Economic performance of the Spanish regional public service broadcasters according to the political party in government: An empirical analysis (1995-2016)

ISSN 2386-7876 – © 2019 Communication & Society, 32(4), 159-170 159 Economic performance of the Spanish regional public service broadcasters according to the political party in government: An empirical analysis (1995-2016) Abstract This article analyses and quantifies the influence that the political party ruling the regional government has upon the economic performance of the corresponding public service broadcaster. An analysis of the economic performance of regional broadcasting companies was undertaken over the period from 1995 to 2016, studying variables such as expenses per employee, total expenses, total assets, ROA, assets turnover and total share; and their relationship with the political party ruling in the region. The correlation analysis, based on the Krukal-Wallis test, confirms the political parties' influence on the broadcasting entities' management approach. The results obtained indicate that there are significant differences in economic performance depending on the political party (Partido Popular, Spanish Socialist Workers Party and nationalist-regionalist parties) which governs the Autonomous Community, and suggest research areas of interest in order to examine in greater detail the impact of these parties in each case. Keywords Governance, regional television, public television, political party, economic performance. 1. Introduction Regional public service broadcasters in Spain originated with the passing of Spanish Law 46/1983, of December the 26 th , which regulated the third television channel at the beginning of the 1980s. Since their creation, they have contributed significantly to the democratic development of Spanish society. In addition, they have helped invigorate local cultural industries and generated economic development in their respective regions (Fernández Quijada, 2009; Guimerá & Blasco Gil, 2012). However, these public entities have been called into question regarding various issues such as their high cost, the independence of their content with respect to the corresponding regional political government, and their management of production activities (Muñoz, 2011; Campos, 2012; Artero et al., 2015; Olmo & Navarro, 2015). In addition, audiences have decreased significantly over time. From the beginning of the 90s until the year 2016 viewership figures declined significantly, from an audience share of 15.6% to 7.4% (Kantar Media, 2016). In other words, there has been a decrease of more than 100%. Víctor Orive Serrano https://orcid.org/0000-0001-6467-7414 [email protected] Universidad de Zaragoza Miguel Marco Fondevilla https://orcid.org/0000-0003-2717-8389 [email protected] Universidad de Zaragoza Pilar Latorre Martínez https://orcid.org/0000-0002-8486-6885 [email protected] Universidad de Zaragoza Submitted December 21st, 2017 Approved March 26th, 2019 © 2019 Communication & Society ISSN 0214-0039 E ISSN 2386-7876 doi: 10.15581/003.32.4.159-170 www.communication-society.com 2019 – Vol. 32(4) pp. 159-170 How to cite this article: Orive Serrano, V., Marco Fondevilla, M. & Latorre Martínez, P. (2019). Economic performance of the Spanish regional public service broadcasters according to the political party in government: An empirical analysis (1995-2016). Communication & Society, 32(4), 159-170. Orive Serrano, V., Marco Fondevilla, M. & Latorre Martínez, P. Economic performance of the Spanish regional public service broadcasters according to the political party in government: An empirical analysis (1995-2016) ISSN 2386-7876 – © 2019 Communication & Society, 32(4), 159-170 160 Numerous investigations carried out in different knowledge areas such as journalism, audiovisual communications, sociology and media economics have looked at regional public service broadcasters in Spain over recent decades. In the latter area –media economics–, there is a growing body of knowledge which analyses and outlines the influence of the economic policies of regional governments on the behaviour and economic performance of regional public service broadcasters (Casado, 2005; Miguel de Bustos & Casado, 2012; Marzal et al., 2015; Casado et al., 2016). However, these studies do not differentiate based on the orientation of the political party which governs the Autonomous Community and which, as such, oversees the production activity of the public television entities. In light of this situation, the principal aim of this study is to analyse and quantify the influence that the political party which oversees the regional television networks has on their behaviour and economic performance. The empirical study covers a period of time with several ruling governments. The paper is structured as follows: Section 2 consists of a review of the literature concerning the governance and economic management of regional public service broadcasters in Spain. Following this, the variables under study and the characteristics of the sample are outlined. The results are presented in Section 4. Finally, Section 5 covers conclusions, limitations and future areas of research. 