VALUE FOR MONEY
AUDITING PUBLIC BROADCASTING PERFORMANCE IN
SWEDEN OLOF HULTÉN
Abstract
The four national broadcasters in Sweden are required by Parliament to present annually self audited public service reports, on how they fulfil their chartered service obligations. Reports have been submitted since 1997. An independent Broadcasting Commission reviews the reports to make sure they meet the needs of the ultimate recipients of the report: the government and the public at large. The purpose of the self-audited performance reports is two-fold: to stimulate public discussion about public service broadcasting and to be useful in the political evaluation of these services. The reports published so far have not led to much public discussion. Its first Parliamentary use will come when a special committee prepares a report to Parliament on the future of public service broadcasting for the next
charter period from 2006.
Olof Hultén is senior strategic analyst with Sveriges Television, e-mail olof.hulten@svt.se.
The history of performance auditing in most countries in Western Europe is as old as public service broadcasting. At regular intervals since its start, first radio and later radio and television, have been reviewed whenever the broadcasting licence was extended. Those where the windows of opportunities for parliamentarians to make changes. In Sweden, where radio was introduced in 1925, these intervals were ten years. From the 1960s, the periods of review became shorter. At present, public broadcasters operate on charters of four years, the most recent being 20022005. The reason given is that the speed of change in broadcasting is increasing, with new forms of services, new technologies and, since the advent of satellites, also increased competition.
The traditional form of performance review, the Parliamentary Committee aided by experts, has given way to reports from expert advisors to the government, joint Parliamentary working groups, and from 1997 also self auditing by public broadcasters themselves, as requested by law. There are also influences from other areas of civil services and cultural policies, as will be described below. In recent years, developments within the EU put pressure on governments in member states to guarantee that public broadcasting is clearly defined, is proportionally financed and that its performance can be reviewed by an independent national authority.
A Bird's Eye View of the Broadcasting Landscape
In Sweden, there are three different public service broadcasters with different remits, owned by the same public foundation. Each broadcaster has its own board of governors, staff and separate economy. The practical examples given in this article will come from Sveriges Television, SVT, responsible for national and regional public television. In addition, Sveriges Radio, SR, is responsible for national, regional and some local radio. Utbildningsradio, UR, is responsible for educational programmes, distributed by SVT and SR, as well as other supplementary material.
The radio and television market in Sweden can be characterised by a few key facts. There are 4 million TV households, out of which 70 percent have access to either cable, satellite or digital terrestrial reception, with anything from 10-40 channels. Basic must carry output on cable is the three national analog terrestrial channels, SVT1 and 2 plus private TV4. The remaining 30 percent of households seem satisfied with these universal channels. This share has decreased slowly but is surprisingly resilient since the early 1990s. Besides the three national channels, there are some Swedish satellite channels carried also by cable and one (Kanal 5) by digital terrestrial transmitters. The largest of them is TV3, reaching 64 percent of all viewers.
Digital television was first introduced by cable in 1997, by satellite and terrestrial a couple of years later. Today, 20 percent of the population has converted to digital, the majority by satellite. They are subscribers to pay TV services. The rate of conversion is slow, due to lack of affordable set-top boxes for non-subscribing viewers. Internet is available in the homes of 76 percent of the population. Connection to fast Internet service by cable, ADSL and broadband is growing and was available to 23 percent of the population at the end of 2002.1
SVTs domestic competitors are TV4, owned by the Bonnier publishing group, TV3 owned by the MTG group (owner and operator of satellite platform Viasat), and Kanal 5, owned by the SBS group. The audience share of SVT was in 2002 43
percent of the population as a whole, and about 35 percent, in multi-channel homes. SVT has launched two thematic digital channels, the Children's channel (December, 2002) and SVT24, news, current affairs and sports (February 2003). Some data might illustrate the market position of the four main competitors on the Swedish television market 2002.
