Media Influence on Public Policy in Kenya: The Case of Illicit Brew Consumption

The media plays a major role in public life, influencing citizens’ focus of attention and providing many of the facts and opinions that shape perspectives on topics of the day. Studies that analyze media influences on policy making have had mixed results on whether the media does actually influence public policy. This case study analyzes the role of the media in Kenya in influencing public policy in relation to illegal brew sale and consumption. The study seeks to understand media coverage mechanisms that led to government actions on an issue. This was done through content analysis of the coverage in two leading daily newspapers over a period of 10 years (2005-2015). The analysis found that intense, congruent, and incident-driven media coverage had influence on government action and consequently public policy on the issue. The findings also found weaknesses in media reporting. The media was very good at telling the politicians what was happening and what to talk about, but they were less able to tell the politicians what to do, which compromised the watchdog role of the media in Kenya. These findings can also be useful for advocacy groups and the public on the ways to engage the media to influence policy makers’ decisions.


Introduction
Media plays a major role in public life, influencing citizens' focus of attention and providing many of the facts and opinions that shape perspectives on topics of the day. The mass media can be considered a mechanism by which to communicate symbols and messages to the public at large (Herman & Chomsky, 1988). The public and policy makers in government at all levels consistently receive messages about what is really important in the world of public affairs. The result of this has been that public affairs issues that are prominent in the press frequently gain importance among the public and government officials (Koch-Baumgarten & Voltmer, 2010).
The study of the relationship between media and policy is fraught with complications related to causation. In developed economies, policy makers are highly sensitive to media coverage of their actions and activities (Campbell, 2008). Political insiders have typically been well aware of the influence the media has in political decision making. However, scholars in political science find it quite difficult to define how the media influences policy except in clear narrow issues. Walgrave and van Alset (2006) reviewed 19 studies published over the last three decades that analyze media influences on policy making and found mixed results on the effects of media and policy making; therefore, more research is needed to explore whether the media ever influences policy making and this research attempts to contribute to the literature on this issue. Another aim of this research is to contribute to literature on media and policy framing. An evaluation of media frames and policy frames will provide opportunities to nuance what possible effects framing has on the policy agenda.
This study seeks to understand the mechanisms that lead to the influence of media on public policy. The research question is "Does print media influence public policy?" and looks at a Kenyan case.
Public policy consists of government decisions and actions that deal with matters of public interest (Cochran & Malone, 2005). In this article, we analyze government response to a particular dominant issue in the media and seek to find out whether this appears in any formal policy formulation process and then investigate how these decisions are eventually shaped in action. Influence can be described as having an impact in the processes and outcomes of political decision making.
We examine one economic and social issue in the Kenyan public domain: illicit alcohol. This issue was selected because it has been a long-standing debate in the Kenyan Special Collection -Interest Groups 764245S GOXXX10.1177/2158244018764245SAGE OpenMwangi research-article20182018 1 Strathmore Business School at Strathmore University, Nairobi, Kenya political and economic sphere, relevant not only at the national level but also having impact in rural areas where many Kenyans live and run their businesses. Similarly, the impact of illegal alcohol trade on consumption, expenditure and saving is clearly understood by the "average" Kenyan and requires no particular expertise to form an opinion.

