has developed towards wider releases and shorter runs.

Another important factor is the number of cinemas in which the films are released: this is also positively correlated with the success rate of EFDO-supported films. In Switzerland, successful films are released on average in 57 cinemas, and the average number of cinemas for all releases is 34. Compared to Austria, which has similar population size, Switzerland uses a greater number of cinemas for all films (34 compared to 8) and also retains films for longer (31 weeks against an average of 7 for Austria). The Swiss average marketing budget is approximately the same size as that for Austria.
 
EFDO-supported films in Denmark are shown in 14 cinemas for 18 weeks on average. The marketing budget for Denmark, the smallest country in our sample, is the eighth largest. Danish distributors use few copies of the films (a maximum of three) and take about 33% of box office income in rents. As in Switzerland, more successful titles have more prints of the film in circulation, and they are shown in a greater number of cinemas.
 
The only other country where EFDO-supported films tend to perform quite well is Portugal. This country appears to be "the odd man out" of our study. On average, films are shown in 10 cinemas and kept for 9 weeks. These figures are lower than those of Belgium and Greece, which have roughly the same population as Portugal. The average number of prints per film is two, and distributors' rent is 43.2% of net box office.
 
The redeeming feature that makes Portugal a successful market for EFDO-supported films is probably its ticket price, by far the lowest in Europe. The highly monopolistic market in Portugal, dominated by the distributor and exhibition chain, Lusomundo, may mean that those films which do enter theatrical distribution in Portugal face less competition than would be the European norm.
 
France and Germany are generally good, although not outstanding, at marketing EFDO-supported productions (see Marketing Budgets below). The high marketing budgets are used to organise a widespread release of films in an average of 139 and 186 cinemas, respectively. German exhibitors show EFDO-supported films for an average of 43 weeks, and their French counterparts for 25. In both countries the positive correlation between success rate and the number of prints and number of cinemas is observed.
 
Concentrating on the country with the worst record of successful European films, the UK, it is immediately apparent that European films are shown by very few cinemas in the country. On average, each film is shown in 34 cinemas (as many as in smaller Switzerland!) for 13 weeks, compared to 99 cinemas and 31 weeks in Italy and 139 cinemas for 25 weeks in France. The most successful European film in the UK, La Belle Noiseuse, attracted total admissions of slightly over 37,000. Compared to results for the top title in Italy, Laberinto de Passiones, which recorded 108,000 admissions, the British figure looks extremely low. The 1991 figure on annual cinema attendance per capita(20) for the UK is 1.8, which is lower than Switzerland (2.2) but still higher than the Italian average (1.5) and therefore does not help justify the low admission figures.
(20) from "European Cinema Yearbook", MEDIA Salles, 1993.
 
British distributors earn the lowest share of box office income in Europe: 29.3% on average. Comparisons with the two countries of similar size (France and Italy) also shows a significant difference in the average number of prints: three subtitled prints in the UK versus 15 in Italy and France, most of which are dubbed. Based on results from our sample, the UK also has the second most expensive cinema ticket.
 
 
Marketing budgets
 
We would tend to expect that variations in performance for a film in different countries should be related to variations in the marketing effort for the film. The greater the effort, the more likely the film is to be successful. Of course, different countries require different kinds of releases, with the larger countries requiring a larger-scale release than the smaller ones. But our analysis failed to demonstrate that, for a given country, a larger marketing spend meant a more successful film.
 
Table 44 shows the average, minimum and maximum marketing budget for each country. The "budget rank position" given in the first column shows countries' ranking according to the magnitude of their marketing expenses, thus allowing for a first instance comparison of the relative size of the budgets.
 

Table 44 
Marketing expenditure
Country
Budget rank position
Average marketing budget 
(ECUs)
Maximum marketing budget (ECUs)
Minimum marketing budget 
(ECUs)
France
1
132214
302743
44664
Germany
2
103180
270894
21768
Italy
3
91936
178081
51122
Spain
4
38615
85881
8826
Greece
5
35312
44996
23032
Switzerland
6
35087
103135
12128
UK
7
26952
59263
6582
Denmark
8
18246
29817
9269
Belgium
9
16945
40238
5690
Netherlands
10
14496
53512
6193
Portugal
11
11332
21052
3324
Austria
12
8295
15013
2905
TOTAL  
52802
302743
2905
 
France, Germany and Italy show the highest average marketing budget in the sample. Their minimum budgets are also among the highest, implying that high marketing costs in those countries are common to everyone and no distributor is able to market at the lower budgets found in other parts of Europe. The reverse applies to Austria, Portugal and Belgium, which are characterised by low overall marketing budgets. The three most expensive countries use the highest average number of prints in our sample (16). Advertising costs play a large role in explaining the marketing expenses in the UK, which are much higher than any other country with the same number of prints.
 
