Размещено на http://www.allbest.ru/

FEDERAL STATE AUTONOMOUS EDUCATIONAL

INSTITUTION FOR HIGHER EDUCATION

NATIONAL RESEARCH UNIVERSITY

HIGHER SCHOOL OF ECONOMICS

Faculty of Social Sciences

Master Thesis

Field of Study 41.04.04 Political Science

Master's Program “Politics. Economics. Philosophy”

Sources of media bias in foreign policy reporting - the case of US media

Filip Lazarevic

Scientific Supervisor Professor

Israel Marques

Moscow

2019

Introduction

News media have come under an unprecedented level of scrutiny in the age of so called “fake news”, where politicians seem to use this phrase to dismiss out of hand reports they disagree with. The public often follows suit, with many switching from traditional news media to new sources of information, with different or absent journalistic standards - including innovative sources such as online citizen platforms (e.g. Vice News), to malicious sources such as the widely discredited Breitbart News https://www.theamericanconservative.com/dreher/breitbart-fake-news-alex-marlow/ . The problems of media bias have been explored in different ways in academia, but these usually lack a general explanatory mechanism - usually simply assuming that partisanship or culture is the root of such biases. Others claim that power, influence, class, race or other attributes of journalists are key for explaining causes of bias. Demand side arguments also exist, saying that news media is in the business of providing news, and its profit seeking incentives will lead it in certain directions - however, these theories usually simplify where the demand comes from - seeing viewers and listeners as the consumers, when in fact media rely also on advertisement revenue. On the other hand, rational choice theories fail to grasp the importance of structures that underpin choices of individuals, and fail to take into account the impact of network dynamics on information flow between such individuals. Information based theories, such as those developed under the umbrella of regulatory capture, come closest to a holistic approach to study complex systems such as the one at hand, but they assume that media is not in itself the object of influence by special interests.

Therefore, such approaches usually do not explain satisfactorily the institutional structures and networks that the media are embedded in. The biggest news media in the US are corporations with hierarchical structures, embedded in a network of other corporations, government agencies, civil society, etc. While there is an extensive body of research on the how special interests aim to influence governments, there is a large gap in studying how special interests influence the media itself to promote their own interests. This research tries to fill this gap, by treating the media as an integral actor in the interplay of special interests.

This paper aims to mount a sustained structural analysis of influences of special interests on large private corporate media in the United States, and thereby to answer the research question:

Does the media show a bias in favor of special interest narratives in its reporting on foreign policy of the US government?

To answer this question, we posed two hypotheses:

Hypothesis 1: top military and oil industry special interests have access to all major media companies through board membership

Hypothesis 2: media reporting is biased towards special interests' objectives in foreign policy, by over-emphasizing foreign government abuses in states closed to US special interests, as compared to open states

These hypotheses were tested based on two linked methodologies, the first examines the networks of special interests, here narrowed down to the largest oil and defense industries, and corporate media which for the purposes of the research was limited to the 7 largest corporations. For this research, Social Network Analysis (SNA) was used to map out the links between media and special interests by looking at corporate board membership links. Board members often sit on both special interest boards, and on media boards, or share a third board room in an intermediary company. The analysis found that in fact, all the observed media have some connection to the special interests observed, and that the network is fairly interlinked, so that no media company was further than seven boards away from any of the special interests. This is a significant finding, as it shows how media is embedded in a network of other corporate actors, who can use this network and strategic positioning to restrict or facilitate information flow between the different actors.

As it is not enough to show the mechanism of bias formation, the second part examined the discourses emanating from the news media and looks if these narratives show a bias in favor of special interests. By looking at cases of opinion articles from three of the abovementioned news media, we compared the narratives used to describe a foreign state that is open, and one that is closed to special interests (open here means open to trade with goods and services offered by the special interest, and open to investment and foreign ownership of such industries, and vice versa). We found that media is in fact biased negatively when reporting on closed states, and that it is more likely to utilize keywords that indicate the need to change or intensify foreign policy measures by the US in relation to such states, with calls for supporting a return to democracy, labeling the governments as tyrannical, authoritarian, etc. even when such labels may be more appropriate to the open states' government.

In the final part, we synthesize the two approaches to formulate a structural theory of media companies as hierarchies embedded in a network of other hierarchies, including special interests, that will use the structural characteristics to gain access to the public ear by influencing media discourses. From this, we formulate recommendations for further research, as well as recommendations for how to help improve media's public interest role in democracies.