Media and Public Policy
The media plays an important role in modern political and economic life. Scholars have defined the media as "guardians of the public interest" (Sparrow, 1999, p. 282) who champion "truth, pluralism, objectivity, balance and accuracy" (Bennett, 1990). However, often the environment in which the media operates encourages journalists to exploit sensationalism at the expense of the broader socioeconomic issues due to the need to beat competition and acquire favorable ratings and advertising revenue.
This "market" environment includes the media among other strategic actors in the policy debate, in attempts to redefine, alter, or "spin" political rhetoric (Hallin, 1992). This also places the media in a unique position, where they can define new frames with which to analyze issues inasmuch as they are a conduit to communicate frames presented by other players.
One matter of concern that contributed to this study of the relationship between media and politics and policy stems from the effect that the media had in the 2007 elections that led to a significant wave of ethnic violence and hatred and over 5,000 deaths. This was primarily through partisan community radio stations that fueled tensions that culminated in violence (Ngui, 2008).
This effect in one way demonstrated the power of the media in influencing the actions of a certain group of individuals to a point where they caused actual harm to their friends and neighbors. It appears in this case the media had a direct impact on what these people thought or perceived about their economy, country, and friends. Conversely, there was also a relationship between politicians' use of the media and the reactions of the public. Television and radio platforms were used to insinuate that they (the politicians) were acting (agitating for change-leading to incitement) on behalf of the disenfranchised Kenyan. Although this is not directly related to policy making, it woke Kenyans to the power of the media in democratic processes.
Scholars have focused significantly on policy makers responding to media as a manifestation of public opinion (Molotch, Protess, & Gordon, 1987). Policy makers assume that the public is heavily influenced by what they read in papers or watch on television, and therefore perceive that adapting to the media would actually be responding to what they think are the expectations of the electorate, resulting in the media actually having an impact on policy makers' actions (Herbst, 1998). Sometimes the response will be symbolic and not substantive. It is the substantive action, such as shifting policy positions, actual policy action taken, amendments made to existing policies or intended policy actions are dropped altogether that would be interesting to observe in this study and consequently termed as influence.
Unfortunately, this is not always a straightforward process. Gavin (2010) explained that media content itself can be influenced by audience demand, and thus journalists are highly sensitive to what the public wants to read or know about, so they judge content by news value. The public itself can decide what it wants covered. Other studies argue that even if it is possible to show a causal connection between media activity and policy developments (Walgrave & Van Alset, 2006), it is possible that the policy changes result from the actions of the interest groups' involvement and not the media action (Gavin, 2010). This then confounds the causal relationship between media and public policy. So Robinson (2001) recommended examining the circumstances in which media may influence policy. He hypothesized that media is likely to have influence on policy processes in times of policy uncertainty. Robinson identified one source of uncertainty as ambiguity between policy subsystems. He defined uncertainty as the inability to analyze the present situation or predict future consequences due to inadequate knowledge, ignorance, or imprecision. Ambiguity is defined as vagueness about intended meanings, multiple interpretations, or ambivalence (Zahariadis, 1999). Robinson suggested that differing interpretations of the same problem or solution in different departments could give rise to uncertainty in the form of ambivalent communication or dissent, and this weakness could provide the media an opportunity to influence policy action. It is expected therefore that with regard to the role of the media in the policy process, their main influence may be in policy debate where different perspectives are presented and interrogated. In such cases, a strong case may possibly have influence.
Agenda setting and framing are the most elaborated approaches that link the study of the content of political news to the knowledge, attitudes, and behavior of mass audiences. This body of research has shown that the salience (alignment with personal interest) of issues on the media agenda not only affects which problems citizens consider most important but also their policy preferences and how they evaluate political officials (Iyengar & Reeves, 1997;McCombs & Shaw, 1972). Salience and agenda-setting studies demonstrate that the more attention there is on an issue, the more the public is aware of it and the stronger the incentive there is for policy makers to frame a response to the issue.
Framing theory explores the way in which political information is contextualized and how this affects the meaning of a news story and ultimately the judgment of the audience (Price, Tewksbury, & Powers, 1997). Described frames as "interpretative packages" that include assumptions about causes and consequences of a problem, affected values, and resulting policy implications. These observations are in line with Kahneman and Tversky's (1984) experimental study that demonstrates that the framing of policy alternatives as either "avoiding risks or achieving gains" strongly affects individual preferences.
Framing effects occur when an issue can be presented using multiple lenses or thematic slants. For instance, home brewed liquor in Kenya is labeled as "killer brews" or "traditional brews," which would evoke different emotions in the public: One would be interpreted a national safety crisis or illegal activity, and the other would be perceived as preservation of indigenous cultures and practices.
We aim to contextualize theories on agenda setting by focusing on issue frames in media coverage and on the policy agenda. In this study, we analyze media coverage and policy documents in terms of "frames" of a policy issue.
Here we assume that a limited number of possible frames of a certain issues exists that can be promoted in media attention and policies addressing the issue. We also assume that in most newspapers, one dominant frame will prevail, even though the issue can be portrayed from different viewpoints and by different actors in one message.
The impact of media also depends on the content of media coverage: the intensity, tone, framing, and consonance. Topics that arouse emotional reactions are likely to have more impact than those that do not. The intensity of published reports and the more similar the messages of different news reports (consonance), the more impact they are likely to have (Linsky, 1986).

Method
One specific area of focus in economic policy is studied: alcoholic drinks control in Kenya. This has been chosen due to the fact that alcohol consumption and regulation is a social and economic issue that affects a population in general and is not targeted to unique communities or ethnic groups.
Print media content analysis was employed to analyze the data and establish the frames. Media content analysis is a recognized specialized subset of content analysis, which is a wellestablished research method. This method was first introduced by Harold Laswell (1972) to study propaganda. Process tracing was used to identify the impact of these frames on policy.
In this particular case, the content we are studying is print media from two leading newspapers in Kenya (The Standard and the Daily Nation who have the highest circulation and readership in the country). The Nation Media Group publishes the Daily Nation and the Standard Group publishes The Standard. These two media houses have close links to the Kenyan business elite and political class, and although these media houses are quite vibrant in playing their watchdog role, they are still at risk of elite capture and their criticism of government activity is typically within bounds.
The period of analysis selected was 2005 to 2015. This was for reasons of data availability and also to compare the responses of two different governments in place. The Kibaki government ran from 2003 to 2012, and the Kenyatta government has been in power since 2013. Every single newspaper over 10 years was thoroughly scoured for any article on the topic. All efforts were made to increase reliability of frames identified by using NVivo to support the coding process. This software enabled the researchers to code the data and retrieve text based on key words to form "frames." Therefore, the frames are not necessarily "found" by the researchers but are computed by the software. The software makes it possible to track the evolution of the data analysis.
The study used the same newspapers to analyze whether any government action, decision, or declaration on the issue had any similarity to what was reported in the newspapers. The Kenyan parliamentary records were also referred to, to analyze debates and identify any overt references to the media. The policy decisions could have been in the form of executive orders, parliamentary discussions, or other policy outputs or decision-making processes that were used by the government to respond to the issue.
In this study, we will use Dahl's (1961) definition of influence that A must make a difference and in consequence affect B to illustrate influence. The media focus and attention only has influence if it leads to action. These actions could be a budget increase, sackings, and regulation on local or national level. To demonstrate influence, the study will seek to identify policy makers and politicians proposing or promising actions relevant to the issues and check the implementation of these proposed actions. Politicians are known to make promises that they do not keep; in cases where promises are made but not kept, the media can be said to have "pseudo-influence." This study will demonstrate causality by using the process tracing approach. Process tracing enables a researcher to focus on unfolding events over time in a descriptive manner by observing changes and sequence in a situation, all the while identifying key events in time, consequently enabling one to provide a good analysis to illustrate causation (Collier, 2011). This study looks at the evolution of the news coverage over a 10-year span, providing a rich set of data to work with and opportunities to investigate policy makers' behavior spanning two governments.