2. Governance and economic management of regional public service broadcasters in Spain The regional broadcasters are public companies whose management and administration falls under the control, through different instruments and mechanisms, of the regional government which is democratically elected by the citizens of the Autonomous Community where the broadcast entity is located (Fernández Alonso & Fernández Viso, 2012). Since their launch in the early 1980s, during a period of development and consolidation of democracy in Spain, these companies have contributed to regional economic development by revitalising the local audiovisual industry and creating jobs (Fernández Quijada, 2009; Guimerá & Blasco Gil, 2012). At present, once the Valencia regional entity concluded activity, 12 of the 17 Autonomous Communities count on, at least, one regional public service television channel. The governance style of public service broadcasters can be shaped by means of four different models: governmental, professional, parliamentary and corporate (Hallin & Mancini, 2008). The first model –governmental–, is characterised by the governing political party exercising strict and substantial control over the management of the television network. In order to exercise this control, the political parties appoint managers whose loyalty to the party is more important than their professional competence. Radio Televisión Española (RTVE, Spanish Radio Television) used this governance model until the passing of Spanish Law 17/2006, of June the 5th, which deals with state-owned radio and television. The professional model involves managing the public service entity in a very different way. Professionals with extensive experience in the sector are responsible for their management. From this perspective, responsibilities relating to audiovisual monitoring, approval and regulation are assigned by the political parties to experts that are considered impartial and that have extensive knowledge and competence in the area of audiovisual production. This is the case with the public service broadcasters in the United Kingdom and the United States. There are also notable differences regarding management between the parliamentary and corporate models. The corporate model is characterised by the existence of a diverse array of people who play a part in the management of the broadcast services. It is considered a very democratic model (Humpreys, 1996) in which various groups (professional, societal, etc.), along with representatives of the ruling political parties, analyse and make joint decisions regarding production at the public service network. Public service television in Orive Serrano, V., Marco Fondevilla, M. & Latorre Martínez, P. Economic performance of the Spanish regional public service broadcasters according to the political party in government: An empirical analysis (1995-2016) ISSN 2386-7876 – © 2019 Communication & Society, 32(4), 159-170 161 Holland has opted for this model of management. Finally, in the parliamentary model, management decisions are taken by means of negotiations arising from the Parliament where the network is located. Thus the political parties, according to their representation in parliament, make the decisions regarding the management of the television network. This is the model which governs the activity of Spanish public service broadcasters at national (RTVE) and regional level: Euskal Irrati Telebista (EiTB), Corporación Catalana de Medios Audiovisuales (CCMA), Corporación Radio y Televisión de Galicia (CRTVG), Radio y Televisión de Andalucía (RTVA), Radio Televisión Madrid S.A, Radio Televisión Canaria, Radiotelevisión de Castilla-La Mancha (RTVCM), Ens Públic de Radiotelevisió de les Illes Balears (EPRTVIB), Radiotelevisión del Principado de Asturias (RTPA), Corporación Extremeña de Medios Audiovisuales (CEXMA), Corporación Aragonesa de Radio y Televisión (CARTV) and Radiotelevisión de la Región de Murcia (RTRM) (Fernández Alonso & Fernández Viso, 2012). Various theoretical and empirical studies have analysed the economic management and administration of regional broadcasters in Spain over recent decades. In order to do this, researchers have analysed different variables and used different methodologies and approaches. Some of the variables that have been considered in the literature are the professional, business and political profile of the directors, presidents and boards of directors of the regional entities (López Cepeda, 2012), the input-output ratio necessary to achieve production efficiency (Campos & Velasco, 2013; Artero et al., 2015), the influence of the management model (classic or outsourced) used by the channel on the economic performance obtained (Fernández Alonso, 2002; Cuesta, 2011; López Vidales et al., 2012; Orive et al., 2016) and the effects that regional public policies have on the activity of regional broadcasters (Miguel de Bustos & Casado, 2012; Marzal et al., 2015; Casado et al., 2016). The research conducted by López Cepeda (2012) analysed the professional profiles of