Literature Review

In this section we will discuss the academic fields dealing with special interests' influence on public discourses, as well as the other literature related to social influence, media bias, and the influence of special interests on societies more broadly. The research question: “Does the media show a bias in favor of special interest narratives in its reporting on foreign policy of the US government?” is part and parcel of the wider field of study of special interests' influence on public discourses. As such, we need to look at the state of play in contemporary research in regards to speciыal interests' influence on media. The media, the public, special interests, and the government all share complex relations between each other. The public is the largest, but least cohesive group in terms of preferences, and has the strongest potential impact on policies in democratic states, if cohesion is high in any given policy field. For example, right to vote or own private property in the United States are issues of high public cohesion, and debate of it among political elites is all but absent. On the other hand, the right of abortion is an issue with less public cohesion, and this is represented in political debate, which is highly contested. Special interests often develop precisely around such issues, with either the aim to pressure political elites or to influence the public opinion, or both (e.g. the National Rifle Association which runs a TV channel for the public, and lobbies the government on 2nd Amendment issues). Finally, the media acts an important catalyst in society, as it informs the public on such issues, while also, as far as they are private, are influenced themselves by special interests. In the following sections we look at these relations from the perspective of different academic fields.

Rational Choice Theories

One of the early theories of special and public interest was presented by Downs (1962) DOWNS, A. (1962). THE PUBLIC INTEREST: ITS MEANING IN A DEMOCRACY. Social Research, 1-36, called the economic theory of regulation (ETR) which argues that individuals will on average follow their self interest in preferences of policies, and that politicians will generally follow these preferences in order to be re-elected. One exception, the author argues, in cases of “system survival” - when the public opinion is against policies which are necessary for the system to continue existing Ibid.. However, this theory did not have a nuanced view on the importance of other groups in society, i.e. special interests'. Olson further complicates the argument by showing that the public may not act to pressure the government, even when their self-interest is at stake. The author shows groups may not act to gain their common interests in situations where public goods (goods available regardless of participation) are in question, and hence voluntary action will be absent (OLSON, 1968 OLSON, M. (1968). The Logic of Collective Action: Public Goods and the Theory of Groups. New York: Schocken Books).

This argument will later become known as the “free rider” problem in rational choice theory, and will play a central part in the discussion of academia in group choice debates. Olson, and later Peltzman, also introduce a more nuanced concept of the interest group, distinguishing between large diffuse and small cohesive groups (later to be knowns as special interests). This theory postulates that governments are influenced by the competition among different interest groups, and that groups with the lowest cost of organizing relative to the size of the group are more likely to succeed (OLSON, 1968 OLSON, M. (1968). The Logic of Collective Action: Public Goods and the Theory of Groups. New York: Schocken Books) (Peltzman, 1976 Peltzman, S. (1976). Toward a more general theory of regulation. Journal of Law and Economics). They briefly also outline other types of groups and find that government regulation (e.g. antitrust law) works in the interest of small-dispersed interests, and not for the big-concentrated ones. For us, the most interesting aspect of the early academic work is that it provides us with the essential parameters of interest groups: their size and cohesion.

With this in mind, we will look at a recent addition to the above discussed rational choice literature of economic theory of regulation, where Dyck et. al. argue that:

“in a representative democracy, profit-maximizing media may play an important role in minimizing collective action costs for dispersed actors and therefore may help to tip the political balance against concentrated interests (including large incumbents) in certain cases. By helping to inform the broad public about certain public policy issues, media can create a more informed electorate to which politicians are likely to try to cater. Recognizing this role that the media can play allows for a richer theory of regulation, where outcomes depend on the relative strength of the media in informing and, in turn, empowering dispersed versus concentrated interests.” (Dyck, Moss, & Zingales, 2013 Dyck, A., Moss, D., & Zingales, L. (2013). Media versus Special Interests . The Journal of Law & Economics, 521-553)

Therefore, the article suggests that the media goes against special interests, informing the public on issues, and hence increasing cohesion of the least cohesive group: the public itself. As such, this paper directly contradicts the research question, which looks at biases in the media, due to special interests. As this is a key article in this field, it is necessary to look at it more closely, and discuss how our theory fits or clashes with it.

Dyck et. al. Ibid. note that the foundations of the ETR are that the public at large has no incentive to invest in informing themselves, as the payoffs are small, while small cohesive groups do. The for-profit media, seek profits by trying to capture an audience as broad as possible, which means that they have an incentive to inform the public on matters that the public is interested in consuming. The public has an interest in consuming political content for whatever the reasons may be (shock, sensationalism, or entertainment more broadly that accompanies politics). With this theory in mind, the authors set out to test it. However, before we go into such data, we need to discuss some of the underlying assumptions in the argument.

