Caught in the Middle: Internal and External Pressures on the Coverage of Organized Crime in Mexico

With 33 journalists killed since the beginning of Andrés Manuel López Obrador’s presidential term in December 2018, Mexico heads the list of the most violent countries for journalists in Latin America—and that of countries not at war. While journalist organizations demand a meaningful protection apparatus to safeguard their physical safety, official corruption, and criminal impunity continue to escalate the pressures to which media staff are exposed, especially in Mexican states where cartels and criminal groups have the largest footprint. This study aims to precisely identify the pressures, both internal and external, facing journalists who report on organized crime in Mexico. To do this, we drew from the Hierarchy of Influences Model, and interviewed 22 Mexican journalists who work on the organized crime beat in the country’s capital and in the most violent states in the Republic. The results suggest that the most influential forces they face are associated with the organizational level (such as editorial line or institutional censorship), and the extramedia level (e.g., anti-press violence from cartels/authorities, and government advertising contracts).


Introduction
Mexico has become one of the deadliest countries in the world for journalists. More than 100 news personnel have been killed during this century, and the number of victims is constantly increasing (CPJ, 2022). However, assassinations are just the tip of the iceberg when it comes to this complex phenomenon, which includes different types of aggressions such as kidnappings, beatings, harassment, and smear campaigns, to mention just a few (González & Rodelo, 2020). In addition, those who commit these crimes enjoy almost complete impunity; the vast majority remain unsolved, while some were never even properly investigated (De León Vázquez & González, 2020). Finally, violence against members of the press is a multi-faceted phenomenon that involves a range of actors and has diverse impacts on reporters and the newsrooms they work for (González, 2020).
Although the potential risk of being assaulted is generalizable across the country, members of the press who consistently cover organized crime activities tend to be more vulnerable (CPJ, 2022;Hughes & Márquez, 2018). Therefore, drawing on the Hierarchy of Influences Model (Shoemaker & Reese, 1996, this paper is guided by the following research questions: RQ1: What are the internal and external pressures at the micro and macro levels that determine the coverage of organized crime in Mexico? RQ2: Which level is most influential when writing a news story on this issue? In the context of this study, internal pressures are related to newsroom practices, whereas external ones refer to outside actors (cartels, government authorities, and even social groups) that try to control news content. In doing so, the study draws on a set of 22 in-depth interviews with reporters and editors from newspapers located in Mexico City and the 10 Mexican states where at least one journalist was killed during 2017.
The study is organized as follows: The first section offers an overview of endemic anti-press violence in Mexico, followed by the study's analytical framework, which draws on Shoemaker & Reese's (1996 Hierarchy of Influences Model. Then, the second section presents the methodological design, while the third includes the study's findings; the article closes with some concluding remarks.

Literature Review: Violence Against Mexican Journalists
Mexico experiences such high rates of violence that, in 2018, its economic impact was estimated at 5.16 trillion pesos (US$268 billion), equivalent to 24% of Mexico's GDP (Institute for Economics & Peace, 2019). Since the beginning of former President Felipe Calderón's militarized response to Mexico's drug war, during his 2006 to 2012 administration, violence in general has been on the rise (Valdés, 2015), including violence against journalists. According to the Committee to Protect Journalists (CPJ, 2022), 124 news workers have been killed in Mexico since 2006; 33 of them occurred during the first 3 years of Andrés Manuel López Obrador's 6-year term, which began in December 2018.
For Hughes and Márquez (2018), in addition to the drug war and the government's response to it, there are other explanations for the violence to which Mexican journalists are subject. Among them, it is especially important to mention the discretionary power that state governors assumed after the 2000 election, when 71 years of authoritarian presidential power came to an end (Espino, 2016;González & Echeverría, 2018;Guerrero & Márquez, 2014;Salazar, 2018). As subnational politics and criminal violence began to merge, journalists who covered either were targeted.