the managers of regional broadcasters using different variables and indicators. A questionnaire was sent to more than 200 members of the boards of directors of the networks. The results of the study show significant differences in these profiles depending on the regional public service entity analysed. In the research undertaken by Campos and Velasco (2013), they determined by means of Data Envelopment Analysis (DEA) that certain regional broadcasters are not efficient in the management of their resources. These networks include the regional public service entities of Aragon, Valencia, Galicia, the Canary Islands, Asturias and Madrid. Later, Artero et al. (2015) confirmed these results and proposed some performance measures in order to address the inefficiency in production management based on variation of inputs. To this end, they suggested that certain networks needed to considerably reduce their external expenses and implement actions aimed at increasing their audience share. The literature has addressed the effects of the management model (classic or outsourced) used by regional broadcasters in order to manage production (Fernández Alonso, 2002; López Vidales et al., 2012). Sánchez and Sarabia (2012), from a theoretical perspective, pointed out that the outsourcing of part of the network's value chain can lead to a reduction in costs and debt at the public service broadcaster. However, this management model can result in a loss of internal capital and asymmetric relations within and outside the broadcast entity (Cuesta, 2011). Recently, Orive et al. (2016) used an empirical approach to determine that the management model implemented influences the economic performance and the productivity levels achieved by the regional broadcasters. As such, broadcasters that utilise a management model based on outsourcing part of their value chain require fewer assets and public subsidies for each audience share point they achieve. In addition, these networks have less financial losses. With regard to the influence of the regional administrations' economic policies on regional broadcaster activity, the work of Miguel De Bustos & Casado (2012), which included contributions from numerous researchers, highlighted the inefficiency of certain policies Orive Serrano, V., Marco Fondevilla, M. & Latorre Martínez, P. Economic performance of the Spanish regional public service broadcasters according to the political party in government: An empirical analysis (1995-2016) ISSN 2386-7876 – © 2019 Communication & Society, 32(4), 159-170 162 which hinder the sustainability of various television networks. The authors thus underscored aspects related to governance, budget control and political independence in production decisions. Later, the work of Marzal et al., (2015) analysed the causes and consequences of the deep economic and political crisis which many regional broadcasters found themselves in. The authors emphasise that significant changes in the networks' management policies were necessary in order to increase the credibility of the entities and improve the results of the networks. Recently, research carried out by Casado et al., (2016) analysed the impact of decisions to reduce public spending on regional broadcasters. The results of their research quantified these effects, showing a significant loss of employment and a considerable reduction in the turnovers of the companies in the audiovisual sector. However, from the review of the literature, it can be seen that the economic and political behaviour of the regional broadcasters is not homogeneous. In fact, in recent years, certain networks have had to shut down (for example, the Valencia broadcasting entity), lay off staff (Telemadrid) or they have seen a notable decrease in their audience (Canal Sur). However, other networks have consolidated and increased their audience (Canal Extremadura) or improved their financial performance (Aragón TV). Moreover, the literature has not analysed the influence of the political party’s orientation on the economic management of the broadcasting networks. Three different groups of political parties have governed and managed regional administrations during recent decades. The Partido Popular (PP), ideologically conservative and supportive of liberal economic policies, the Spanish Socialist Workers Party (PSOE in the Spanish acronym), which has a progressive and social democratic ideology, and the nationalist-regionalist parties (NAC) of each Autonomous Community. To the latter group belong the Basque Nationalist Party (PNV), Convergencia i Unió (CIU), Foro Asturias and Coalición Canaria (CC). All of these are ideologically conservative and have a strong sense of their own regional identity. 