Firstly, it is not at all clear that media is seeking to inform the general public as its primary goal - media corporations seek to increase profit by selling advertisement content, rather than news. As such, their primary customers (for most online content, cable television, and radio based media) or highly important customers (for print media, magazines, and subscription based limited advertisement online media) are corporations interested in advertising their products or services. A Pew research paper shows that over “two-thirds (69%) of all domestic news revenue is derived from advertising” MITCHELL, A., & HOLCOMB, J. (2014, March 26). Revenue Sources: A Heavy Dependence on Advertising. Opgehaald van Journalism: https://www.journalism.org/2014/03/26/revenue-sources-a-heavy-dependence-on-advertising/ in the United States. The research paper points out that there are other sources of revenue, such as purchases of print media, cable subscriptions, or online subscriptions, but that these never present the majority of the revenue sources for profit seeking journalistic media. (Mitchell & Page, 2014

Mitchell, A., & Page, D. (2014). The Revenue Picture for American Journalism, and How It Is Changing. Washington, D.C. : Pew Research Center.).

Native online news media are less transparent, due to the fact that they are not legally seen as news media for tax purposes. Nevertheless, these platforms are estimated to receive the majority of its revenue from subscriptions as well. Another category discussed is the non-profit news media, which also depends on ad-revenues, however, they depend on consumers to a much larger degree, if not due to revenues, but because they are smaller in size and have specific audiences without which they would be far less viable, particularly those independent of foundations and other large block grants. (Mitchell & Jurkowitz, 2013

Mitchell, A., & Jurkowitz, M. (2013 ). Nonprofit Journalism: A Growing but Fragile Part of the U.S. News System. Washington D.C.: Pew Research Center. ). As such, the assumptions by Dyck et. al. that the news media have an incentive to inform the public is true only insofar as the positions taken do not interfere with the business interests - i.e. ad revenues.

As they define special interests as cohesive groups, we need to take into consideration that the business community of large corporations with foreign policy interests may find issues in foreign policy where cohesion would be expected. A particular interest that is more or less ubiquitously shared across the large business sector, is the search for new markets, hence the particular interest of this research on foreign policy. While large companies compete against each other in domestic markets, foreign markets that are closed are in the common interest of all large international businesses to open those markets up, and hence this may lead to cohesion. media policy public reporting

The “strength” of the media may not be the most decisive factor in creating cohesion in the public, due to economic incentives of the media themselves. Here Dyck et. al. seem to forget to apply their own rational choice theory on the media themselves. The media incentives can be described as follows: for profit news media seek to maximize revenues and profits; in order to do so, they must maximize their audience, to be more appealing to large advertisers; advertisers, in turn, may interfere in the news media operations through threats of withdrawing advertisement funding, if their direct interests are threatened by the media reporting; media organizations will try to predict the interests of the advertiser to prevent this from happening; this threat is greater with increased cohesion in the business community; the public can decide to consume a competitors news, but paradoxically the public is not cohesive as per the Dyck theory, unless media create cohesion. Dyck, A., Moss, D., & Zingales, L. (2013). Media versus Special Interests . The Journal of Law & Economics, 521-553

This logical progression would then lead us to consider that the media increase public cohesion against special interests only when there is competition between special interests; as soon as cohesion among the special interests arises, media stop being adversarial, and may actually support public divisiveness or cohesion with the special interests' objectives, based on their economic interests alone.

Information Based Approaches

Another approach to researching special interests was developed in the general field of information flow and access theories literature. A primary model in this field was developed by Levine and Forrence (1990)

Levine, M. E., & Forrence, J. L. (1990). Regulatory Capture, Public Interest, and the Public Agenda: Toward a Synthesis. Journal of Law, Economics, & Organization, 167-198., who specifically look at information access by different interest groups in society. They suggest that information flow and access is key to understand the issue of regulatory capture, which happens when a regulator is faced by special interests, and the public is under-informed. The authors call this information discrepancy “slack” - a regulator has slack, in cases where the public is not informed, i.e. the cost of information is high. The authors explain this concept as follows:

“High information, monitoring and organization costs create "slack," which shields officials from accountability to the general polity... This slack can be used by a regulator or her political sponsor to pursue officeholding self-interest (reappointment, reelection, or postregulatory pecuniary interests) by pursuing regulatory policies that benefit special interests. The regulator does this by using slack to avoid detection or effective restraint while she pursues policies of benefit to special interests who are willing to make political contributions, to sponsor the regulator for appointment or reappointment, or to provide lucrative postregulatory employment.” (Levine & Forrence, 1990

Levine, M. E., & Forrence, J. L. (1990). Regulatory Capture, Public Interest, and the Public Agenda: Toward a Synthesis. Journal of Law, Economics, & Organization, 167-198.)

The authors conclude with a list of institutions that reduce slack, listing (1) self-publicity - i.e. by being visible, some regulators can pursue their preferred policies; (2) political competition where politicians inform the public about their opponents' and their own policies; (3) other interest groups which compete in the marketplace of regulations; (4) academics; and (5) the media, which is characterized as “perhaps the most powerful” Ibid..