Corruption often puts journalists in a defenceless position since they fear reporting threats to the authorities when those same authorities are suspected of colluding with the criminal organization(s) in question (De León Vázquez & González, 2020). According to the independent organization Article 19 (2019), this invites further aggression. Furthermore, it exemplifies the various effects of impunity and the clear absence of rule of law in the exercise of freedom of expression in Mexico (Article 19 y CENCOS, 2009;De León Vázquez & González, 2020). Brambila (2017) goes further, arguing that violence against the press and subsequent impunity reflect a serious problem in how both state and political institutions function.
It is precisely the journalists in charge of investigating the links between criminals and power players who receive the most threats. Many times, they come from the government itself or from the police instead of directly from criminal groups (Del Palacio, 2018;González, 2018;Waisbord, 2002). Piccato (2003) explains this by remarking several features of crime in 20th-century Mexico City. Some of them, like the corruption of the authorities-whom the population perceives as a source of insecurity and unfair harassment rather than of protection (p. 21) -continue to this day and reveal the difficulty of being a journalist dedicated to covering organized crime in this democratic country.
The violence exerted by organized crime is not the only pressure to which journalists are subject. Although Mexico is a democratic country, its government exerts pressure by way of advertising contracts and agreements with the mainstream media. De León Vázquez et al. (2018) argue that media entrepreneurs prefer to yield to these pressures instead of losing government advertising contracts, and the favor of politicians. In this sense, Espino (2016) concludes that, most commonly, state governors control local powers and, consequently, dominate the public space, as well as local media.
Along the same lines, González (2018) points out that the use of official advertising and "chayotes" are not recent phenomena. The word "chayote" is the Mexican equivalent of the word "embute" in other Spanish speaking countries (Scherer, 2015). Both are euphemisms for the money journalists receive as bribes, which is how power corrupts media professionals (Riva Palacio, 2013). Based on this use, "chayote" is the nickname given to the General Law of Social Communication that has been in force since January 1, 2019. Among other things, it allows for discretionary allocation of government advertising to the media. Many journalists identify this law as another form of pressure exerted on the media. That is, they lose advertising revenue if they do not publish the messages expected by those who govern (Gallegos, 2018;González, 2018;Palacios, 2018).
Since government advertising is the main source of income for Mexican communication companies (Espino, 2016;González, 2018;Salazar, 2018), the media-especially local branches-receive equal pressures from organized crime and the government. Their content often follows an official editorial line since state governments can co-opt the agenda in exchange for economic resources. For her part, Márquez-Ramírez and Hughes (2016) indicates that up to three out of every 10 journalists in Mexico believe that state actors, as well as the military and security forces, government officials, politicians, and lobbyists are highly influential or extremely influential.
Beyond advertising agreements, "chayotes" or bribes are part of the economic aggressions to which Mexican journalists are commonly subject. When a journalist refuses bribes, politicians and public officials use other types of psychological pressure or aggression, including exclusion, smear campaigns, and defamation (both online and offline) (González & Rodelo, 2020).
Under these circumstances, recent studies on anti-press violence in Mexico have tried to explain the impact of the endemic risk associated with journalism. Brambila (2017) argued that journalists are more likely to be murdered in regions with high levels of social violence, internal conflict, low democratic and economic development, and human rights violations. Besides these social determinants of violence against news professionals, threats, economic pressure (Cepeda, 2018;Reyna, 2017), isolation and dangerous beats (such as crime news, organized crime, and government corruption) are predictors of censorship, both individual and organizational (Hughes & Márquez, 2018).
In addition, the practice of watchdog journalism increases the chances of receiving a threat (Salazar, 2019). However, the effectiveness of any given threat depends on the existence of "strategic allies"-for example, NGOs, strong opposition parties, and journalist networks-who support and accompany reporters in their efforts to demand accountability and expose wrongdoings (Salazar, 2019). Thus, constant assaults on the Mexican press have not just constrained reporters and newsroom operations; instead, in response, some have improved their professional practice at the individual and organizational level (González, 2020).
