Abstract
Introduction
The consolidation of media ownership is often considered a major civic risk to democratic societies (e.g. Baker, 2007; Picard and Pickard, 2017: 22–23). An owner may influence their outlets’ agenda and skew the public discourse in their favor – more ownership consolidation thus means concentration of discursive power. An owner may also seek to reduce production costs by reproducing content across many outlets, or by reducing investment in the product and trading journalistic performance for cost savings (e.g. Lacy and Blanchard, 2003; Peterson and Dunaway, 2023). Yet the trend of ownership consolidation is common across Europe and largely unopposed by policy (Trappel and Meier, 2022).
This study focuses on the Finnish newspaper sector, which is relatively concentrated: four leading companies command 75% of the market (Mäntyoja and Manninen, 2023). Regional-level figures are even more alarming, as most Finnish provinces are dominated by virtual newspaper monopolies (Hellman, 2022). High ownership concentration has consistently been noted as one of Finland’s main risks in the Media Pluralism Monitor (cf. Manninen, 2017; Mäntyoja and Manninen, 2023). These warnings are balanced with arguments according to which consolidation is necessary to keep struggling newspapers in operation (Lehtisaari et al., 2024: 174; see also Esser and Brüggemann, 2010: 52–53). For example, induction into a large media conglomerate might offer savings in infrastructure (e.g. software, premises and administrative services) and access to syndicated content.
Decades of research have yielded inconclusive results on the effects of concentration. Much of this research is from the United States, but even their empirical contexts and methods are varied, which complicates making sense of what changes in ownership mean for audiences. There is evidence pointing to two directions: Either ownership consolidation is a threat (e.g. Becker et al., 1978; Grotta, 1971) or a boon (e.g. Demers, 1996b; Lacy, 1991) to audiences. This paper seeks to clarify the situation by expanding the field of research into a non-Anglo-American context and by utilizing a modern framework for assessing journalistic performance. We compare the output of 69 local newspapers published in Finland, some of them owned by major newspaper corporations and others being independent. The comparison focuses on the newspapers' ability to satisfy audiences’
Media ownership and social responsibility
Our research is built around a certain ideal of what journalism (or media in general) ought to be like: the
Commercial media’s concentration has been a particular point of concern. However, extant evidence has often been found lacking in explanatory potence. For example, Compaine’s (1980: 98–100) review found no empirical support for ill effects stemming from consolidation or chain ownership. Instead of proof, much of the critique is based on theorisation. Baker (2007) offered a direct rebuttal to Compaine nearly two decades later, but still had to rely mainly on abstract arguments. According to Baker (2007), the dispersal of communicative power should be seen as a goal in itself for democratic societies (6–16), and that the threat of malpracticing media proprietors is too great to ignore even if there is currently little evidence of abuses (20–22). This is not to suggest
The United States is not the sole locus of these worries (Christians and Nordenstreng, 2004). What combines these concerns is a strong emphasis on audiences’ needs rather than the needs of the states or media companies (cf. Siebert et al., 1956). This pursuit for “common good” has become the basic tenet of modern journalism in liberal democracies across the world – and most certainly in so-called democratic corporatist countries like Finland (Hallin and Mancini, 2004: 195–197).
What, then, does social responsibility to audiences entail, and how does one define “common good” in journalism? Furthermore, what ownership structure is most likely to deliver that common good? Research into media structures and journalism’s social responsibility has continued for some 70 years, and the literature offers an abundance of study designs. In the following review we will focus on
As far as we know, first studies on the effects of newspaper ownership appeared in the United States nearing the end of 1940s (i.e., Bigman, 1948), with some of their data going back to at least 1939 (Nixon and Jones, 1956: 303). This rising scholarly interest stemmed from the same developments that prompted the Hutchins Commission Report: authors first cite the appearance of “single-ownership cities” (e.g. Nixon, 1954), then the spread of “newspaper chains” (e.g. Wackman et al., 1975) and later the rise of “corporate newspapers” (Demers, 1996a; 1996b, 1998). When similar changes have hit countries beyond the United States, familiar worries have followed: In Canada (e.g. Trim et al., 1983), Finland (e.g. Herkman, 2005), Sweden (e.g. Ots, 2012), Switzerland (Vogler et al., 2020) and Belgium (e.g. Hendrickx and Van Remoortere, 2021).
Overview of research on newspaper market’s development.