Limitations
However, we cannot know whether these policy decisions would have happened "anyway" without any media intervention, but great focus is put on using process tracing to demonstrate causal links and explanations that would support any conclusions. The research attempts to address this limitation by employing the process tracing approach that demonstrates cause and effect.

Findings
In a speech, Hon. John Mututho (2014), the Chairman of the National Authority for the Campaign Against Alcohol and Drug Abuse (NACADA), admitted that alcoholism problems have been a struggle for Kenya even as early as 1902 when serious challenges with the drug were experienced in Fort Hall District in 1902 (current day Murang'a County).
The Colonial British Administration, through M. Francis Hall, appointed a Chief named Wangu Wa Makeri on recommendation of senior chiefs in Central Kenya to assist in eradicating the alcoholism. The successes of these efforts led to the enactment of a Prohibition Act and African Drinks Control Act that was meant to control indigenous alcohol production and consumption.
After the colonial era, there has been a checkered history of regulating traditional alcohol consumption in Kenya. President Moi in 1978 banned the production and sale of traditional brews, stating that they retarded development. However, after President Mwai Kibaki's government took over in 2003, Hon. Amos Kimunya, the Finance Minister, made history by introducing the Licensing Laws (Repeals and Amendments) Act (2006), which repealed the Traditional Liquor Act of 1971. Hon. Kimunya came up with one law, the Liquor Licensing Act, which broadened the licensing of traditional brews and enabled the mass production, bottling and sale of these "traditional" products countrywide. This law came into effect on May 1, 2007.
The Act defined Traditional Liquor as ". . . any intoxicating liquor manufactured by traditional African methods other than distillation, which is offered, or intended to be offered, for sale in a state of continuing fermentation without further processing" (Liquor Licensing (Repealed) Act, 1957, p. 4).
This excluded Chang'aa, a popular homemade cheap alcoholic drink, from the definition of a traditional brew, because it undergoes further processing through distillation. The law made it illegal to brew because it neither followed the regulatory guidelines of beer or alcohol processing nor was it approved by the Kenya Bureau of Standards.
However, this understanding was not obvious in government agencies. Different reactions to this change revealed that government agencies themselves had understood that traditional liquor including Chang'aa could be brewed at home, as long as one had a license.
In May 2007, following the enactment of the law, media attention on increased consumption of alcohol spiked and it was evident that the actions taken by Finance Minister Amos Kimunya to allow the manufacture and sale of traditional brews led the country through a downward spiral. Prior to that, there were no articles or major reports or concerns about alcohol consumption in the country, but coverage intensified after 2007 as in Table 1.
Media reported that the government had allowed the brewing of Chang'aa (Amran, 2007;Anyuor & Martin, 2011). This could have been influenced by communication from the then Deputy Nairobi Provincial Commissioner who had explained to journalists that there were licenses available to brew traditional liquor at a cost of KShs 500 (US$5) and those without licenses could get them on the grounds that in various parts of the country, the licenses were being provided. The Provincial Commissioner also at the time reportedly felt that the move to charge for brewing licenses was welcome and would increase government revenue through legitimate licensing and taxation of these brewing activities (Njagih, 2008).
The government approach to dealing with the repealing of this Act seemed inconsistent. Following the media's exposure of conflicting government positions on the manufacture of traditional brews, the Minister of Finance clarified his position in Parliament, stating, "The Government has not, and I repeat, has not authorized anyone to start brewing traditional liquor unless they follow the law as captured within the Liquor Licensing Act . . ." (Government of Kenya, 2007).
The reason why Hon. Kimunya wanted to disassociate his government with Chang'aa brewing was because unlike traditional brews which are made from fermented fruits and undergo no further distillation process or chemical additives, Chang'aa is a mixture of ethanol, methanol, and other deadly intoxicants. It has no formal recipe and is formulated using chemical additives that are subject to regulation.
Another problem Kimunya faced in 2007 was that the Liquor Licensing Act he had introduced had no clear distinction between processed beer and traditional beer, and in fact the word" Traditional" brew appears only in the definition of terms. The Liquor Licensing Act defines itself as "an Act of Parliament to make provision for regulating the sale and supply of liquor, and for matters incidental thereto and connected therewith." (Liquor Licensing (Repealed) Act, 2007, p. 1) The brewing requirements of the Traditional Liquor Licensing Act of 1971 were very strict, and there were stipulated guidelines on how to get authorization to brew traditional liquor. There were also clear guidelines on health and safety, which indicated that only premises approved by the medical officer of health for the area could be granted a license.
This was not the case in the new Liquor Licensing Act. The repeal of the Traditional Liquor (Repealed) Act (1971) made the safety guidelines redundant, and there was no clear regulatory framework to sit in its place. Media coverage focused heavily on specific themes when discussing the issue of alcohol consumption in the country. Newspaper articles represe different perspectives, several times doctors, public affairs consultants gave their technical positions on matters of addiction focusing on the responsibility of brewers big and small, in being responsible about advertising to impressionable youth. There was also a lot of focus on the importance of civic education on dangers of excessive consumption of alcohol and underage drinking.