However, the authors stop short of exploring the possibilities that special interest have an active role in these institutions themselves. Special interests often own shares, have board seats, donate, or otherwise have a power relation in academic institutions, the media, NGOs, think thanks, and other slack reducing institutions. This is particularly important in an age of massive consolidation (most media came from six sources in 2011, down from from 50 in 1983 Lutz, A. (2012, Jun 14). These 6 Corporations Control 90% Of The Media In America . Opgehaald van Buisness Insider: https://www.businessinsider.com/these-6-corporations-control-90-of-the-media-in-america-2012-6 ), cuts in public support for universities and rising share of private universities, and an increasing awareness that NGOs often side with their donors in their opinions Ingraham, C. (2019, January 17). Massive new study traces how corporations use charitable donations to tilt regulations in their favor. Opgehaald van Washington post: https://www.washingtonpost.com/business/2019/01/17/massive-new-study-traces-how-corporations-use-charitable-donations-tilt-regulations-their-favor/?noredirect=on&utm_term=.d9bb8024f6fb but it may also be a way of influencing politicians by giving to nonprofits linked to politicians Porter, E. (2018, April 3). Charitable Giving by Corporations Is Also About Getting, a New Study Finds. Opgehaald van New York Times: https://nyti.ms/2H6gx0D . In other words, special interests have an incentive to increase slack, by decreasing the impact of slack-reducers.

Considering the interlinking nature of media, government, the military, powerful business interests, and think thanks, as these groups often share direct or indirect cross-ownership (in case of the private groups) and professional “revolving door” movement of staff (as in the case between the government and private actors), we can conclude that basic attitudes may also overlap. This could lead to cross-pollination of ideas, creating self-referential loops, where governments get their information from vested interests like think thanks, pass these on to the media, who then check this information with the same vested interests. As such, there is a possibility that media show the public a skewed picture of US foreign policy, despite its formal independence from governments. Furthermore, there is a cross-pollination of individuals who are heads of media businesses, who may also be part of other organizations, such as think thanks, international corporations, while also military and government sources and experts often provide commentary in the media.

Figure 1 - Realtions between actors

When it comes to foreign relations of the US and oil producing countries, the stakes are much higher than in some other relations, as oil is both economically and geo-strategically important. As such, the relation between the US and Venezuela on the one hand, and of Saudi Arabia on the other, are particularly interesting. We would expect that there is a general overlap of interests between the government and major military and oil industries special interests, since more government access means more industry access. The paper will hence explore the cross-links between military and oil industries one hand, the and media on the other. This relationship is presented in the following figure:

This paper will look at this neglected link between information flow between media and interest groups, in order to see if the media align with the interest of captured regulators and their captors. The links between the media and special interests will be explored through links in board membership.

Ideological Media Bias Studies

Another field of research is that of media leanings in terms of US politics. A frequent charge of the US media is that it has a “liberal bias”, i.e. that its reporting is slanted towards the interests of liberals. While this claim if often dismissed as a political narrative, some research seems to confirm it. Particularly interesting are the studies by Povich (1996 Povich, E. (1996). Partners and Adversaries: The Contentious Connection Between Congress and the Media. Arlington, VA: Freedom Forum), who shows that there is a higher level of democratic voters among journalists working as Washington correspondents than the general population, in terms of voting habits. She shows, for instance that 7% percent of correspondents voted for the Republican G. H. W. Bush in 1992, while the general population voted at 37%. As journalists' political leanings don't necessarily represent their reporting, such research was usually dismissed as not predictive of actual bias.

However, a more recent study shows a more nuanced set of results, where the authors analyzed media content to establish a list of references to think tanks, and comparing them to congressional references to think tanks (Groseclose & Milyo, 2005 Groseclose, T., & Milyo, J. (2005). A Measure of Media Bias Author(s): . The Quarterly Journal of Economics, 1191-1237). They find that media sourcing of think tank articles correlate with the congressional scores for the liberal voting based on the Americans for Democratic Action (ADA) score, where 100 is maximum correspondence with the ADA view on what “liberal” voting is.

While there are some issues with this paper in terms of its results and conclusions, for our purpose it is important to note that we are not looking at the liberal-conservative bias of media, but rather the public interest - special interest axis. However, we will note that the US Congress has been acting in ways that are not popular with the US public - e.g. US public view of relations with Saudi Arabia have been negative since early 2000, while policies of arms sales and general partnership have continued in Republican and Democrat congresses Taylor, A. (2018, October 19). Did Saudi Arabia have a reputation to ruin? Opgehaald van Washington Post: https://www.washingtonpost.com/world/2018/10/19/did-saudi-arabia-have-reputation-ruin/?noredirect=on&utm_term=.4e42bac088e1 . Furthermore, recent data shows that in 2018 both self-declared Democrat and Republican voters identified mostly negatively towards Saudi Arabia D - Unfriendly/Enemy at 50 and Ally/Friendly at 27; R - Unfriendly/Enemy at 42 and Ally/Friendly at 34. YouGov poll, see Frankovitc, 2018 Frankovic, K. (2018, October 17). Americans had limited trust in Saudi Arabia even before Khashoggi. Opgehaald van YouGov: https://today.yougov.com/topics/politics/articles-reports/2018/10/17/america-had-limited-trust-saudi-arabia-even-disapp .