Table 1: Audience, Content and Cost Share of Swedish TV Channels
	Audience share (%)	National content (%)	Programme cost (mill. SEK)
SVT	43,0	76	2,500
TV4	25,3	47	790
TV3	9,9	17	500 (est.)
Kanal 5	7,8	10	260 (est.)
As in other EU countries, public service broadcasters in Sweden struggle with increasing competition from the private sector and shrinking shares of young audiences. Pressures from both outside as well as internal ambitions demand new policies for digital services and programmes attractive to younger viewers and listeners (Hultén 2003).
History of Public Service Auditing
Public service broadcasting in Sweden is regulated by three different pieces of legislation: the Constitution, and its Fundamental Law of Freedom of Expression; the Radio and Television Act; and the Charter between public broadcasters and the government. The constitution forbids censorship; i.e. the State or public authorities cannot explicitly forbid or demand of mass media certain contents. The constitution, of course, does not allow certain forms of expression listed.
The Radio and Television Act, most recently revised in 1996, enables the government to demand certain conditions of broadcasting channels using terrestrial frequencies allocated for national services. Such conditions concern the character of news and information (impartiality and independence), the amount and forms of advertising (advertising can be and is denied public broadcasters), and stipulations about variety and pluralism of the total output. The Radio and Television Act delegates the authority to licence local radio broadcasting to the Radio and Television Authority. All other forms of wired broadcasting require no licence and need only to respect the Fundamental Law of Freedom of Expression.
The government licences the three national public services, plus national private TV4 with some public service obligations:
•	SVT, national and regional television,
•	SR, national, regional and some local radio stations,
•	UR, educational radio and television,
•	TV4, a privately owned national network with 16 mostly owned and operated
regional stations.
The conditions laid down by the Government in the charters are more detailed than in the Act. They specify for SVT, SR and UR that no commercial revenues can be used (except SVT, which may accept sponsors for certain live sports events) for
programmes broadcast in Sweden or for their web services. The charter specifies the range and scope of activities (universality, national reach by terrestrial networks, pluralism, minority services, services for handicapped people, regional presence, number of main channels, etc; see e.g. www.svt.se). The conditions for privately owned TV4 are fewer, but when it comes to volumes of output they are absolute (minimum level).
The legal and political tradition in Sweden makes it necessary to set up general goals and guidelines for broadcasters, focusing on the nature and character of services. Public broadcasters then are required to operationalise these goals and guidelines in productions and output profiles. Chartered broadcasters are scrutinized by the Broadcasting Commission, an independent authority, appointed by the Government. It reviews complaints about programmes from the public, and it reviews the broadcasters' performance against the broadcasting charters. All reviews of the Commission are ex post only. In specific cases, the Commission reports when it interprets a programme to be in breach of impartiality, when a programme broadcast does not follow the sponsoring rules or when it finds output service levels to be deficient. In 2002, the Commission received 677 complaints regarding SVT programmes. In 31cases the Commission upheld the complaint, most often on grounds of partiality and a few instances where SVT had not followed the sponsoring rules.
Since 1997, the four chartered national broadcasters are required to submit a special annual report to the Broadcasting Commission on how they performed the previous year in relation to their public service obligations. This annual report is called the public service account. SVT and the other three chartered broadcasters have submitted such accounts for one whole charter period (January 1997 to December 2001), and for the first year of the current period starting in January 2003, to December 2005.
Why Auditing of Public Service Broadcasting?
The traditional parliamentary method of public broadcasting was based on a political discourse, much influenced by cultural and moral values. Television viewing in itself was for a long time considered "stealing time" from other more worthwhile activities. The radio and television schedules were biased towards news, serious information and culture. A light music channel on radio was introduced only when broadcast "pirates" started commercial stations on board ships at high sea along the coasts of Sweden in 1961. The introduction of a second public television channel in December 1969 gave viewers an opportunity to zap between channels. Commercial television was allowed by parliament in 1992 when foreign satellite channels could reach Swedish homes by way of individual satellite antennas and not only via cable networks. According to the cable regulation from 1986, cable operators were until then obliged to filter out satellite channels "aimed at only Swedish viewers."