All these studies coincide on at least two issues: First, although extreme anti-press violence tends to concentrate in certain states (Tamaulipas, Veracruz, and Guerrero), safe haven states as such do not exist because media across the country face a wide array of threats and aggressions. Second, there is a growing emphasis on the relevance of journalist networks, both formal and informal, to promote collaborative work, foster professionalization through constant training, and provide aid to the victims of threats and aggressions.

Analytical Framework: The Hierarchy of Influences Model
The Sociology of Journalism approach has two conceptual pillars: (1) News is a constructed reality and, hence, (2) the outcome of a process (e.g., McNair, 1998;Schudson, 1989). In that sense, Shoemaker & Reese's (1996 Hierarchy of Influences Model contributes to the understanding of journalism because it proposes an explanation of the news-making process. The authors suggest a multi-level approach to media content production. In other words, there is a set of micro and macro factors that determine the way an event or issue is reported. Whereas the micro factors are associated with individuals and their involvement in the production of news, the macro factors are related to economic, political, and cultural systems. The model includes the following levels: Individual Level: News workers' demographics (age, gender, ethnicity, etc.) and psychographics (religion and political preferences) shape their perception of the events they cover and, thus, the way they communicate them to their audiences. In addition, journalists' own professional orientations-for example, neutral observer or watchdog-also influence their work. Media Routines Level: The news-making process is structured by a set of practices that organize the production of content. In other words, journalists within a media system share-and are expected to follow-common standards related to gathering, processing, and distributing information.
Organizational Level: Individual journalists and their routines are incorporated into a larger institution with its own rules and goals. That is, newsroom staff are part of a more complex organization, which includes other areas such as administrative and legal departments. Therefore, every single department's objectives are subordinated to the overall organizational goals. For instance, commercial media's ultimate aim is to generate revenue, hence, the content they produce seeks to appeal to wide audiences and, as a result, advertisers. Extramedia Level: This includes all the external actors that-to different degrees-shape content production. For example, individual information sources only provide the angle that corresponds to their interests. Thus, relying on a limited set of sources fosters biased coverage. The government also has a role in the news-making process, not only as a source, but also through the diverse laws and regulations that determine media operation. Finally, media content responds to advertisers and audiences because revenue originates with them. That is, ratings and readership sustain advertising investments. Ideological Level: Any given media system operates within a cultural and ideological system. For that reason, news organizations and their journalists are producers of meanings. In other words, the messages they produce are ideologically aligned with the interests of power holders (media owners, businesspeople, government, church, etc.); these powerful groups provide the framework for interpreting reality. Through a wide array of messages, the media constantly reproduce and legitimate the dominant class's discourse. For instance, it is highly unlikely that commercial news outlets would give prominent space to anti-capitalist or anti-free market voices.
As its name suggests, the Hierarchy of Influences Model indicates a hierarchical relation between the levels with topdown influence, that is, the micro levels are subordinated to the macro ones. Therefore, in the news-making process, the Ideological Level is significantly more prominent than the Individual Level. For that reason, the production of journalistic messages is based upon a series of decisions that go from situational (event coverage) to institutional (organizational gatekeeping), and from there to systemic (ideological and/or cultural alignment) (Shoemaker & Reese, 1996. However, rather than analyzing content as such, this approach studies the forces that shape the messages produced by journalists (Reese, 2007). That is, it focuses on the actors and factors that determine and intervene in the production of news.
As one of its authors (Reese, 2007) said it should be, this model has been widely used in diverse studies worldwide (e.g., Fahmy & Johnson, 2012;Kwanda & Lin, 2020;Milojević & Krstić, 2018;Xu & Jin, 2017). Regarding the Mexican case, Relly and González (2014) analyzed antipress violence in the Northern region, adding a new level to Shoemaker andReese's (1996, 2014) scheme with Intermedia Influences related to journalist networks. As these inquiries point out, this theoretical framework is suitable for explaining journalistic practice around the globe. Nonetheless, context matters when assessing each level's relevance. That is, the specific characteristics of each media system and/or the particularities of the case study determine each level's impact. Thus, whereas macro level influences are sometimes the most important forces (see, for instance, Milojević & Krstić, 2018), in other environments, micro level forces are the main determinants (Relly & González, 2014;Xu & Jin, 2017).