As such, the Groseclose and Milyo research is important to keep in mind as a methodological inspiration, rather than a challenge to the underlying assumptions or the research question itself. Finally, the interesting question that arises in regard to their research is whether they really measure a bias in the media when they compare the sources used with sources used by the politicians, given that Page and Benjamin find that economic elites have much higher impact on policy change as compared with the average voter. Gilens, M., & Page, B. I. (2014). Testing Theories of American Politics: Elites, Interest Groups, and Average Citizens. Perspectives on Politics, Vol. 12, No. 3, 564-581 At the same time, think tanks are themselves special interests, and as such, Groseclose & Milyo are really measuring ideological biases within the special interests, and ignore the wider public Groseclose, T., & Milyo, J. (2005). A Measure of Media Bias Author(s): . The Quarterly Journal of Economics, 1191-1237.

Policy-Making Theories

In this chapter we will lay out the primary theoretical frameworks used to analyze stakeholder influence on policy-making in democratic systems. We start with a short overview of the theories, and the body of research testing these theories which were conducted in the US.

The Political Science on stakeholder influence of US Policy development has been widely researched, and as such provides the most comprehensive body of research in this field. The primary theories, which dominate the academic discourse on stakeholder influence on policy development in democratic systems include the “Majoritarian Electoral Democracy, Economic-Elite Domination, and two types of interest-group pluralism - Majoritarian Pluralism, and Biased Pluralism” (Gilens & Page, 2014 Gilens, M., & Page, B. I. (2014). Testing Theories of American Politics: Elites, Interest Groups, and Average Citizens. Perspectives on Politics, Vol. 12, No. 3, 564-581).

Now we will look at each briefly, to understand how the academia predicts and measures the impact of different stakeholders on policy development. As the primary focus of this research is to understand media bias on foreign policy, it is crucial to understand how policy is made in the first place. For this, we will look at basic models by which policymaking is researched, with a focus on different interests and how they are represented. This, in turn, will help us understand the conditions under which policy (including foreign policy) in the United States is influenced by special interests, ultimately linking the special interests to the media itself.

The Majoritarian Electoral Democracy model was developed primarily by the fusion of the traditional theories of democracy (as stemming from the early-modern thinkers, such as de Tocqueville Who was quite fearful of the tyranny of the many, stating that “Society is endangered not by the great profligacy of a few, but by the laxity of morals amongst all.” Alexis de Tocqueville, Democracy in America ) and Rational Choice theories. This theory postulates that policies are developed as a result of parties' desires to be (re-)elected, and who seek to propose policies based on the median-voter principle (the party wins elections if it can capture the median voter, in addition to all those on their side of the spectrum of debate). This theory has been primarily criticized for the fact that it assumes a single-issue based model, while real politics tends to have a multitude of issues of interest to the public (Gilens & Page, 2014 Gilens, M., & Page, B. I. (2014). Testing Theories of American Politics: Elites, Interest Groups, and Average Citizens. Perspectives on Politics, Vol. 12, No. 3, 564-581). By contrast, empirical research has been undertaken based on this theory, and it is still regarded as a useful tool for understanding policy setting in democracies.

Particularly important is the book “Macro Polity” which tries to empirically test the idea that public preferences have a direct impact on government activity over time. “Borrowing from the perspective of macroeconomics, it treats electorates, politicians, and governments as unitary actors, making decisions in response to the behavior of other actors” and hence “argues that the electorates and governments that emerge from these analyses respond to one another in orderly and predictable ways” over long periods of time (Erikson, Mackuen, & Stimson, April 2002

Erikson, R. S., Mackuen, M. B., & Stimson, J. A. (April 2002). The Macro Polity. Cambridge: Cambridge University Press.). While the book was criticized on a number of levels, such as small sample sizes in some variables (Morris, 2004

Morris, I. L. (2004). Review: The Macro Polity. The Journal of Politics, 304-306.), for this discussion it is more important to note that there is a general lack treating campaign spending in the book (Morris, 2004 Ibid.), as well as the power of “issue framing” by elites, and particularly by the political parties and the media.