Key Media Frames on the Alcohol Debate
This research provides an interpretive account of the media frames identified from the data and considers broader cultural elements. Several scholars have used this approach in studying media frames (Haller & Ralph, 2001) .These frames can be considered as "categories" that classify the text. These observed frames are experiential and contextually contingent rather than empirical, and careful description is provided to validate this approach.
Key words in the textual analysis demonstrated the areas of focus for the media in reporting the issues. Children were a key concern, and the categorization of the alcohol being discussed as illicit or illegal was significant. The prominence of regulatory agencies being mentioned in the print media versus other stakeholders such as mainstream beer production companies such as East Africa Breweries Limited was also notable. This had the effect of creating a conflict frame that pitted the effects and negative outcomes against lax government action. Productivity as a theme was often cited as well as the "youth." The mention of specific regions in the country was significant because it demonstrated the media focus on the disproportionate effects of alcohol use and/or abuse. Details of the occurrence of key words in the 10-year period are provided in Table 2.
Specific themes that were pulled out of the data and categorized above are described below.

Social Fabric, Morality, and Families Compromised
Alcohol was depicted as a vice that causes disruption to families (Gitonga, 2012;D. Okoth, 2009). Coverage of entire pages sometimes had nothing but pictures of seized illegal alcohol in residential areas. The use of images was rich and would depict men sleeping on the roadside in broad daylight, too inebriated to reach home. The plight of wives and mothers was emphasized with focus being on wives losing their husbands' attention and affection and mothers losing their children to alcohol addiction (Kemei, 2009). Domestic violence was also attributed to the increase in alcohol consumption. Children were also reportedly dropping out of school due to the negative effects of parents' alcohol abuse.

Regions: Central Kenya
The effect of illegal alcohol production and consumption was severe in Central Kenya, particularly in Kiambu and Murang'a Counties. There are reports of broken families, violence, thefts by addicts and declining population rates as well as an increase in the number of spinsters as a result of not having any appropriate suitors (Gikandi, 2008). This was depicted as a matter of grave concern to these communities because leaders were worried that their lineages, culture, and way of life may change as a result (Weru, 2009a). Murang'a town was particularly notorious for excessive consumption of alcohol because the licensing of liquor outlets was uncontrolled. This was directly mentioned by the media as having directly contributed to school drop-out rates and declines in population growth, leaving many kindergarten schools empty. The Kikuyu traditionally have been the largest tribe in Kenya, and numbers were linked to political power. The alcohol menace was seen as a direct threat to political supremacy, and many suspected the alcohol abuse prevalence in Central Kenya was politically motivated to reduce eventual political power as detailed below.

Alcohol Use/Abuse as "Political"
Another angle the media took was to politicize the issues by implying that the government was legalizing traditional brews to please the public and secure votes (Kendo, 2007). Another theme with a political angle is when the media covered a member of Parliament from Naivasha who claimed that Central Region has become the dumping ground for cheap liquor manufactured in the region, in an attempt to destroy its social fabric and weaken the region to destroy its ability to have a huge population of voters or even leaders (Jamah, 2011;Kiberenge & Kiarie, 2011;Tanui, 2014).

Poverty
It was often highlighted that the brewing dens were located in low-income areas, and the dangerous alcohol was mainly consumed by poor people who could not afford better beer. The general tone was that poverty was the cause, and the young people wasting away their time drinking would not do so if they had opportunities to engage in meaningful recreational and employment activities (Iraki, 2014;Kamau, 2014;Muchiri, 2014).

Productivity
Vision 2030 was referred to as an important goal for the nation and would be threatened by illegal alcohol consumption (Ombati, 2009). President Kibaki several times acknowledged that the situation was dire, and severe action needed to be taken by all leaders to ensure that the country's manpower does not waste away in Chang'aa dens (Njagih, 2010).

Police/Law Enforcement Corruption
Cases of police corruption are covered extensively and told in narrative form with real life stories told from a first-person perspective, depicting the challenges of operating traditional breweries and even premises where Chang'aa was sold. The media coverage demonstrated an industry that was enabled by the same law enforcement that is meant to crack down on it. Stories told of a disconnect between the Chiefs who live in these locations and are the representatives of the Office of the President, representing the law and the police who have powers to arrest and detain lawbreakers. Other stories focused on the authority responsible for spearheading the campaign against drug and alcohol abuse being caught up in fake licensing scams operating in its premises with its officers pocketing license fees and authorizing production of poisonous substances as beer (Munuhe, 2013).