As mentioned in the previous section, journalist networks have emerged as a topic of interest in Mexican journalism studies, particularly in a violent, anti-press environment (see for instance De León Vázquez, 2018; González, 2020;Salazar, 2019;Hughes & Márquez, 2018;). Nevertheless, during the interviews, our informants did not raise the issue. Therefore, we decided to exclude the Intermedia Influences level proposed by Relly & González (2014), and maintain the original Hierarchy of Influences Model instead.

Materials and Methods
The study is non-experimental, with a descriptive scope and a qualitative approach. In-depth interviews were chosen as the research method since they aid in uncovering latent phenomena (Taylor et al., 2015). In total, we interviewed 22 journalists dedicated to organized crime coverage in Mexico's principal newspapers. To contact the journalists in the sample, we used a snowball sampling method. We first interviewed seven journalists located in Mexico City from the main print media there, including the newspapers El Universal, Reforma, La Jornada, Excelsior, Milenio, and the newsmagazine Proceso. This initial sample was very knowledgeable of organized crime and/or drug trafficking news coverage. This first contact was necessary based not only on the experience of those interviewed, but also on the fact that, because they work for the principal Mexican media outlets, they were able to help us confirm our initial list of possible interview candidates throughout the country. Fifteeen interviews were then conducted in Mexico City and in 10 states, all of which had at least one journalist murder in 2017, including Guerrero, Veracruz, Chihuahua, Baja California Sur, Morelos, Sinaloa, Jalisco, Michoacán, Baja California, and Tamaulipas.
As seen in Table 1, the identity of the interviewees was anonymized in order to protect them. The sample stands out because of their experience-the average interviewee with 23 years in the field. At the same time, 40% (n = 9) of the journalists were editors/supervisors at the time of the interview, while the rest 60% (n = 13) were reporters; 54% (n = 12) of the professionals were working in national media outlets, and 46% (n = 10) in subnational media outlets, that is, for regional or local platforms. The final list of interviewees is as follows (in alphabetical order according to their initials): All the interviews were carried out at the informants' workplaces, except in the cases of the journalists from Chihuahua, Tamaulipas, and Baja California Sur. At the time of their interviews, these professionals were displaced in Mexico City as part of an initiative to protect their physical safety after having received threats. The interviews were carried out between the months of February and November 2018; they were all face-to-face and lasted between 1 and 4 hours. The pre-established questionnaire had 10 blocks of questions and all 22 interviews were carried out with the same structured script (Valles, 2007). The analysis process consisted of the following steps. We first transcribed all the interviews. Second, we used an open coding process in order to identify consensus and dissention within the answers (Taylor et al., 2015). This was performed within a matrix in which answers were carefully grouped according to key words or detected indicators. In third place, the indicators revealed categories, which were connected to concepts from the literature, specifically to the Hierarchy of Influences Model (Table 2). Finally, we identified and discussed the results in accordance with the interpretation scheme indicated.

Results
Following the logic of Shoemaker andReese's (1996, 2014) Hierarchy of Influences Model explained above, the results obtained from the interviews are presented below. Therein, we locate the internal pressures at the micro level and the external pressures at the macro level.

Micro Level: The Internal Pressures That Informants Face
Unlike other types of coverage, organized crime tends to be handled differently in Mexican newsrooms due to the pressures that related information professionals face in Mexico Espino, 2016;González, 2018;Merchant, 2018). The Hierarchy of Influences Model (Shoemaker & Reesse, 1996 proposes the individual as the first level of influence. In this sense, journalists must deal with dangerous scenarios, as well as with habitual confusion surrounding the limits between personal and professional life. This is why job turnover in this area tends to be high: "I have fired three of my employees, four, because they can't handle the death, the bodies, the fear of being close [to a crime scene] and of having them [the assailants] return to finish off those left alive. They don't last long" (12).