Coming from the field of Sociology, the Economic-Elite Domination Theory predicts “asymmetrical influence between economic units… [where] asymmetrical influence [is] exerted by economic firms, industries, organized groups, the public sector, and nation-states… and complete balance of power is seldom achieved in the social universe because of the imperfect reciprocity of the actions and reactions of human agents and groups.” (Bocage, 1977 Bocage, D. (1977). The Sociological and Institutional Background of Perroux's Economic Domination Theory. The American Economist, vol. 21, no. 1, , 39-43). As such, this theory stands in stark contrast with the Majoritarian Theory, since it takes a more nuanced approach to the distribution of power in societies, including democratic societies. Here we start seeing how different actors can have a disparagingly high impact on policy setting, based on their social and economic status in society.

One of the earlier conceptions of the Economic-Elite Domination Theory come from the seminal work of the writer C. W. Mills in his book “The Power Elite” where he traces psychological, social, and other factors which lead elites to have a strong “class consciousness”, i.e. a sense of unity in goals and motivations, while also maintaining surface differences in specific issues:

“All the structural coincidence of [elite] interests as well as the intricate, psychological facts of their origins and their education, their careers and their associations make possible the psychological affinities that prevail among them, affinities that make it possible for them to say of one another: He is, of course, one of us. And all this points to the basic, psychological meaning of class consciousness. Nowhere in America is there as great a 'class consciousness' as among the elite; nowhere is it organized as effectively as among the power elite. For by class consciousness, as a psychological fact, one means that the individual member of a 'class' accepts only those accepted by his circle as among those who are significant to his own image of self.” (Mills, 1956 Mills, C. W. (1956). The power elite. New York: Oxford University Press)

With such psychological, economic, and social affinities among the elites, Mills goes on to assert that this leads to an institutional type of convergence of interests (like in conventional lobbying, where organized groups push for policy), but even more importantly to informal convergence of interest (informal lobbying). The second, as Mills explains can be seen as “'high-level lobbying.' [as] all over the country the corporate leaders are drawn into the circle of the high military and political through personal friendship, trade and professional associations and their various subcommittees, prestige clubs, open political affiliation, and customer relationships” Mills, C. W. (1956). The power elite. New York: Oxford University Press.

This theory, as it is highly nuanced in terms of different actors, groups, and institutions, which often overlap in membership, makes it more difficult to measure in quantitative terms. As such, it can be criticized for being untestable, having many moving parts, and not very good at predicting change. Nevertheless, empirical work has been done, both explicitly by proponents of the theory See “Ferguson, Thomas. “Industrial Structure and Party Competition in the New Deal: A Reply to Weber.” Sociological Perspectives, vol. 34, no. 4, 1991, pp. 493-526.” and the `Institutionalists', such as the work by Daron Acemoglu and James A. Robinson, who show how elites dominate certain institutions (such as colonial elites), and how stability and change occur in different institutions over time, opening up the discussion for a more comprehensive understanding of elites through the lens of institutions (Acemoglu & Robinson, 2002 Acemoglu, D., & Robinson, a. J. (2002). Why Nations Fail: The Origins of Power, Prosperity and Poverty. New York: Crown).

Finally, Bealey further diversifies the Theory by looking at the relative importance of certain elites in comparison to others, arguing that some elites need to be more independent than others, since they have values which foster democracy (unlike others, which impede it). As such societies will be more democratic if they support those:

“"key elites" which uphold the principles of free expression and assist citizen participation in the democratic process-parliaments, political parties, the academic-intellectual elite, and the media-autonomy becomes much more vital… The Thatcher government's attack on the BBC, the universities, and local government (not hitherto mentioned but surely an autonomous source of power), while not significantly successful, gave cause for concern. Perhaps of even more concern was the lack of public indignation at these attacks on such elites: their autonomy is vital for democracy.” (Bealey, 199 Bealey, F. (1996). Democratic Elitism and the Autonomy of Elites. International Political Science Review / Revue internationale de science politique,Vol. 17, No. 3,, 319-3316).

Finally, two versions of interest-group pluralism - Majoritarian Pluralism, and Biased Pluralism explore policy making as neither individualistic rational choices nor as hierarchical processes, but in terms of special interests - factions. Majoritarian Pluralism assumes that a society with many interest groups (economic, identity, social, etc.) will lead to “struggles among the diverse factions that would be found in an extensive republic would lead to policies more or less representative of the needs and interests of the citizenry as a whole - or at least would tend to defeat "tyrannical" policies” (Gilens & Page, 2014).

This theory is criticized both by formal political theory (e.g. game theorists argue that free-rider effect will diminish the participation in groups representing large sections of society, while small groups may more easily get their way) and by empirical analysis, as done by many researchers looking at marginalization of some groups in society (e.g. racial and ethnic minorities, the poor, etc.). Because of these objections, the second variant of the pluralistic model was developed, i.e. the Biased Pluralism. This variant of the theory adds relative power of different groups into its calculations. Several tools were developed based on this theory, such as the impact of intensity, visibility, direction and scope of issues in society, where elitist interest groups, such as businesses, have disproportionate power when the scope of conflict is small, and visibility is low, as the public is not mobilized easily (Schattschneider, 1957).