3. Variables under study and sample characteristics Audiovisual corporations and, in particular, television broadcasters are complex organisations in terms of their economic and financial structures. In order to analyse and evaluate the economic performance of public service broadcasters, the literature distinguishes between three classes of variable (Picard, 2003; Orive et al., 2016). These variables are: market share, financial variables and productivity measurement variables. Data relating to operating income, total assets, number of employees, expenditure on material, expenditure on personnel, economic performance and the aggregate audience share of the corporation's channels have thus been collated as variables under study. Given this starting point, the study of the economic performance of the audiovisual corporations and broadcast entities over time presents significant limitations and challenges, related to their particular economic structure and the impact that subsidies from the public sector have on their income and economic performance. Furthermore, over the two decades under research there have been major changes in the field of accounting (the adoption of the new General Accounting Plan of 2007, for example), organisational restructuring and changes in ownership, and different taxation schemes (pro-rata VAT, for example), among others. These changes have had an obvious effect on the annual accounts of the organisations studied, especially with regard to figures dealing with income and profits. In addition, the worldwide economic and financial crisis of 2008-2012 had a significant impact on broadcast corporations and entities, especially in Spain, both as regards the amount of subsidies and income received, and employee numbers and volume of expenditure. Finally, it is difficult to study the economic and financial behaviour of the country's different audiovisual corporations taken on an equal footing, without bearing in mind significant socio-political considerations such as the number of potential viewers, the Orive Serrano, V., Marco Fondevilla, M. & Latorre Martínez, P. Economic performance of the Spanish regional public service broadcasters according to the political party in government: An empirical analysis (1995-2016) ISSN 2386-7876 – © 2019 Communication & Society, 32(4), 159-170 163 political and cultural mandate upon which they were founded and to which they must respond, and other particular conditions that affect only some of the organisations, such as broadcasting in various languages. The problems outlined have been taken into account when designing the methodology to be used in the proposed analysis, and their influence has been minimised by the decision to focus on relative trends instead of on absolute values. In order to study the impact that the particular political orientation of the regional government has had on economic performance, it was decided to incorporate the relative year on year trend changes for each variable under study, such as percentage increase or decrease, thus eliminating the distorting effect of absolute values. Specifically, the main aspects of the methodological design employed are the following: • The use of the variables ‘result’ and ‘income’ as variables under study was discarded, replacing them with ROA and Asset Turnover ratios, which better facilitate the comparative study of year on year trends. • The analysis of the relative trends for each organisation is incorporated, calculating the positive or negative percentage change observed year on year, for each of the variables under study. In this way the effects of legislative changes, accounting practices and the different organisational structure of the organisations, on the variables selected for study (expenditure, number of employees, the ratios previously mentioned) is limited. • In the same way, given the impossibility of determining what percentage of the corporations' total costs, employees, assets and income corresponds solely to the broadcast entities, the year on year study of relative trends allows for assessment of the evolution of these variables in each organisation, assuming that the aforementioned percentage does not change significantly. • The ‘audience share’ variable has been taken into account in aggregate form, as the total of all the networks present in each region. The analysis of the aforementioned variables covers the period between 1996 (the first year for which sufficient data is available) and 2016 (the last year for which complete data is available). A longitudinal analysis was performed, covering more than 20 years for the entities created in the final decades of the last century and the first decade of the current century (see Table 1). The data was obtained by way of the Annual Accounts of the 13 regional public service broadcasters and completed using the SABI (Iberian Balance Sheets Analysis System) economic and financial database. From a statistical point of view, each year for each corporation represents one case in the sample, linked to the dependent variable 'political orientation in the regional government' of that year (PP (1), PSOE (2) or Regional/ Nationalist (3)), such that the total sample covers 214 cases. Political orientation being the dependent variable, defined as the political party in the regional government in each year, the variables under study for each of these years and audiovisual corporation/ television broadcasting entities, are the following, shown in Table 2. Orive Serrano, V., Marco