As for journalistic routines-the second level within the micro sphere-when there are pressures, they are related to bosses prohibiting coverage of a certain topic if it is likely to put reporters' lives in danger. Likewise, another aspect related to routines lies in the increased quality and journalistic rigor that many bosses demand of news reports related to organized crime. That is to say, this pressure is related to greater diversification and verification of sources, and in general taking greater care with regard to content that may end up causing public alarm or lead to the loss of public or private advertisers.
At the organizational level, demand for a high degree of accuracy, which can even lead to the exclusion of certain content if doubts arise about its veracity, is due to the complexity of the coverage, including the established relationship between these topics and Mexican institutions like the police, political representatives, and businessmen: ...issues that are silenced involve, let's say, a big investigation that might call the authorities into question even more so than a cartel. Because... media managers depend so much on advertising, they depend more on, or fear more, threats from authorities than from a cartel. We could say, because they collude (16).
Other internal pressures associated with the organizational level-such as the outlet's editorial line-do not disproportionately affect these journalists. Within Mexican newsrooms, there are other internal pressure mechanisms, such as infiltrated colleagues; most of the journalists consulted claimed to have worked with or met peers whose interest in the sector goes beyond merely journalistic pursuits. However, infiltrators usually work as freelancers or reporters, or in other professional spheres, rather than in established newsrooms. As the following informant comments: On the other hand, journalists' salaries are a source of internal pressure associated with the organizational level since their allocation depends on the medium's salary policy. Given this reality, cartels often offer enticing amounts of money to restrict journalists' professional freedom. In that sense, almost all the professionals consulted agreed that their salaries are not higher than those of other colleagues. Source. Authors' elaboration based on sample. *Though the publication is national, they report as correspondents from their respective states.
Despite the fact that journalists who work on crime and related issues face numerous risks, many media often take advantage of and pay freelancers for the entire story without considering how complex it is in terms of follow-through or the quantity and quality of its resources (11).
For their part, other interviewees argue that their salary is unfair because they have to invest part of it in paying for costs associated with the coverage. Given this job insecurity, the journalists interviewed admit that they come up with other alternatives to financially support themselves. The most common is moonlighting, that is, selling resources or informative content to various media. But we also found others, like selling extra services, including translations and style corrections.
In some of the responses, journalists acknowledged receiving special support from the news outlet they work for, be it salary supplements or security bonuses, per diem for travel, a better vehicle, insurance with coverage greater than that of other journalists, more paid vacation, etc. But it must be reiterated that this kind of support depends more on each organization's distinct policy than on global consensus. Likewise, the journalists who mentioned having these benefits worked at the largest media organizations, which have greater resources and deeper institutionalization.

Macro Levels: The External Pressures That Journalists Face
Within the macro level, we found the highest density of responses in the area of extramedia influence. In analyzing the interviewed journalists' discourse, we identified the primary extramedia pressures that they perceive in their daily practice. It should be noted that external actors intervene depending on numerous factors, such as the organization's size or ownership, financing mechanisms, their degree of approximation to the different actors through economic investment or otherwise: (a) Infiltrators work in sectors beyond the media. Some of the journalists consulted assured us that some officials exercise dual functions in institutions like city councils as press officers for a criminal group: "... the government did not want to admit that [name omitted] was the Zetas' press chief in Acayucan, and also the mayor's press chief" (15). (b) When configuring the positive and negative elements associated with criminals' interests, especially in their work as gatekeepers, infiltrators are in charge of strategically projecting-directly or indirectly-the reputation of the group from which they receive a salary or chayote. For their part, criminal groups offer firsthand information to certain infiltrators who are the first to arrive on the scene. Thus, a symbiosis is established around journalism, which necessarily ends up benefiting, in symbolic terms, criminal groups, by helping to silence bothersome issues; it also benefits journalists in economic terms, when they receive a greater number of scoops and double, under-the-table payment. Additionally, infiltrators are in charge of monitoring their colleagues' work (13). Journalists paid by organized crime groups not only influence general disclosure, but also get involved in a kind of espionage with the aim of giving first-hand information to the cartels whose interests they defend. In some cases, the interviewees acknowledged that the media knows about this reality, although they prefer to ignore it to avoid conflict with cartels. In addition to working for a cartel, some infiltrated journalists are part of Mexican state forces. As one interviewee recognized, undercover police officers sometimes double as journalists to safeguard professionals responsible for covering organized crime: Yes, here the Centre for Investigation and National Security sends journalists-some trained by them and some by our colleagues. They serve as informants, pass on photos, and so on. They are concerned with the risks we take; they know where we are going and everything, you understand (19).