To compare the different theoretical models, the authors measure the impact of the different income levels of survey respondents and their policy preferences, to measure its impact on actual policy change. The survey respondents' policy preferences are divided into the “average citizens policy preferences (preferences at the fiftieth income percentile); the policy preferences of economic elites (measured by policy preferences at the ninetieth income percentile); and the stands of interest groups (the Net Interest-Group Alignment Index)” (Gilens & Page, 2014 Gilens, M., & Page, B. I. (2014). Testing Theories of American Politics: Elites, Interest Groups, and Average Citizens. Perspectives on Politics, Vol. 12, No. 3, 564-581). Subsequently, to distinguish between the Majoritarian Pluralism and Biased Pluralism, the authors look at the differences between mass-based groups and business groups.

The paper finds that Majoritarian Theories have substantially less causal impact than the other theories. The authors add that when the “preferences of economic elites and the stands of organized interest groups are controlled for, the preferences of the average American appear to have only a minuscule, near-zero, statistically non-significant impact upon public policy” (Gilens & Page, 2014 Gilens, M., & Page, B. I. (2014). Testing Theories of American Politics: Elites, Interest Groups, and Average Citizens. Perspectives on Politics, Vol. 12, No. 3, 564-581). Furthermore, the interest-group system does not represent the median individual either, with business interests negatively correlated with the average citizen's preferences. The preferences of economic elites have much higher impact on policy change as compared with the average respondents.

Finally, Biased Pluralism theory is supported by the results more so than for Majoritarian Pluralism theory as “both mass-based and business-oriented interest groups are positive and highly significant statistically, but the coefficient for business groups is nearly twice as large as that for the mass-based groups” (Gilens & Page, 2014 Gilens, M., & Page, B. I. (2014). Testing Theories of American Politics: Elites, Interest Groups, and Average Citizens. Perspectives on Politics, Vol. 12, No. 3, 564-581). As such, the results of this research article informs us that policy making in modern democracies can, and often is, disproportionately influenced by elites, followed by business interest groups, then mass-based groups and finally the median income population. Considering the authors clearly show the level of impact that elites and special interests have on public policy, we can assume that this will hold true for foreign policy too.

This said, the Page and Benjamin article only answers the questions of which actors' preferences are most likely to be implemented, but it does not answer the question of how these issues are shaped and the power of elites to shape public opinion itself.

“Of course our findings speak most directly to the "first face" of power: the ability of actors to shape policy outcomes on contested issues. But they also reflect - to some degree, at least - the "second face" of power: the ability to shape the agenda of issues that policy makers consider. The set of policy alternatives that we analyze is considerably broader than the set discussed seriously by policy makers or brought to a vote in Congress, and our alternatives are (on average) more popular among the general public than among interest groups. Thus the fate of these policies can reflect policy makers' refusing to consider them rather than considering but rejecting them. (From our data we cannot distinguish between the two.) Our results speak less clearly to the "third face" of power: the ability of elites to shape the public's preference” (Gilens & Page, 2014 Gilens, M., & Page, B. I. (2014). Testing Theories of American Politics: Elites, Interest Groups, and Average Citizens. Perspectives on Politics, Vol. 12, No. 3, 564-581)

These three levels of influence on choosing policies, shaping the agenda, and influencing public opinion itself pose a conceptual and empirical problem for any study of public policy. Particularly in studies of media influence on public policy it is important to clearly distinguish between these levels. As such, we will return to the three levels of influence on public policy in our discussion of how the media themselves may be impacted by elites and special interests.

This body of research is key in terms of theory and empirical research findings of the influence of elites on policy making and media biases in democracies. The theoretical frameworks presented below are the primary foundation upon which the research question was formed. By looking at these theories, specific testable hypotheses were formed. The discussion also looks at empirical findings, in order to determine the frontiers of research, and hence ensure the research adds to the overall research in these fields. While the research focuses specifically on the United States, the aim of the paper is to test some of the prominent theories, and see how well they can explain some of the phenomena observed in this country, with the hope that this case can further inform the development of the theories for media in democracies in general.

Conclusion of the Literature Review

The overall paper looks at underlying structures of public life that leads to media biases, through the lens of structural analysis. While a significant corpus on internal biases exists, a larger picture of deep causal links is absent. As such, this paper looks at two proximal causes for the overall bias in the US media. The first part looks at social networks that link special interests to media companies, and proposes an institutional and structural web of influences, based on ownership and closeness between different actors in the network, and proposes that this leads to both direct power to control media through shared voting seats on boards of directors, and through information flow choke points - i.e. where media companies save on resources that would go into research of topics by relying on ready-made studies made under the auspices of special interest funded and controlled think tanks. The second part explores how the information flows actually influence reporting, by comparing different narratives of media and key think tanks.