First on the horizontal axis is
The first category on the vertical axis is
The above, brief overview demonstrates how plentiful and varied the existing research is. It also underscores the need to precisely position the current study in a particular niche. In Table 1, our study would fall in the top-center square. We compare one newspaper
Several studies suggest being part of a large newspaper chain or bearing hallmarks of “corporate newspapers” has benefits 1 . They are more critical towards authorities (Demers, 1998: 583), have more innovation-oriented managers (Krumsvik et al., 2013: 102–103), publish more highly-rated journalism (Becker et al., 1978: 107), publish more news on politics (Sjøvaag et al., 2019: 74) and certain other “hard news” categories (Hamilton, 2004: 152–153), allow their journalists more time per article (Lacy, 1991: 40), and they are generally more profitable (Lacy and Blanchard, 2003: 956). In contrast, other studies suggest corporate (or chain) ownership may lead to increasingly negative tone and more cynical framing in election coverage (Dunaway, 2013: 35; Dunaway and Lawrence, 2015: 52), less substantial coverage (Dunaway, 2008: 1198), less space reserved for editorial content (Lacy, 1991: 40; Litman and Bridges, 1986: 21) and fewer journalists in employ (Lacy, 1991: 40; Lacy and Blanchard, 2003: 956; Litman and Bridges, 1986: 20). Further still, many studies conclude there are little to no significant differences between owner types (Borstel, 1956: 221; Coulson, 1994: 408; Demers, 1996b: 301; Drew and Wilhoit, 1976: 439; Fletcher and Bell, 1981: 37–38; Lacy and Fico, 1990: 49; Hale, 1991, 44; Matthews, 1996: 349–350). Further adding to the uncertainty, much of this body of research is already decades old and focuses on the US, while very little of it has been done on Finnish newspapers.
Implications of newspaper ownership consolidation in Finland
The number of newspaper owners has declined in Finland for some time (Hellman, 2024). While many newspapers have gone out of business, the number of owners has mostly declined due to consolidation: larger companies buying out smaller ones, and perhaps abolishing some of the newly redundant papers in their portfolio.
Consolidation can still aid struggling publications through economies of scale (for an overview, see Van Kranenburg, 2004). First, financially stable companies can help their component parts by lending resources (e.g. Lehtisaari et al., 2024). Second, large corporate owners can offer savings by removing redundant administration and infrastructure costs (e.g. Hendrickx, 2020). Both of these benefits are financial and therefore difficult to confirm, as expense structures are rarely accessible to researchers. Third, large corporations can pool and focus their development resources, allowing even small newsrooms to launch state-of-the-art services such as mobile apps. Survey evidence from Norway supports the idea: managers in larger newspaper chains appear more prepared to keep up with technological development (Krumsvik et al., 2013: 102–103). Fourth, commonly owned outlets can share content. Content sharing—or duplication—is an established practice in many media companies, including some of Finland’s leading newspaper publishers (Lehtisaari et al., 2024).
Conversely, consolidation also entails risks. They are outlined, among others, by Napoli (1999) and Picard and Dal Zotto (2015). The most salient argument against ownership concentration relates to political capture, that is, concentration of discursive power into fewer and fewer, politically motivated hands. Use and abuse of this power could threaten democracy by privileging the opinions and aims of large media owners such as Rupert Murdoch or the late Silvio Berlusconi. There is little evidence of political capture in the Finnish media. All major media companies in Finland tend to have lots of small-time stock owners (joined by few capital management firms) with no single-owner majorities in effect (e.g. Mäntyoja and Manninen, 2023; cf. Picard and Van Weezel, 2008: 26).
Other risks seem more plausible for Finland. One is content homogenisation through syndication: sharing content among several dozen newspapers could erode the overall diversity of views and information. This is something Lehtisaari and colleagues (2024) studied extensively and found evidence of widespread content duplication among Finnish regional and national dailies. However, content duplication is less likely an issue at the most local level of journalism, which is the focus of current study (cf. Yan and Napoli, 2006: 797–798, on local television journalism). These papers’ central value proposition is local content, which is difficult to outsource or recycle from elsewhere – although some companies may be willing to do so even against audience demand, as shown by Martin and McCrain (2019).
For local Finnish newspapers, the most worrying threat involved with consolidation is a sort of cannibalisation, described in the US by Lacy and Martin (1998: 63) and later Benton (2021, 2023). It is a process where profit-prioritizing companies acquire newspapers and begin aggressively seeking savings and stretching profit margins. A study by Peterson and Dunaway (2023, 85) indeed suggests “investment owners” like hedge funds and private equity companies tend to reduce journalistic staff more than other owners — particularly staff assigned to “hard news” beats like politics. Reducing staff, increasing the use of syndicated content and raising subscription prices will inevitably decline the papers’ appeal, but the readers (to whom the paper is often the only local news source) will be slow to abandon it. In the interim, the new owner will reap high profits at the expense of the papers’ civic utility and long-term viability.