Even so, one-fourth of the journalists interviewed stated no knowledge of infiltrators in their newsrooms: "No, I don't think so. There may be reporters. . . But inside the newsroom, no" (14).
(c) Pressures that come from commercial advertisers. Like practically all media that receive financing through commercial investment, more than half of

Macro level
Ideological: Cultural proximity between reporters and members of organized crime, particularly in certain regions or small cities Extramedia: Antipress violence from cartels and government authorities; advertisers do not invest in crime news sections; government authorities exert pressure through advertising contracts Micro level Organizational: Editorial line; organizational censorship related to the coverage of certain issues; precarious labor conditions; newsrooms infiltrated by cartel spokespersons; certain news outlets give bonuses to investigative reporters Media routines: Factual reporting, fact-checking, diversification of sources, accuracy Individual: Journalists believe that they have to keep society informed based on their professional values Source. Author's elaboration based on the hierarchy of influences model (Shoemaker & Reese, 1996. the journalists interviewed acknowledged that commercial advertisers exert external pressure on the coverage of organized crime. This pressure takes place, firstly, because some companies do not want their brand associated with these issues. Furthermore, advertisers may withdraw their funding when they do not like the coverage of certain issues, as several of those consulted acknowledged: "Yes. There are companies that say they won't advertise in our [medium] because of the coverage of drug trafficking" (1). In fact, a few of the interviewees indicated that some media have created secondary headlines detached from police news to avoid the economic downfall associated with advertisement desertion. For those that keep police news in local and regional media, the gamut of advertisers varies and can include unconventional companies, such as bars of dubious reputation, or brothels, among others. We observe that some of the interviewees, influenced by advertisers, have moderation strategies that help them prepare their content. For example, in the case of articles with photographs, they try not to show gory details. Another strategy involves omitting the scene of the crime or event, given that they know that advertisers tend to request such an omission when the scene in question is linked to their business. Pressure is thus exerted through written and unwritten advertising investment agreements. In these situations, publishers tend to filter references and anonymize scenarios to avoid advertisers' displeasure.
The media can also join in as accomplices, when managers receive a bribe or economic boost through advertising. In this sense, information omission or bias is typical of some media.
(d) Pressure from governments. More than half of the journalists interviewed agreed that Mexican governments (national, state, local) are responsible for external pressure because they are the largest advertisers in the country. In some cases, government advertising exerts pressure to get journalists to promote official sources to the detriment of other sources. In other words, as recipients of financial investment, some media choose to relax key professional values, such as comparing/contrasting and verification. The pressure exerted by state entities is strongest among the regional and local media, which are also the most dependent on advertising money: Well, I can tell you that the media here in Michoacán survive because of state or municipal agreements. . . I've heard that the municipal government has exerted pressure to fire reporters that are uncomfortable with the agreement, whether the state one or the municipal one (11).
Media with higher subscription numbers are less vulnerable to government influence because they have an additional source of income.
(e) Pressure from the cartels. Almost all the journalists interviewed confirmed that the cartels put pressure on what they investigate and publish; more than half mentioned having received threats, especially death threats: "Well, yes. It happened to me twice; twice they ordered someone to kill me because we published about small planes, about an airfield, because I had access to military intelligence" (13). In interviewees' responses, we found that they find out about these threats through direct contact with criminal groups, usually from someone with a press connection, through videos generated by criminal gangs directed at journalists or their media group, or through reports from the United States Intelligence Community, which is more sophisticated than the Mexican counterpart: U.S. authorities have greater investigative capacity in terms of infiltrating these groups or tracking their phones and listening to their conversations. The threats that I have learned of have been thanks to that system (1).