Private competition was introduced into Sweden gradually, beginning in the 1980s, first with quasi-commercial local radio stations posing as public access broadcasters, later by international satellite channels carried by cable and then by Swedish direct-to-home satellite services, in 1992 by domestic private terrestrial television broadcasters, and a year later by commercial local radio. It became clear to a growing number of observers inside and outside public broadcasting that there
was a need for a different set of "success" criteria for public broadcasting as compared to commercial channels.
A Danish academic researcher, Preben Sepstrup appears to have been the one to first introduce the concept of "public service auditing" in 1990, as part of a bigger research project at the university of Aarhus, called The Aesthetics of Television. Sepstrup was then also a member of the board of governors of Danmarks Radio, DR, the national public broadcaster. In a paper presented at the Nordic conference for Mass Communication Research in 1993, Sepstrup identified two reasons for such formalised criteria of public service accounting: as a managerial tool for nonprofit organisations and, as an instrument for media policy to differentiate public service broadcasting from private, for-profit, broadcasting. Ideally, he saw a need for similar reporting from private broadcasters as well in order to facilitate comparisons. The benefit of such comparisons would be to create an understanding of the costs and benefits of different forms of financing broadcasting. The pillars of the accounts were identified by Sepstrup as related to the various goals of public service broadcasting: output and schedules, audience use and appreciation, production and productivity, quality in the sense of pluralism and variety. DR Audience Research Department worked on the implementation of a standardised method using available statistics (Sepstrup 1993).
In Sweden, support for the idea of regular public service broadcasting accounting appears to have been inspired by two other non-academic, separate but related, discussions in the early 1990s. One was tied to the parliamentary review of state cultural policies and the other process was connected to the general study of productivity in the civil services. In 1993, a parliamentary committee had begun a review of policies in the cultural sphere, such as literature, theatre, film, music, art, and of the division of financial responsibilities between the central government and local communities. Twenty years earlier, in 1974, Parliament had passed the first comprehensive legislation on cultural policies. As the review committee reported its work in 1995, it put forward a proposal on public service broadcasting accounts, aimed at helping government to ascertain how these services meet their obligations.2
The same year, 1995, a permanent expert group (called ESO) within the Ministry of Finance studying the productivity of the public sector, published a report on the productivity of SVT (ESO 1995, 31). This report found that SVT, on the whole, was an economically sound operation, but the experts made a number of recommendations to improve the reporting formats and statistics of SVT. Among these were use of quality assessments of important productions and improved internal procedures of bookkeeping to facilitate analyses of efficiency. The ESO experts listed among their references a report from DR 1992 on public service auditing. "DR's model would facilitate analyses of productivity and goal achievement if applied to SVT" (ESO 1995, 136; author's translation).
In 1996, the Swedish government presented its bill to Parliament on the new charters for public broadcasters, to take effect on January 1, 1997.3 Among the proposals approved by Parliament was a new condition added to the charters: annual self-auditing of public service performance and goal achievement, to be reported to the Broadcasting Commission. In its discussion of this amendment, the government referred to the recommendation made by the parliamentary commission on
cultural policies mentioned above. It reiterated the kinds of statistics to be included: output volumes, costs and revenues, key indices on productivity and quality. This information should be relevant also from a managerial point of view, as well as from an external perspective. The government pointed out that this annual report would be useful as a stimulus for public discussion on public service. The formal status of the report was not, however, to be on a par with the formal closing of the books.
The Swedish government was thus first to formulate a formal obligation of public broadcasting to report public service accounts. It invited public broadcasters to present a suitable operational model of reporting. Since few of the public service obligations are expressed as specific volumes, shares or as detailed levels of activities, the government expected the accounting reporting to be flexible and open to change over time. Furthermore, the editorial and managerial integrity of the broadcasters must not be compromised. This open invitation required that the 1996 public service output be used as the reference year. Each successive year is compared to the year before and to the reference year.