These proximal causes can then be linked to the distal, i.e. ultimate causes, which represents deep structural mechanisms of influence, control, and power in society. These distal causes can then be better understood, such as how does private ownership of media influence their public interest mission? It is widely recognized that lobbying is a legal way of accessing policymakers to try to impact policy, and this probably works, if we take into account that expenditures on lobbying have increased in the US over time, but empirical studies also seem to confirm this increasingly (GRASSE & HEIDBREDER, 2011 GRASSE, N., & HEIDBREDER, B. (2011). The Influence of Lobbying Activity in State Legislatures: Evidence from Wisconsin. Legislative Studies Quarterly, 567-589 and Richter, Samphantharak, & Timmons, 2009

Richter, B. K., Samphantharak, K., & Timmons, J. F. (2009). Lobbying and Taxes American Journal of Political Science . American Journal of Political Science, 893-909.).

Finally, lobbying is not limited to domestic issues, but as Skonieczny (Skonieczny, 2017) Skonieczny, A. (2017). Corporate Lobbying in Foreign Policy. Oxford Research Encyclopedia of Politics. concludes: “[in] summary, corporate lobbies in particular often have a huge stake in foreign policy decisions. Whether the issues relate to the opening of markets abroad, environmental regulation, intellectual property rights, or the size of the defense budget, corporations see engagement in foreign policy decisions as critical to business success in a globalized world.” Ibid. As such, we can see that special interests are in the field of foreign policy, and wield significant influence.

Yet it is much less recognized that media companies, which are predominantly for-profit, can also be influenced or enticed by corporate power. The editor-in-chief, producers, and other top management and journalistic professionals are, afterall, hired by the president of a media company, who is appointed by the CEO, who is hired by the board of directors. This top-down structure opens up a way of direct influence of media by ownership, despite the claimed independence of the editorial from the commercial. As Mat Thaibi shows, however, in today's editorial rooms, advertisement and commercial interests are present and have significant influence - countering the 1970's standard “that if you were in the sales force [advertising department], you weren't even allowed on the editorial floor”, which changed with deregulation of media, and a changing business model which aims to maximize profits over all else (Taibbi, 2019

Taibbi, M. (2019, February 18). The News Media and Manufacturing Consent in the 21st Century. (D. Kofinas, Interviewer)).

With the merger of the sales and editorial, the CEO / president gains much more power over the editorial policy, which was significant even before these changes. As such, the network analysis will link power that can potentially be exerted through shared board of directors' seats, while the case studies of discourses in media will try to show how narratives are shaped through information sources that are also linked to these special interests, and which reduce information costs to the media by providing ready made opinion, research and other forms of information. In exploring these two proximal causes, the research will attempt to hint on the deeper distal causes of bias in US media, particularly of ownership and economic incentives. In the following section, we will present the methods, methodology, analysis and results of the research, using two approaches: a social network analysis and a case study of media narratives.

Methods and Methodology

The research methodology was designed as twostep process, that will allow us to answer the research question:

Does the media show a bias in favor of special interest narratives in its reporting on foreign policy of the US government?

In order to answer this question, we need to show (1) that special interests have access to media, and hence can insert it's interests and narratives as hegemonic in the mass news media field; and (2) to show that narratives presented by the media do, in fact, favor the interests of those special interests. The first step involves a social network analysis of media and special interests, in order to test the following hypothesis:

Hypothesis 1: top military and oil industry special interests have access to all major media companies through board membership

In the second part, we will look at the media narratives themselves, to see if they show clear bias in favor of the interests of the special interests, which we define more precisely below, by testing the following hypothesis:

Hypothesis 2: media reporting is biased towards special interests' objectives in foreign policy, by over-emphasizing foreign government abuses in states closed to US special interests, as compared to open states

Before proceeding, it is important to define the term “bias”. To do so, we will look at the definition presented by Groseclose and Milyo:

“Most important, the definition has nothing to do with the honesty or accuracy of the news outlet. Instead, our notion is more like a taste or preference. For instance, we estimate that the centrist United States voter during the late 1990s had a left-right ideology approximately equal to that of Arlen Specter (R-PA) or Sam Nunn (D-GA). Meanwhile, we estimate that the average New York Times article is ideologically very similar to the average speech by Joe Lieberman (D-CT). Next, since vote scores show Lieberman to be more liberal than Specter or Nunn, our method concludes that the New York Times has a liberal bias. However, in no way does this imply that the New York Times is inaccurate or dishonest just as the vote scores do not imply that Joe Lieberman is any less honest than Sam Nunn or Arlen Specter.” (Groseclose & Milyo, 2005 Groseclose, T., & Milyo, J. (2005). A Measure of Media Bias Author(s): . The Quarterly Journal of Economics, 1191-1237)