Could Finnish newspapers be facing a similar fate? Even companies dedicated to journalism may need to squeeze out additional profits, perhaps to invest in infrastructure or to ensure the parent company’s survival. Companies may also seek to appease stockholders by distributing generous dividends even when the resources would be better invested in improving products. This threat is the driving motivation behind the current study: to find out whether corporate ownership is reflected in local newspapers’ performance.
Method
We quantitatively analyzed one issue each from 69 Finnish local newspapers, in accordance with a code book developed by the authors. Our data consists of one self-selected issue per newspaper, each submitted to the 2022 local newspaper competition, held annually by the trade organization News Media Finland (formerly The Finnish Newspapers Association). The competition is open to all local newspapers that are part of News Media Finland (as the overwhelming majority of Finnish newspapers are). The competition is split into two divisions based on whether the paper is subscription-based or a free-sheet; our sample only covers subscription-based papers. The participating newspapers were instructed to select one issue published between March 1st and April 30th 2022 and submit it for review. The submitted issues should, therefore, reflect the competing newsrooms’ understanding of their best performance 2 .
This sample is a cross section of the local newspaper genre in Finland, covering 52% of eligible News Media Finland member papers 3 . These kinds of newspapers are typically published in small towns and rural municipalities, and they are almost exclusively non-dailies (i.e. they publish 1–3 issues per week). Larger urban areas on the other hand sustain dailies (4–7 issues per week) and/or free-sheets. All newspapers in our sample are non-dailies, which corresponds with the general composition of the Finnish local newspaper genre; in 2022, only one News Media Finland member paper was a “local daily”, but it did not participate in the competition that year. Our data comprises 2133 individual items including news shorts, investigative articles, feature pieces, editorials, and readers’ opinion pieces 4 . We have included opinion pieces in the analysis, as they may also contain information relevant to the readers and selection of those texts is in the hands of journalists. Excluded from the sample are advertisements, comic strips and other non-editorial content.
Each item (i.e., individual text) was analyzed with regards to the public’s 1. Emergencies and public safety 2. Health 3. Education 4. Transportation Systems 5. Environment and Planning 6. Economic Development 7. Civic Information 8. Political Life
We have slightly adapted this framework to the Finnish context by systematically including certain ecclesiastical topics in the Political Life category. In Finnish society, churches provide many non-spiritual services such as social housing, food aid, domestic assistance, summer camps, and event space rental. Despite generally being secular, some two-thirds of Finns still belong to either the Evangelical Lutheran Church of Finland (65%) or the Orthodox Church of Finland (1%), both of which practice democratic decision-making through church council elections.
Each CIN category was coded based on the presence of CIN-relevant information: if an analyzed text contained any information related to, for instance, public safety, it was coded as such. The codes were not mutually exclusive, and a single text could have several CIN topics present. It should be noted that Friedland and colleagues’ (2012) definitions for the categories can be interpreted quite broadly, and so we limited the categories to information that facilitates active participation. To give an example, the original schema includes “cultural and arts information” as well as “recreational opportunities” (p. 7) under Civic Information – huge swathes of newspaper content could thus be included. We have considered CIN-relevant only information that allows participation beyond that of a customer, visitor or spectator. Information about an art club, its activities and opportunities of joining has been coded as CIN-relevant; information about an upcoming (or indeed, past) art exhibition of said club has been excluded.
For complementary analysis, all texts were categorized based on their length, localness, and authorship, following the performance measures adopted by Napoli and colleagues (2016, 2018). Item size was estimated visually and coded into three categories: ‘short’ for texts that could fit into a single column, ‘major’ for texts that take up at least one full page, and ‘basic’ for all others. All examined newspapers were of similar physical size (the tabloid format).
Localness refers to the texts’ geographical reference point in relation to the newspapers’ self-stated area of coverage. The papers’ coverage areas were determined by consulting their websites. The code ‘local’ was used for topics that were particular to the paper’s area of coverage (e.g. news on a municipal council’s decision), ‘non-local’ for topics that were not specific to the paper’s locality (e.g. news on domestic politics), and ‘localized’ for otherwise non-local topics that had a local tie-in (e.g. an item on national health services with an additional paragraph on local services). Texts with no discernible locality (e.g. an article providing general nutrition advice) or with unclear location were coded as ‘undetermined’.
Lastly, authorship was coded based on the items’ bylines: the code ‘staff’ was used when the item was credited to journalists named as the paper’s employees or when it was anonymously credited to the newspaper collectively. ‘Other’ was used for items credited to a named person that was not identified as a staff writer, ‘other organization’ for named organizations other than the newspaper itself (e.g. an opinion piece signed collectively by a political party), and ‘unnamed’ was used when the item had no byline whatsoever.