The type of print also has an impact on the level of the marketing expenses. Relative prices may vary greatly among countries, even for the same film and similar order sizes, as pointed out in the first EFDO report(21). Most countries use a mix of dubbed and subtitled copies. Denmark, the Netherlands, Portugal, Switzerland and the UK only use subtitled copies. Italy is the only country in our sample to use dubbed prints exclusively.
(21) EFDO report, "1 June 1988 to 31 December 1989, Pilot Phase of the European Film Distribution Office"
 
When comparing marketing expenses, it is useful to take into account population size. Three of the largest countries in the sample are also the top spenders. This highlights the peculiar position of the UK, the second largest country by population and the third largest theatrical market, but only the seventh largest in terms of marketing budget.
 
A similar point can be made about Austria (less than 8 million inhabitants), which has the lowest expenses overall, when compared to the other small countries like Switzerland and Denmark ranking sixth and eighth respectively. The low marketing expenses in Austria, however, can be explained by the fact that distributors often use the same material and campaigns used in Germany, so that Austrian national marketing represents a relatively small additional cost to the marketing expenses in Germany.
 
When we compare the average marketing budgets and number of prints in our study with those described as "ideal" in the first EFDO report, we find a mixed picture. With the exception of Italy and France, all countries use fewer prints than prescribed by EFDO's pilot study. Whereas France and Spain's marketing budgets are very close to their ideal levels, Greece and Italy use much higher marketing budgets and print numbers. Belgium, Portugal, the Netherlands and the UK are all below their ideal budget levels. There is no "ideal" data for Austria and Switzerland, and figures for Germany only relate to the western part.
 
 
Conclusions
 
The analysis of the two most extreme cases in our data set has clearly shown that aside from national characteristics that may make a country relatively more (or less) receptive to European films, the distribution patterns adopted in the country have a role in shaping its ability to market European films successfully. In Switzerland, as we discussed, three European languages and high per capita rate of attendance are not enough to explain this country's superior performance. Conversely, the UK has no national feature that could explain the poor performance of European productions.
 
The study shows very clearly that, although no single factor can overwhelmingly condition the outcome of European films in a given country, some aspects of the distribution process can combine to improve or hinder such outcome. In particular, the availability of a reasonable number of prints and the widespread exhibition in national cinemas strongly influence a film's chances of being successful.
 
These results must, however, be put in a sobering context. In Table 45 we compare, for each country, the average box office for all films, the average box office of non-domestic European films, and the average box office for the films in our sample.
 
With the exception of Greece and Germany, the performance of EFDO-supported films is well-below that of non-domestic European films (i.e. European films outside of their country of origin). And non-domestic European films perform well-below the average film (i.e. including US and national productions).
 
It would not be commercially viable to release most of the films supported by EFDO were it not for that EFDO support. But no amount of support at the distribution level will compensate for an insufficiently commercial production and, by the standards of domestic successes, such as national comedies, and international blockbusters, no European film in a considerable time has had the necessary commercial attractiveness. But this is not an argument against European film-making: European films produced and marketed by the US majors tend to perform just as well or better than the majority of US titles; one thinks of A Fish Called Wanda, produced by MGM and distributed by UIP.
 
Table 45 
Average Box Office, All films, Non-Domestic European films and EFDO Sample Films (in 000 ECUs)
 
Average box office (1992) - all films
Average box office (1992) - non-domestic European films
Average box office - EFDO sample films
 
Austria
206
n.a.
15
Belgium
308
176*
29
Switzerland
255*
212*
108
Denmark
312
42
20
Spain
844
584
78
France
1,554
559
151
Germany
1,586
219
185
Greece
139*
34*
39
Italy
806
248
118
Netherlands
220
43
17
Portugal
180
57
10
UK
1,682
119
52
* London Economics' estimates
Source: London Economics' analysis of MEDIA Salles/BIPE and EFDO data
 
The logic of EFDO - and of any scheme to support the circulation of European films in Europe - is to make up for the diseconomies associated with the niche market of specialised film: high print costs, advertising costs concentrated on a relatively narrow release, the expense of dubbing and subtitling for small markets. At the margin - but only at the margin - EFDO support makes possible bigger numbers of prints and higher advertising budgets than would otherwise be commercially possible.
 
In order to improve fundamentally print availability and advertising support, far higher levels of subsidy than those contemplated by EFDO or sanctioned by the MEDIA programme, would be necessary. On the print side, it is precisely this kind of subsidy that the French and German governments have instituted; no public authority has yet intervened to provide comparable assistance for film marketing, with the possible exception of some of the German Länder.
 
From the exhibitors' standpoint, a scheme like EFDO does improve the availability of prints to a degree (as does various MEDIA Salles and Europa Cinémas initiatives) and probably enhances the performance of the films in commercial and, in the case of the UK, non-commercial (i.e. publicly-supported) cinemas alike. But it is unlikely that it improves the commercial attractiveness of the films to the extent that an exhibitor would contemplate booking a film who did not have a prior commitment to programming and winning audiences for non-mainstream films.
 
In other words, it does not improve the competitiveness of non-domestic European films relative to the US alternatives. To improve competitiveness, the scheme would need distributors and exhibitors to negotiate lower rentals, encouraging them to book films for longer runs despite lower capacity utilisation. But this strategy would be expensive and fruitless if it did not translate into greater acceptance by cinema-goers' of non-domestic European films. This would be a cultural change which would lead not only to larger audiences for those films, but higher levels of cinema-going overall.