The Broadcasting Commission reviews the audit and approves the report when it finds it sufficiently comprehensive. It is important to note what the Commission's role is, as specified by the government, to review the broadcasters self audit reports to make sure that the reports contain sufficient information for the government's and the public's evaluation of how the broadcasters fulfil their public service obligation. The Commission then communicates this approval to the government.4 Other ingredients of its communication with the government are of course the commission's own decisions on complaints received, its own studies of various cases and its commissioned research. Among the latter is an annual report from the university of Gothenburg on the content and schedules of all major Swedish TV channels (8 channels included for the time being).
Compared to SVT, private TV4 produces very simple reports, often compilations of titles of programmes purchased, produced and transmitted. The three public broadcasters SVT, SR and UR report independently of each other, but they are entitled to coordinate their services for minorities. So far, however, there is no formal cooperation or coordination of such services. In 2001, Parliament approved an amendment to the audit procedure: the information in the reports should be checked and validated by the broadcasters' own corporate auditors, before being submitted to the Broadcasting Commission.
The Structure of the SVT Self-Auditing
In 1995, when SVT began thinking of how to meet this new obligation, there were at least two factors to consider: the scale and scope of the self audit, and how to make it useful as a managerial tool, a planning instrument, for the company as well as a summary of activities of the year passed. In short, these two aspects boil down to what kinds of information and comparisons should be used, and what is the work and benefit involved? Since the remit of public broadcasting is expressed in rather vague terms, the first task was to sift through all relevant documents and find all mandatory obligations and operationalise them. The only specified measure given by Parliament is that at least 55 percent of original in-house productions (except news and events) must come from outside Stockholm, the capital. All other demands are relative: level of services must equal or be higher each year and com-
pared to the last year of the previous charter period (1996; at present 2001). There are no explicit demands made on viewing shares. There are no explicit demands on how the budget must be spent.
Internally, the annual public service reports to the Broadcasting Commission have resulted in the process of collecting non-financial statistics being made quicker. The report is to be presented two months into the new year. The auditing process also requires a review of definitions used: are all units in the corporation making the same operationalisation. For example, does "independent production" have the same definition everywhere and all of the time? Are resources used recorded identically everywhere and all of the time? Over time, changes in organisational structures and administrative routines usually generate demand for new sets of statistical information and continuity may be jeopardised.
The public service report from SVT is divided into three main parts: output; audience information; and use of resources. Output relates to channels and services, to programmes and their origins; audiences are described by their viewing of the programmes as well as what viewers say about SVT; use of resources relates to financial means, and costs of the programmes broadcast. Programmes are divided into seven main categories - news, factual and information, fiction/movies, music, entertainment, sports, and others, further divided into a number of sub-categories (a total of 54).
The core services of SVT are the two analog channels SVT1 and SVT2, the second channel also carrying regional news from 11 different news desks around the country. Internet services are offered on www.svt.se and extensive videotex service is available on the two national channels. Two digital thematic channels have been launched, Children's Channel and SVT24. The output and usage of these channels are not specified in the public service audit, because they are still available to only a minority of viewers (23 percent in May 2003).
On the scale and scope, the one extreme is: the reporting should be comprehensive and detailed; there is nothing to hide. The other extreme position is: report only the bare necessities, otherwise the autonomy and integrity of the company is at risk and too much openness will encourage successively more detailed instructions. The model chosen was to report fairly extensively and at the same time with due respect for the corporation's autonomy and managerial integrity.
SVT has until the writing of this paper submitted six public service reports. The report of 2002 marks the beginning of the new charter period. The format during the period of the first five years was kept the same, always with 1996 as the reference year. Some changes were made in 2002 (with 2001 as reference year), concerning the categorization of output, to make comparisons with other broadcasters easier.