Economic Activity
The coverage of the Chang'aa and traditional alcohol brewing in the country was presented from various angles. There was balanced coverage in terms of portraying it as a social ill, as well as representing it as a fully fledged industry that had support from the local communities it serves, particularly in low-income areas and slums all over the country. This industry was reflected as also providing employment in the form of daily wages to "scouts" looking out for police as well as laborers in the distilleries. The impression given sometimes was that the operations of these illegal alcohol distilleries were too sophisticated for the average slum dweller and that the Chang'aa business was protected by well-connected business people (Kareithi, 2008).

Youth/Underage Drinking
The newspapers covered stories of college students drinking away their youth and using alcohol to cope with the demands of school and life. The issues of peer pressure that perpetuate drug abuse in Kenyan schools were regularly featured. The frequency and intensity of the articles on teenage alcohol consumption implied that the drinking problem in the country was larger than anticipated. The media slant was one of the hopelessness that the youth feel in their future prospects due to dismal employment opportunities (Kithi, 2008;Ombara, 2008). The writing would be emotional and appeal to nationalism by stating, "A country that fails to invest in its youth is a country without a future; it is time to stand up and save the youth and our communities from self-destruction" (Ombara, 2008, p. 14). The responsibility for managing the youth and protecting them would be placed on parents, the government, and sometimes the entire Kenyan society.

Rehabilitation
The reporters would also focus on how rehabilitation efforts could complement the government's initiatives to prevent alcohol abuse. This coverage would focus on the need for funding of government agencies to support rehabilitation efforts (Mwakio, 2009) and even stories of people who have been successful in recovering from addiction and finding new preoccupations in income-generating activities like farming Lucheli, 2008).

Public Health and Safety
There were reports especially during spikes in media coverage of deaths that had occurred due to consumption of these brews. There were reported cases of people getting blind immediately after consuming these drinks (Ombati & Kareithi, 2010). Other reports had graphic explanations of poisonous and harmful additives in the brews to ensure potency and to "bewitch" their clients to beat competition (Wanyonyi & Muruka, 2010). Other reports highlighted the use of river banks-where many city low-income dwellers get their water supply from-as prime locations for makeshift breweries and these locations become hot spots for "crime and violence" (Muiruri, 2009). In keeping with the health and safety frame, the media also covered road accidents that were suspected to be caused by drunk driving. Although there was no hard evidence to determine the actual percentage of deaths caused by inebriated drivers. The most dominant media frames were those that focused on the effects of alcohol abuse among the poor as well as its impact on the social fabric and families. The health risks of alcohol brewing with regard to the environmental impact and human health were also dominant and used repeatedly in the print media. Images that were used frequently, reinforced these frames. Table 3 shows the number of articles on alcohol consumption and production across the years in this study.