Fondevilla, M. & Latorre Martínez, P. Economic performance of the Spanish regional public service broadcasters according to the political party in government: An empirical analysis (1995-2016) ISSN 2386-7876 – © 2019 Communication & Society, 32(4), 159-170 164 Table 1: Year the regional public service broadcaster was created. Organisation Commencement of broadcasting Political Party in government No. of cases Euskal Irrati Telebista 1982 1996-2009 NAC. 2010-2012 PSOE 2013 - 2016 NAC. 21 Corporación Catalana de Medios Audiovisuales (CCMA) 1983 1996-2002 NAC. 2003-2009 PSOE 2010 - 2016 NAC. 21 Corporación Radio e Televisión de Galicia (CRTV) 1985 1996-2005 PP 2006-2009 PSOE 2010 - 2016 PP 21 Radio y Televisión de Andalucía (RTVA) 1989 1996-2016 PSOE 21 Radiotelevisió Valenciana (RTVV) 1989 1996-2013 PP 21 Radio Televisión Madrid S.A 1989 1996-2016 PP 21 Radio Televisión Canaria 1999 1999-2016 NAC. 18 Radiotelevisión de Castilla-La Mancha (RTVCM) 2001 2002-2011 PSOE 2012-2015 PP 2016 PSOE 15 Ens Públic de Radiotelevisió de les Illes Balears (EPRTVIB) 2005 2006-2011 PSOE 2012-2015 PP 2016 PSOE 11 Radiotelevisión del Principado de Asturias (RTPA) 2005 2006-2011 PSOE 2012 REG. 2013 - 2016 PSOE 11 Corporación Extremeña de Medios Audiovisuales (CEXMA) 2006 2006-2011 PSOE 2012-2015 PP 2016 PSOE 11 Corporación Aragonesa de Radio y Televisión (CARTV) 2006 2006-2011 PSOE 2012-2015 PP 2016 PSOE 11 Radiotelevisión de la Región de Murcia (RTRM) 2006 2006-2016 PP 11 Source: Own elaboration. Table 2: Variables under study. Organisation Autonomous Region x1 Autonomous Region x2 … Autonomous Region x13 Year 2016 2015 … 1995 2016 2015 … 1995 2016 2015 … 1995 (V1) Political party P 1-3 (V2) Expenditure / employee (G1) Absolute and relative value = (G1 200x -G1 200x-1 )/G1 200x-1 (V3) Total expenses (G2) Absolute and relative value = (G2 200x -G2 200x-1 )/G2 200x-1 (V4) Total assets (A) Absolute and relative value = (A 200x -A 200x-1 )/A 200x-1 (V5) ROA (R1) Absolute and relative value = (R1 200x -ER1 200x-1 )/R1 200x-1 (V6) Asset Turnover (R2) Absolute and relative value = (R2 200x -R2 200x-1 )/R2 200x-1 (V7) Share total (ST) Absolute and relative value = (ST 200x -ST 200x-1 )/ST 200x-1 Orive Serrano, V., Marco Fondevilla, M. & Latorre Martínez, P. Economic performance of the Spanish regional public service broadcasters according to the political party in government: An empirical analysis (1995-2016) ISSN 2386-7876 – © 2019 Communication & Society, 32(4), 159-170 165 4. Results With the objective of quantifying and empirically analysing the variables in relation to the political party in government in each Autonomous Community (PP, PSOE or a nationalistregionalist party), a descriptive analysis of the variables under study was first undertaken. Subsequently, a Kruskal-Wallis contrast was carried out. The reason for choosing this nonparametric test was determined by the characteristics of the sample, which does not meet the requirement of normality, nor are the variances for the analysed variables homogeneous. To achieve these objectives, the SPSS 22.0 statistical program was used. 4.1. Descriptive analysis of the variables Table 3 presents the average values for each variable under study. The results thus indicate that the highest level of expenditure per employee tends to occur in regions with socialist governments, while the highest levels of total expenditure, volume of assets and asset turnover can be found in communities with nationalist or regional governments. In the communities with PP governments, however, the greater return on investment (ROA) stands out, despite the fact that the regional broadcasters administered by nationalistregionalist party governments have a much higher operating income than the networks overseen by PP or PSOE governments. In fact, the amount of operating income of broadcasters managed by nationalist or regionalist parties is more than double that of broadcasters that operate under populist governments. Although these results must be assessed with caution, taking into account the particular income structures of the broadcasting entities, which are closely linked to public subsidies, and the frequent occurrence of negative ROAs as a direct consequence of equally negative results; the observed trend suggests the probability of political influence on economic management, which varies depending on the political party in power. The higher level of expenditure and income (taking into account that the latter depends heavily on public subsidies) observed in communities with nationalist or regionalist governments can be explained by particular features, such as the commitment to a second official language, with the consequent increase in production costs, or that they look to forge a stronger socio-cultural relationship with their audience. In fact, networks managed by the nationalist-regionalist parties obtained a greater total audience share than the other two network groups, with a difference of more than 2.5 audience share points. This result is especially evident in communities where one or more television networks broadcast in the regional