The coding was done manually by the study’s first author, following a code book created by the authors. The code book was developed iteratively, with the authors cross-coding sub-samples of the data, comparing results and negotiating to resolve conflicts. Cross-coding was only implemented on the CIN categories. The number of cross-coded articles was 440 items in total, approximately 20% of the full sample. Changes were made to the code book after each round of cross-coding, and the code book was approved for the main analysis when coding had reached a minimum of 94% level of unanimity in recognizing each CIN category. Reliability problems were caused almost entirely by the secondary coder producing false negatives in rarely occurring categories. This issue was resolved with small additions to the code book, and the analysis proceeded without further intercoder reliability testing at the time. During final analysis difficult coding decisions were made jointly by both authors. A post-hoc test with a different data set (involving the same main coder as here but a different cross-coder) produced satisfactory reliability scores (Cohen’s kappa >0.65) for the CIN instrument. As a caveat, the instrument was slightly refined for this post-hoc test, but in a way that does not impact negatively on the test’s applicability to the prior analysis.
The newspapers in the sample were divided into two categories according to their owner type: either as
Only 31 of Finland’s 268 newspapers are dailies, and the companies that publish them represent major economic power in the sector. Of the nine corporate owners in our sample, eight are among Finland’s ten leading newspaper publishers by net revenue – and together with the ninth company (HSS Media) they comprise over 70% of the entire newspaper sector’s revenue (Kauppalehti, 2023; Statistics Finland, 2022). Our sample includes a total of 25 newspaper owners (with subsidiaries collapsed together), of which 16 are independents (i.e., owners of only non-daily newspapers) and 9 corporate owners (i.e., owners of both daily and non-daily newspapers).
Results
Descriptive variables in corporate and independent newspapers.
The data show corporate and independent local newspapers are remarkably similar in structure and size. They have about 31 items of editorial content per issue. Most articles take up less than one page, with major stories and news shorts both in the minority.
The (non)localness of items is near-identical in corporate and independent papers. They all consist mostly of local items, even if the papers would have access to syndicated non-local content. We also explored the localness of particular CIN categories, but the number of items per category remained too low to draw meaningful conclusions.
The papers’ production models also appear consistent: less than half of items are credited to a staff writer or the newspaper collectively. However, slight differences emerge when looking at the sources of the remaining items. In corporate papers 32.8% of items are credited to a person or organization outside the resident staff. In independent papers this proportion is smaller (26.3%), but the share of entirely uncredited items is higher (26.8%, compared to 23.0% in corporate papers). The shares of items credited to a combination of staff and others are negligible in both types of papers (0.4% and 0.6%).
Turning to the frequency of CIN-relevant items, an average newspaper in our sample contained 11.6 of them – less than half of the average papers’ contents. Their size distribution (basic, major or short) is in line with the distribution of all items. No statistically significant difference in CIN-relevant items’ size was observed between the newspaper types (Pearson’s χ2 = 0.524 and
Proportions of CIN categories in corporate and independent newspapers.
Politics is clearly the most common fare in the studied newspapers, outweighing most other CIN categories by a factor of three. Civic information is the second most common category by a broad margin, despite being a wide-ranging category comprising information on local associations and public services alike. The remaining six CIN categories are roughly as common, their share sizes all within 3.1 percentage points. This, we believe, gives a rough idea of the papers’ priorities with regards to CIN categories: politics first, civic information second, and all others then.
Pearson’s χ2 test results for ownership type and CIN category frequencies.
An open question remains: could the results be skewed by some items containing many CIN categories? It is possible that one type of newspaper owner tends towards broad-ranging articles and the other towards tightly focused ones. Of the 797 CIN-relevant items, 78% contained information pertinent to only one CIN category, 20% information from two categories and 0.8% (i.e. seven items) from three categories. Only two items contained information from more than three CIN categories. Could the few multi-CIN items still hide differences between corporate and independent papers?
We investigated differences in the occurrence of any CIN-relevant items by collapsing all CIN categories into a single, binary variable that indicated whether an item contained any CIN-relevant information. Here, a statistically significant difference between owner types finally emerges: corporate papers contain more CIN-relevant items than independent papers (χ2 = 5.153 and
Isomorphism across local newspapers
The findings of this paper reveal remarkable, and somewhat unexpected, similarities between corporate and non-corporate (i.e. ‘independent’) local newspapers in terms of fulfilling citizens’
It has been suggested that consolidation enables and incentivises recycling of content from one newspaper to another (Beckers et al., 2019). Membership in a large corporation should, then, result in an increased use of syndicated content from other papers: non-local content authored by non-staff writers. However, this hypothesis does not hold up in our data. The proportion of local articles is the same in both types of newspapers, approximately two-thirds. Same is true for the contents’ authorship, which is about two-fifths staff writers in both categories. Perhaps it is the papers’ local orientation that prevents corporate-owned newspapers from using more syndicated content; news on municipal politics, for example, might be important locally but irrelevant elsewhere.