Intensity of Media Coverage
It is important to note that the coverage and salience of the alcohol issue grew in significant across the years from 2007.
Media coverage on the woes of alcohol abuse had been increasing in salience and newsworthiness from 2007 to 2009 with intense coverage in 2009 of 74 articles in both newspapers on the issue of alcohol alone compared with 23 in 2007 and 29 in 2009. The coverage in the newspapers was consistently graphic, showing images of dirty unkempt environments that were filthy and unhygienic where these "traditional brews" were manufactured for public consumption. The media reports were consistently negative, graphic, and had emotive depictions of the effects of the illegal brews, using words like "raid," "den," "alcoholism tears the family apart," and similar heart wrenching titles. The media liberally used photographs of tanks of the illegal brew being poured out into roads after police raids on premises, men sleeping in trenches-too drunk to get home, and even of women protesting their husbands' absence from home due to alcoholism.
On Wednesday, May 20, 2009, The Standard ran a twopage special feature titled "Dens where police dare not patrol" that detailed the challenges the police were having in implementing the prohibition of Chang'aa brewing. From January 2009 leading up to May 2009, there were 30 articles on illicit brews which was more than the coverage on the topic in 2007 and 2008.
The NACADA, a government agency, promptly responded. Through an advertiser's announcement in the newspapers on May 29, 2009, NACADA explained its position on alcohol abuse to the public. This was only 9 days after the special feature covered by The Standard. The statement from the NACADA focused on the health and social cost of excessive alcohol consumption, particularly among the youth and women. The policy would be focused on restricting accessibility to alcohol, especially to the youth; mitigate risky behaviors such as drinking and driving; and protect at risk groups such as young people. The contexts of alcohol consumption, such as bars, clubs, and other settings, as well as alcohol product safety would also be targeted as harm reduction interventions to ensure the overall safety, security, and productivity of Kenyans. These were the exact ways in which the media communicated to the public on the alcoholism vice, as demonstrated in the categories in the previous section. Illicit alcohol was specifically mentioned, and in the statement, Busaa, Muratina, and Banana which are formally recognized as traditional liquors were categorized as illicit alcoholic beverages, in the same class as Chang'aa. The statement called for a new legislation declaring all the traditional liquors and Chang'aa illegal as well as imposing punitive fines and/or imprisonment for contravention of the law. In the same breadth, the notice indicated that the new law would "Empower the regulatory authority board to issue permits regarding all forms of non-commercial alcohol" (NACADA, 2009).
Local leaders' approach to the increased consumption of alcohol was clearer than that of National Government at times. Members of Parliament in various constituencies adversely affected by the consumption of illegal alcohol declared "war" against the producers and sellers of the alcohol and would urge the public to support their case. One member of Parliament from Imenti North declared that he would lead "door to door" campaigns against the brews because the Provincial Administration (Chiefs and District Officers) had failed to squash production and sale (Weru, 2009b). Some government officials like District Commissioners would declare that licensed bars would be closed if chiefs gave in reports that licensed establishments also acted as a retail outlet for illicit alcohol.
In some cases, the police and the provincial administration worked hand in hand to arrest individuals caught brewing illegal alcohol and would seize all the products. For instance, in Biomet central division in February 2009, 23,000 liters of the brew was seized and 150 people arrested. In Nakuru in May 2009, 31,000 liters of the brew was impounded by the District Commissioner. There were successes when the provincial administration and police worked hand in hand to crack down on illicit brews. The media vividly covered these stories with pictures accompanying the reports of these seizures.
One notable member of Parliament took serious action to combat the vice. Mr. John Mututho got government support in 2009 for a bill that was seeking to legalize Chang'aa and traditional liquor by repealing the Liquor Licensing Act and the Chang'aa Prohibition Act that had outlawed the production and sale of Chang'aa. This was the Alcohol Drinks Control Bill. The bill proposed that these drinks could be produced and consumed as long as they followed government regulations. The then NACADA Coordinator was interviewed by journalists on the decision to legalize all brews and was quoted in the newspapers saying: "since we cannot eliminate all illicit liquors, legalization of Chang'aa and controlling its production will enable many access to legally produced drinks at affordable prices" (Jamah, 2009b, p. 8). She also claimed the move would boost local investment and tax revenues, because traditional brews are estimated to contribute to at least half of the KShs 42 billion alcohol industry in 2009.
It was envisaged that the 2007 legalization of the production, consumption, and sale of Chang'aa would make it an honest business by ensuring that it followed government approved standards of quality health and safety. Industry stakeholders felt differently and were not supportive of this proposed bill. The perspective presented by the media implied that mainstream breweries did not support the law as it would erode their profits by increased competition (Jamah, 2009a). Although this was true to some extent, it is also fair to note that these heretofore unregistered brewers were not known by the Kenya Revenue Authority and did not have to meet government regulatory standards and so could sell their alcohol dirt cheap, with no regard for safety, unlike the registered companies and it would be unfair to deregulate the market without these considerations.
It is evident from the data and the themes that were emphasized in the media coverage that the bill was inspired by the persistent coverage of the media by publishing 74 articles in 2009 alone on the alcohol menace. This was an average of six articles a month on the alcohol issue alone. This was more coverage than the previous 2 years combined.
In August 2010, the Alcoholic Drinks Control Act (2010) was signed into law by President Kibaki and implemented in November 2010. The media reports following the enactment of this law were balanced with a mixture of hope and positive stories depicting a change in the sale and production of alcohol that would lead to a sober, safer, more productive society (Njagih, 2010;Orengo, 2010). The media also highlighted different viewpoints from stakeholders in the alcohol business.
Government officials regularly quoted, phrases, and stories from the media to support alcohol legislation. According to the Official Record of the National Assembly, in 2010 a minister while supporting the then Alcohol Drinks and Control Bill declared that ". . . if nothing is done [to stop the youth from drinking], we are going to lose generations to come"; he also said, . . . our population growth is rapidly changing because young men can no longer be productive. Our daughters have no men to marry! My worry is that Wahu, my daughter, will she ever get a husband . . . (Government of Kenya, 2010).
These are phrases that are heavily inspired by newspaper reports as this is the angle a lot of the reporting has taken, depictions of hopeless youth, frustrated wives, declining enrollment in kindergartens, and retarded population growth (Ayodo, 2007;Kemei, 2009). This suggests that media coverage is a source of information for members of Parliament and actually used media reports as evidence to support their policy positions. It appears that media reports enabled members of Parliament to demonstrate that public debate is translated into legislation, as well as to enliven the debate and make it recognizable to the public and emphasize arguments.