On the whole, there has been little use of the reports by the media, by lobbying groups or by the public at large. This can be appreciated as a sign of approval, or alternatively perhaps as a sign of "don't really care." The amount of detailed statistics makes the audit reports rather inaccessible to the ordinary viewer or citizen. One of the ambitions formulated by the government for these reports is that they stimulate public discussion of public service broadcasting. As this has not happened, the government expressed its disappointment to Parliament in 2001 when the bill for the new public service charter period was presented.5
The very first test of the political usefulness of the material will be made in 2004, when a parliamentary commission on public service broadcasting gets to work, preparing the ground for the next charter (from January 2006). Parliament determined in 2001 that there had been no "proper" political review of public service broadcasting since the 1980s. Traditionally, parliament and successive governments have wanted to achieve solid majorities behind legislation and regulation of public broadcasting. Over time, this has become more difficult, because the structure and operations of broadcasting grow more and more complicated. New distribution platforms, new electronic services and new technologies present themselves. In this expanding industry, public service broadcasting has to define its role anew. Parliament wished to be part of a thorough and broad debate on the issue. At the time of writing, the Ministry of Culture is preparing the remit of the committee's work.
Obviously, this parliamentary committee will be using the self-auditing reports from the public broadcasters, as well as input from the Broadcasting Commission, other official bodies and its own review to report to parliament on the future of public service.
Quality Indicators
Quality is a complex and complicated concept in television. Parliament asked the public service broadcasters to develop indicators to describe quality of programmes and services, and not only quantities. Efforts are made by researchers in many countries to make a contribution to this area (e.g. Ishikawa 1996) but there are no simple procedures ready to be used. Such research is time consuming, and thus expensive to apply. Results yielded depend mainly on interpretations made case by case by individual observers, and thus they are difficult to generalise.
SVT is using three different procedures, none of which is original. One is the use of in-house, quality-enhancing feedback channels for planning and production. A second one is external professional evaluation, i.e. prizes at festivals. A third procedure is to track what viewers think about SVT services. Programme contracts are developed for each major programme project, specifying target groups, content and resources used. This is followed by evaluation of what was achieved. A test method is developed to find out more about how different audiences perceive and react to a range of programmes, to new programme formats, styles of presentation, etc.
Regular surveys are commissioned by SVT to describe viewers' attitudes toward public service programmes and their image of SVT. Public service output scores highest on news, children's programmes, culture, current affairs and documentaries, but has lower scores on entertainment, sports, movies and foreign fiction. This is quite a normal pattern for most public broadcasters. The image scores of many public broadcasters, and also of SVT, is high on trustworthiness, professionalism, quality, but lower on exciting and innovative. SVT has in later years made many changes in schedules, programmes and formats, as well as in presentation, graphic design and marketing. Table 2 illustrates how survey respondents have reacted.
Table 2: The Image of SVT in the Eyes of the Viewers
Item	Percent of viewers agreeing that an item properly describes SVT 2002 1996	
Trustworthiness	91	81
Professional	83	69
Quality	82	71
Variety	82	77
Important to all	78	64
Entertaining	63	54
Innovative	46	31
Exciting	41	30
Source: SVT (Annual surveys from University of Göteborg, Dept of Journalism and Media Studies.)
Other Reviews of Public Service Performance
There are no comparisons to other channels or broadcasters in the public service reporting from SVT. The interested reader who wants to compare SVT to other services will have to use other sources, such as the Broadcasting Commission. Its annual report on what is on all the Swedish free to air channels reflects the much greater variety on SVT.6
The Broadcasting Commission publishes every year a study carried out by the University of Göteborg on the total output from 8 Swedish television channels. It presents measures on volumes of hours, channels profiles, scheduling on prime time and indices of pluralism and variety. While the SVT self-audit is based on total broadcast hours, the Göteborg study is based on a selection of 17 weeks per year. It resembles the German studies by Udo Krüger of the Institut für empirische Medienforschung in Köln, reported annually in MediaPerspektiven (see e.g. MediaPerspektiven 2002, 10, 512-530).