Along with the direct recycling of content, there is a suggestion that being integrated into a large media conglomerate might lead to cost savings in infrastructure, administration, and marketing. This could, in turn, be reflected in resources available for reporting: more and better stories. However, regardless of whether such savings exist or not, they do not show up in our results. The average number of items per paper was the same in both corporate and independent papers, and those items were similarly distributed between longer and shorter articles. Same applies to human resources, as the production models of the papers in terms of authorship (staff vs other writers) were similar. While the results seem compelling, this line of argument comes with two caveats. First, our study did not touch upon profitability: it is possible corporate papers did find savings, but they were used to expand the profit margin, not the newsroom’s resources. Second, our study is not able to touch upon the contrafactual: could the corporate papers have been doing worse than independents without the savings brought by consolidation? Perhaps some of them may have even failed and gone bankrupt without corporate support. These hypotheticals are well beyond our methods, but they are often brought up in public debates. As far as we know, no research evidence exists either way.
Few uncertainties aside, our data convincingly shows a strong similarity between corporate and independent local newspapers. What could account for this homogeneity? One possible explanation is the professionalization of the journalistic field. The Worlds of Journalism survey finds that most Finnish journalists have tertiary level education and their views on professional norms appear fairly uniform (Väliverronen et al., 2023: 69). Most working journalists are unionized and thus readers of the trade journal
It is also worthwhile to consider whether some external factor could have biased our scrutiny. Did the major news events of spring 2022—such as Russia’s full-scale invasion of Ukraine and the subsequent influx of Ukrainian refugees and Finland’s NATO discussion—push the newspapers into uniformity? This explanation is possible, but even if true it would only account for some of the observed similarities. Aforementioned topics were mainly discussed in localized and non-local items (e.g. how refugees settled in specific Finnish municipalities), constituting only 12% and 23% of the content, respectively.
Another point of caution arises from our quantitative methodology. Our analysis did not assess the items’ grammatical correctness, their textual flow or narrative engagement, investigative rigor, or the amount and accuracy of provided information. These qualitative details are lost in our current analysis, and this omission might belie more nuanced differences between newspaper types. Furthermore, our focus on printed newspaper issues ignores possible differences between their online editions. While print is likely the flagship product, online content might supplement it in different ways based on the owner and its resources. One must also remember the limitations of our sample: it represents local, Finnish, non-daily newspapers. By titles this group comprises the largest segment of the Finnish newspaper sector, but in number of journalists and readers it is dwarfed by regional and national dailies, not to mention broadcast media. It would be important for future research to expand this investigation beyond local outlets.
Lastly, it is important to clarify potential misconceptions arising from our paper. Our primary aim was to investigate whether media ownership consolidation affects the papers’ civic potential – whether it undermines local newspapers’ social responsibility. To address this, we employed the CIN framework as our analytical tool. However, supporting local democracy is not the sole purpose of local newspapers (in as much as any purpose, aside from staying in business, can be universally applied). We by no means suggest that newspapers should maximize the number of CIN items and prune out, for example, human interest stories and sports results. Many items in our sample do not raise the CIN flag but have a role in providing pastime and support for a shared, local identity. They may well be appreciated by audiences, regardless of their traditionally understood news value (cf., Costera Meijer, 2020: 360).
In this research, we examined a range of data points in our sample, comprising independent and corporate local, non-daily newspapers. Yet our analysis detected barely any statistically significant difference between the types of newspapers. We can still speculate that some differences likely exist somewhere, even if not in content – why consolidate otherwise? The conditions, infrastructures and social frameworks that undergird the content may change, perhaps very slowly over time as a result of consolidation. Different methods and sampling techniques are necessary to complete this investigation into ownership’s effects on newspaper content. However, we still maintain that future research should depart from the audiences’ perspective. It should look at the content and its delivery as well as the reception of journalistic material. Such an orientation is most compatible with the concerns of social responsibility theory, that is, concern for the common good.
Footnotes
Acknowledgements
We thank News Media Finland for granting access to the research material. We acknowledge the contribution of Essi Stenbäck, who as research assistant completed the initial processing of the material.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was partially funded by the Kone Foundation.