language. Orive Serrano, V., Marco Fondevilla, M. & Latorre Martínez, P. Economic performance of the Spanish regional public service broadcasters according to the political party in government: An empirical analysis (1995-2016) ISSN 2386-7876 – © 2019 Communication & Society, 32(4), 159-170 166 Table 3: Descriptive analysis of the variables. Variable under study and political party N Mean Standard Deviation Minimum Maximum Expenditure per employee (in euro) PP 86 48,654.47 33,496.54 27,712 338,426 PSOE 77 53,937.01 19,971.61 21,882 119,218 Nationalists 51 47,643.10 8,506.32 29,640 67,494 Total 214 50,314.17 24,791.58 21,882 338,426 Total expenses (in euro) PP 86 78,545,431.74 58,491,806.11 1,407,722 216,719,333 PSOE 77 75,007,380.88 54,630,175.46 9,696,444 214,694,532 Nationalists 47 104,096,423.55 70,408,100.52 19,666,821 232,963,235 Total 210 82,966,701.74 60,823,976.39 1,407,722 232,963,235 Total assets PP 86 68,229,388.23 45,081,691.82 3,947,367 216,655,103 PSOE 77 77,713,133.60 77,562,787.84 8,502,291 339,881,990 Nationalists 49 109,186,595.55 96,202,531.20 7,044,967 357,789,882 Total 212 81,140,480.46 73,115,664.03 3,947,367 357,789,882 ROA PP 86 -0.81 0.61 -2.16 0.16 PSOE 72 -0.70 0.62 -2.17 0.02 Nationalists 49 -0.50 0.76 -3.31 0.00 Total 207 -0.70 0.66 -3.31 0.16 Turnover of assets PP 86 0.54 0.46 0.00 1.70 PSOE 77 0.88 0.72 0.03 2.97 Nationalists 49 1.18 0.74 0.05 3.19 Total 212 0.81 0.68 0.00 3.19 Total audience share PP 85 11.46 6.35 0.00 22.20 PSOE 75 12.29 6.45 1.70 26.10 Nationalists 51 15.61 6.68 5.40 29.30 Total 211 12.76 6.65 0.00 29.30 4.2. Kruskal-Wallis Contrast In the expenses section, non-parametric tests confirm the correlation between the political party in the regional government and the variables of 'Expenditure per employee' and 'Total expenses', as well as the regional networks' ratio of total Audience Share. Thus the hypothesis of independence for the year on year variation ratio of expenditure per employee cannot be refuted, nor that for the year on year variation ratio of total audience share. A similar analysis of the variables linked to the assets and income of the different entities yielded the following results, in which the dependence between the ROA and the ratio of asset turnover and the political party in the regional government is confirmed. However, dependence with volume of assets or the year on year variation of this was not confirmed. Orive Serrano, V., Marco Fondevilla, M. & Latorre Martínez, P. Economic performance of the Spanish regional public service broadcasters according to the political party in government: An empirical analysis (1995-2016) ISSN 2386-7876 – © 2019 Communication & Society, 32(4), 159-170 167 Table 4: Non-parametric test. Null hypothesis Test Sig. Decision 1 The distribution of expenditure per employee (V2) is the same across the political party categories (V1) Kruskal-Wallis test for independent samples 0.005 Reject the null hypothesis 2 The distribution of expenses (V3) is the same across the political party categories (V1) Kruskal-Wallis test for independent samples 0.042 Reject the null hypothesis 3 The distribution of total assets (V4) is the same across the political party categories (V1) Kruskal-Wallis test for independent samples 0.86 Do not reject the null hypothesis 4 The distribution of ROA (V5) is the same across the political party categories (V1) Kruskal-Wallis test for independent samples 0.018 Reject the null hypothesis 5 The distribution of asset turnover (V6) is the same across the political party categories (V1) Kruskal-Wallis test for independent samples 0.000 Reject the null hypothesis 6 The distribution of total audience share (V7) is the same across the political party categories (V1) Kruskal-Wallis test for independent samples 0.008 Reject the null hypothesis 7 The distribution of variation of total audience share is the same across the political party categories (V1) Kruskal-Wallis test for independent samples 0.212 Do not reject the null hypothesis 8 The distribution of variation of total assets is the same across the political party categories (V1) Kruskal-Wallis test for independent samples 0.735 Do not reject the null hypothesis Asymptotic significances are shown. The significance level is 0.05. Table 5 shows the results of the Kruskal-Wallis contrast for the three groups of broadcasters depending on the political party in power. As can be seen, there are significant differences for different economic variables depending on the political party in government in the Autonomous Community. Table 5: Kruskal-Wallis Test. Variable under study Expenditure per employee Total expenses Total assets ROA Turnover of assets Chi-squared 10.557 6.333 3.364 8.183 26.010 g.l 2 2 2 2 2 Asymptotic sig. 0.005 0.042 0.86 0.017 0.000 Variable under study Annual variation in expenditure per employee Annual variation in total expenditure Annual variation in total assets Total audience share Annual variation in audience share Chi-squared 0.189 0.563 0.617 9.641 2.983 g.l 2 2 2 2 2 Asymptotic sig. 0.910 0.755 0.734 0.008 0.225 a) Kruskal Wallis test. b) Cluster variable: political party.