Eleven content categories are divided into a total of 134 subcategories. One summary from the 2001 report is included to illustrate the profiles of the analysed channels, according to orientation: informative-entertaining (Figure 1). The two public service channels SVT1 and SVT2 score the highest total diversity in both informative and entertaining genres. The changes between the two years of SVT1 ands SVT2 reflect adjustments of their schedules; a number of programmes were moved from one channel to the other, so the total score of SVT output was unaltered.
The two sets of data on SVT output can be compared to each other. The SVT output audit represents every minute broadcast during 2002, classified by the internal unit for audience and programme analyses. In total, 10 131 hours were broadcast on SVT1 and SVT2 that year. The output analysis made for the Broadcasting Commission by the University of Göteborg represents about a third of that volume (3 268 hours from 17 weeks), classified with the help of programme titles. As table 3 shows, there is a lot of similarity but also difference. The SVT study divides programmes for special target groups, children, youth and minorities, according to genre. Mixed genre talk shows are classified as a subgroup of entertainment by SVT but get their own headline in the university study.
Figure 1: Diversity of Program Profiles 2000-2001 (indexed)
Great diversity
100 _,
	1 SVT1 * SVT2 TVd U v
_ t*fvs . TV6 J ZTi 1 1 1 1	v ytoral 5 \ TV3 1 1 1 1
_Little diversity_Entertainment_Great diversity
Source: http://www.grn.se/PDF-filer/info/Svenskt%20TV-utbud%202001.pdf, p. 66 (TV6 and TV8 are thematic channels).
Table 3: TV Output Classification of SVT1 and SVT2 (2002)
	SVT %	Göteborg Uni. %
Factual	27	24
News	26	24
TV fiction/Film	21	17
Music	4	-
Mixed format	-	7
Entertainment	7	6
Sports	12	10
Children/Youth	-	10
Minorities	-	2
Other	2	2
Total	99	102
Volume, hours	10 131	3 268
Other =presentation, continuity, self promotion		
Source: SVT Public service report 2002, Table 6b, www.svt.se; Svenskt TV-utbud 2002, Table 2, page 94. Broadcasting Commission, Report no 11, May 2003, www.grn.se.
Conclusion
The information about what is broadcast on Swedish television channels is today fairly substantial and it allows comparisons over some period of time, although genuinely comparative data are available only since the late 1990s. Comparisons between channels are possible thanks to the study carried out by the University of Göteborg. In SVT's self audits, output and costs are reported, making possible the tracking of productivity, as well as how obligations in the broadcast license are met, such as the origin of programmes. SVT is mandated to produce 55 percent of general programming (excluding news and sports) outside the capital, to acquire 10 percent from independent producers and increase services to various special target groups (hard-of-hearing, ethic and language minorities).
Measures of programme quality are difficult and costly to produce. Output quality as perceived by the viewers can and is measured and included in the SVT self audit. Safeguards to improve quality are included in the production and planning processes, both through increased professional competence and improved audience research. One ambition of SVT is to harmonise the ex post auditing for external use with ex ante managerial and planning usage. As Preben Sepstrup (1993) noted when he was developing his idea of public service audits for Danmarks Radio, the audit is useful to the organisation if it is incorporated into the planning and production processes of the broadcasting organisation.
There are good reasons for developing performance auditing for public service broadcasting, not least because those who pay the license fee (or tax payers) are entitled to know how the money is used. There are also good reasons for a careful and respectful approach to this need: it is difficult to create media quality or quality media by law or regulation. The value of public service broadcasting is the end product of several factors: resources, talents, professional experience, leadership and integrity, as well as organisational autonomy.