Journalists would report the facts and the news but rarely gave insight into the possible challenges the implementation of this law would face. The responsibility for alcohol control and regulation fell within multiple government agencies such as the Kenya Revenue Authority, NACADA, Kenya Bureau of Standards (KEBS), Ministry of Public Health, Provincial Administration, and the Kenya Police Service, and these agencies did not always work well together or even understand their role in fighting the illegal production and sale of unlicensed alcohol. The courts as well are a key enforcement mechanism because it has been customary for courts to give offenders minimum penalties and fines of as low as KShs 5,000 (US$50) because maximum penalties are defined by law and minimum penalties are decided by individual judges, creating a corruption loophole and a subversion of law. Courts at the national level also seem to be some lack of clarity of what can be prosecuted. A ruling in 2011 by the high court declared that being in possession of Chang'aa or consuming it was not a crime (Thuku, 2011). The only crime was selling. This also contributed to weakening enforcement of illegal production because possession was not a crime.
The contradictions in government policy made it difficult for media to provide accurate information on government policy and inadvertently led to misinforming the public on various occasions. The data proved that the media approach was to report the actual happenings from any corner of the country and from any stakeholder or during any "event." Investigations into deeper causes of the issues were lacking. The media focussed more on actual deaths, negative social outcomes associated with alcohol, almost taking sensational twists full page spreads of dismal circumstances and images of the public breaking into premises suspected of brewing or selling the brews. Analysis into the complex nature of issue in question is lacking, not looking at correlations between slums, police officers, and regulatory agencies. The media did not seem to have a clear position on what they wanted the public to do with the information they were exposed to. However, expert opinion was regularly sought from public health professionals, sociologists, psychologists, and even economists. This would provide some deeper analysis into the issue, more than the regular writers could provide.
The progress of the implementation of the Mututho law was given entire pages of newspaper coverage with pictures of revelers still enjoying their drinks in bars in residential areas, which was supposed to be illegal. Stakeholders such as the Pubs, Entertainment and Restaurants Association of Kenya (PERAK) would use the media to demonstrate that it is difficult to legislate people's "freedom" and that is why many of their members would have their bars open even after 3:00 a.m.-which was the mandated closing time for all bars and entertainment facilities. The media also highlighted the fight against a section of the law that required alcohol manufacturers to have large labels on all their products, and the courts supported the companies in barring the Internal Security Minister, the Attorney General, and the NACADA from implementing the Alcoholic Drinks and Control Act (2010; Kwamboka, 2011). There was a lot of interest in lawmakers and even President Kibaki to crack down on the poisonous brew trade (Ngige, 2011). There was still consistency in media coverage of the issue (104 articles in 2011), but it turned out that President Kibaki had made promises that he would take decisive action.
Kenya's local government system changed after the March 2013 elections, and the management of local affairs was structured under devolved government; this required that the provisions of the "Mututho law" had to be ratified and implemented by the individual county governments and their county assemblies, which are their lawmaking bodies. This also proved to be a substantive blow to the initiatives and gains made in fighting the menace because another layer of government procedure had been employed.
The impact of the media on the policy makers depends on the content of media coverage, its topic, intensity tone, framing, and consonance. The topics of illicit brews and alcohol consumption focused more on the negative social effects and displayed this through colorful language such as "filthy brewing dens," "peddling death," "and a drowning nation." These terms typically arouse negative emotional reactions and have a strong impact.
The clearest example that the print media in this case influenced public policy outcomes arose from the intense and daily coverage of the illicit brew menace in May 2014, following a number of deaths from illegal alcohol consumption (Muchiri & Nyawira, 2014). The articles covered by the media expressed the public's frustration with the government in dealing with the drinks menace. Numerous pictures of drunks on road sides, grieving families, and hospitalized beer patrons were juxtaposed against a government regulatory mechanism that was broken. This had happened before 2014, specifically in 2011 when the NACADA Chairman wanted alcoholism declared a national disaster in the wake of deaths in Central Kenya from the brews. The Kibaki government did not take any decisive action in terms of policy changes, sackings, or budget allocations. The media coverage was significant with 121 articles on the issue that year.
The Kenyatta government took decisive action, following deaths from consumption in May 2014, and significant media coverage by articles averaging more than two a day, there was the high-profile sacking of the CEOs of NACADA and the Kenya Anti-Counterfeit Agency. Other public officers also relieved of the duties were from the Provincial Administration and Intelligence officers who had failed to curb the spread of the illegal brews (Olick, 2014). This was done by the Interior Cabinet Secretary in consultation with the Inspector General of Police, the Health Cabinet Secretary, and NACADA Chairman. The government also ordered the closure of a company that traded in spirits that contained 100% methanol and sold two of the products that caused more than 50 people in May 2014 (Mbaka, 2014). It is important to note that Kenyatta and Kibaki are both presidents from the Central Kenya region, and this may have given them additional impetus to deal with illegal liquor in the country.
There are arguments that this would have happened anyway following the deaths, but there had been deaths in the years preceding 2010 and no decisive action such as this had been taken by the government. The months where there had been a clear spike in media coverage with articles on killer brews appearing almost daily, and with images of impounded warehouses full of alcohol and other human interest stories on the plight of people affected by impact of the illicit brew trade, it is not impractical to assume that the media attention did have an impact. This demonstrated a government that is sensitive to media coverage.
The media had continued to cover stories on the devastating social effects on alcohol abuse by writing 147 articles on alcohol over the 12 months from June 2014 to June 2015. This was an average of 12 articles a month on the issue. President Uhuru Kenyatta in July 2015 then issued a verbal directive for a 4-day sting operation to destroy all "Second Generation Alcohol" (Gikandi, 2015). President Uhuru, while issuing the order during a meeting with Central Kenya legislators who were considered to be the worst hit by the menace, stated that provincial administrators who were being lax in carrying out their duties would be dismissed and asked members of Parliament in the affected regions to collaborate with the Police Commanders to shut down the outlets selling liquor without valid licenses. At the end of these raids, 448 illegal drinking dens had been destroyed, 542 illegal bars closed, 15 million liters of harmful spirits seized, and 200,000 suspects arrested-according to NACADA (Chee & Presidential Strategic Communications Unit, 2015). Chiefs were also casualties of the crackdown and 99 of them were fired and 15 senior police officers also sacked over their failure to curb the spread of illicit brews in the country (Ombati, 2009). The government again had responded. During the implementation of this directive, the public got involved in destroying the "killer brews" so much so that they would destroy legitimate, retail businesses. Property worth millions of shillings was reportedly lost during the raids (Nyawira, 2015b).
In the wake of negative and unpopular reports, policy makers can opt to do nothing, hoping the coverage will quickly end or they can react to minimize the anticipated effects on the general public. From the data, it is clear to see that Kenyan print media has an effect on Kenyan policy makers in the current government especially with the use of negative frames and increased prominence of an issue. The intensity of media coverage had an impact in President Kenyatta's government more than it did for President Kibaki where the media only influenced the policy agenda, considering that the issues, frames covered by the media, and the approaches used in coverage in both administrations were the same.