The present model of public service self auditing introduced in 1996 fits into the Swedish tradition of ex post-evaluation, the results of which is fed into the legislative process, which periodically produces revisions of the Radio and Television Act and sets the parameters for the public service charters of the following period. Annually, the government reports to parliament about the financial status of public broadcasting and proposes the necessary license fee each television household has to pay, as a result of inflation.
There are no immediate sanctions if the audit reports show changes between individual years. Increases or decreases of individual measures can have many causes, some of which are beyond the control of public broadcasters. Over a whole period, steady decline of e g minority services could result in an explicit demand to improve the situation, if the particular service is required. SVT has managed to keep the levels of output stable or slightly higher each year in comparison to the reference years. The exception was the volume of original domestic Finnish-language programming offered in 2002, it was slightly lower than during the previous charter period.
One advantage of self-reported audits of public service is that the goals and guidelines, the operationalised remit as it were, are internalized by the organisation. The first years, a special group of staff experts were involved in the process and the report was then signed by the chief executive of the organisation before being sent to the Broadcasting Commission. Gradually, the process has become
more closely integrated with the planning and managing process of SVT. Relevant managers are being made aware of what is required of them to meet the goals for the year and for the charter period. By referring to the audit report internally and externally, the chief executive is giving it a higher status than before.
Whether Sweden will, in the future, have public service charters in the shape of annual contracts such as in some other countries, is hard to predict today, but it seems less likely. The nature of broadcasting as a mixture of creative cultural processes, industrial production processes in many respects and as a vehicle of societal opinion formation, makes it very difficult to encapsulate in a contractual agreement based on simple indices.
Political traditions tend to change rather slowly. It seems more likely that the licence fee might be abandoned in Sweden as a source of financing public service broadcasting, with everything else left more or less intact, rather than that the present structure would be dramatically changed in character or in quantity. This could be explained by the weakening support for the licence fee as such among Swedes in general, rather than a weakening support for the idea of a comprehensive national public broadcasting service universally available. The opinions about the services from SVT and the use of them are more positive than the attitude towards the licence fee.
At the same time, however, viewers are less and less tolerant of advertising breaks on commercial channels. And only a minority opt to subscribe to pay television. Perhaps we find an Internet logic developing here: high demands on the services but let somebody else pay the bill. The logic is impossible in the long run. The only solution for public service broadcasters is to convince the audience(s) to finance the service one way or another. The focus is on the value for money. The public service audit is instrumental in keeping the focus. To avoid externalizing editorial decisions into the hands of public accountants or politicians, self-auditing might serve that purpose best.
Notes:
1.	Mediemätning i Skandinavien, MMS, Basundersökning 2002:2. MMS information pertains to the situation in October-December 2002.
2.	Kulturpolitikens inriktning. Kulturutredningens slutbetänkande, SOU 1995:84
3.	En radio och TV i allmänhetens tjänst 1997-2001. Prop 1995/96:161.
4.	Registration no. 231/03-60. Decision of June 11, 2003. Available at www.grn.se.
5.	Available at http://www.grn.se/PDF-filer/info/Svenskt_TV-utbud2002.pdf.
6.	The most recent data for 2002 available at http://www.grn.se/PDF-filer/info/Svenskt_TV-utbud2002.pdf (in Swedish only).
References:
ESO. 1995. Kostnader, produktivitet och mäuppfyllelse för Sveriges Television AB. ESO, DS 1995:31
Hultén, Olof. 2003. The Future of Public Broadcasting in Scandinavia. In P Donges and M. Puppis (eds.), Die Zukunft des öffentlichen Rundfunks. Internationale Beiträge aus Wissenschaft und Praxis, 170-188. Cologne: Halem. Ishikawa, Sakae, ed. 1996. Quality Assessment of Television. London: John Libbey Media. Sepstrup, Preben. 1993. Skitse til et planleagnings - og regnskabssystem for public
servicevirksomheder med Danmarks Radio TV som exempel. XI Nordic Conference on Mass Communication Research, Trondheim, Norway, August 8-11.
0 S
3
a
o
£ m
CO
o