Conclusion
Examination of media attention to test potential influence on public policy reveals that intense, congruent, and incidentdriven media coverage has had various impacts on public policy.
This study indicates that media attention does lead to influence in some cases. In this illicit brews case, media coverage spurred on legislation and influenced government action by way of the Alcoholic Drinks Control Act (2010), and the sacking of senior government officials and several targeted crackdowns on alcohol manufacturing factories leading to numerous arrests. The success of the alcohol regulation coverage was driven largely by vivid depictions of the plight of Kenyans and a consistent use of negative frames across media houses.
The content of the coverage as well was also rich, demonstrating different viewpoints from affected parties, the general public, experts, politicians, policy makers, and even representatives from diverse government agencies. This contributed to providing policy makers with a comprehensive view of the issue from perspectives they otherwise would not have considered, thus influencing their course of action (Wolfe, 2012). For instance, when a minister in a parliamentary debate quoted a newspaper article claiming there are no more men for young women to marry due to alcohol dependence. The minister used this frame to underline his argument that the alcohol scourge in the country would affect future generations of Kenyans.
Another interesting observation from the research was the congruence of media reporting. Nation Media Group and Standard Group almost had similar coverage of the issues. Studies have shown that congruent media coverage evokes strong responses from political actors if various media outlets focus on the same issue, frame it in a similar way, and cover it persistently (Eilders, 2000).
In the use of frames, human interest frames dominated media reports. In relation to recommendations for government action, these often came from third parties and not necessarily the media itself, and in such cases there was often a governance frame that was addressed to the relevant government bodies. This governance frame usually identified the governance systems that needed to take action on the issues.
The findings of this study reinforce the theories in literature on framing and its power to influence policy. A Norwegian study found that strong frames accepted by society and accompanied by extensive media pressure can lead to changes in legislation or influence legislation itself (Ihlen & Thorbjørnsrud, 2014). The findings are also consistent with interviews from British political actors who conceded that legislative debates often are influenced by media attention and may even accelerate the speed of response to an issue (Davis, 2009). This unique study that investigates a case of intense media coverage and the potential influence of media on public policy is the first of its kind to attempt to shed light on the Kenyan media's role in public policy. This study concludes that there is media influence on policy. The findings of this research illustrate that policy makers in Kenya are not immune to media attention especially in cases where there is intense and congruent coverage that is incident-driven and legitimate.
The Alcoholic Drinks Control Act (2010) was a result of intensive, consistent, and congruent media coverage.
This research also demonstrates weaknesses in media reporting in Kenya. The analysis shows that the media was very good at telling the politicians what is happening and what to talk about, but they did not seem to have the capacity to tell the politicians what to do about the issues. The findings of this research can be used by communications and media training institutions to sharpen the analysis skills of media professionals as well as to reemphasize the watchdog role of the media that requires deep analysis into complex social issues and policy environments. This conclusion resonates with agenda-setting literature that explains that media may focus on reporting the news and not analyzing it (McCombs, 2005;Weaver, 2007). It appears apparent that there needs to be strong policy implementation mechanisms in place for policy action to hold. Further research could look into how media influences policy implementation because policies can be introduced and largely ignored by policy actors.
In terms of a robust democratic environment, it is reassuring to find out that in Kenya, media coverage does elicit policy makers' attention and even forces them to discuss reported issues in their debates as well as make attempts to enforce laws that protect the public.
Further research in this area could look into how this relationship between media and public policy plays out in the digital media and especially in social media platforms such as twitter. It would be revealing to see the relationships between the public, journalists, and policy makers on economic and social issues and whether the findings of this study are similar or replicable in the virtual world.

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 funding for this research was generously provided by the Business Advocacy Fund.