This external pressure usually occurs after a given news story has already been published. Only in very few cases we observed pre-publication pressure, that is, pressure that begins when a cartel learns that a certain journalist is putting together a story on a given topic. In this case, and to prevent its publication, cartels offer money or direct various types of threats at journalists, which include and go beyond physical threats, including those against their reputation, their family or friends. According to the interviewees, criminal gangs particularly issue threats when they feel offended because a piece alludes to their leaders' private matters, or when a piece presents what in other areas could be considered libel. Threats are also prevalent when details are published about a cartel's activities or its political or institutional relationships, issues that can alert public opinion to complex links between criminal elements and the Mexican government.
Similarly, cartels demand coverage of content that they consider relevant and beneficial to their image, and so they also threaten journalists who refuse to cover them. In this regard, responses from the journalists interviewed reveal an interesting balance. On the one hand, cartel members are sometimes used as key informants to help generate exclusive news, be it about politicians, police officers, or warnings about criminal actions, such as the murder of a rival, for example. But, on the other hand, using these key informants can turn into a dangerous alliance both because of the association that police forces can establish (the journalist or news outlet may be considered to be colluding), and because of cartels' constant demands for coverage of topics that benefit their reputation.
Likewise, some journalists underlined the existence of socalled "narco-journalism," which alludes to informants who, paid by criminal gangs, are responsible, for example, for omitting any fact or description that may harm said gangs: ...I think that at that time the Sinaloa Cartel had not yet been divided and they came to a reporter of mine and that person came to me and said, "Look, so-and-so told me that he was giving me four thousand pesos not to publish.' So, we started investigating and he was an "engager." At one time, dozens of reporters were on the narco payroll. I can tell you about 30 reporters and photographers who were, and he was just one of the "engagers," but there were 2 or 3 more (8).
"Narco-journalists"-as some of the interviewees referred to them-are more common in regions with greater cartel influence. Their presence is more frequent in local media, which tend to offer inferior salary conditions to their employees. At the same time, the interviewed journalists claimed that professional training is inadequate in these smaller news outlets. That is, the line between professional ethics and individual interest is often blurred. On the contrary, journalists who work for national press outlets less frequently mentioned the existence of infiltrators in their newsrooms.
To avoid pressures from criminal gangs, journalists covering these issues often employ an explicit reference to the sources behind the news: Many times, well, maybe you are not going to tell a drug trafficker that they have to read carefully, but if you say that you got something from a source, that will take the load off you: "It's not me; it [the information] is official, it is from a government institution." It appears in these places (7).
Approximately a quarter of the interviewees stated that they had not received direct threats from the cartels, but they do know that sometimes happen to other colleagues. In this sense, Mexican government administrations can incentivize these stories, turning them into an agenda item, or they can discourage them by omitting them from the public agenda. They mostly exert pressure through government advertising, which is the main source of funding for much of the media, or through the work of official spokespersons or communication directors. Another major pressure has to do with content's impact. From this perspective, through social networks, journalists are often very aware of how public opinion receives the information they disclose. For example, metrics associated with number of likes, retweets, or times shared have a direct influence since journalists learn to build impact through the interest their content stirs up. Still, online participation is openly seen as a threat to some publications, such as magazines that are financed by the sale of subscriptions or copies, because decreased income tends to make journalistic practice more precarious. Broadly speaking, like in the case of internal pressures, a quarter of the interviewees affirmed that they do not perceive any type of pressure, whether from the cartels, the government, or advertisers. This group is mostly made up of journalists from nationally oriented media. Finally, like the individual level in the micro sphere, the informants mentioned the ideological level the least. In this sense, there may be great complicity between some journalists and cartel members, while in certain enclaves the bonds forged by friendship and similar shared cultural and ideological values predominate. This is especially common in smaller towns. In this way, external pressure is also exerted from a cultural point of view, that is, through assimilating environmental practices and challenges.

Conclusions
As stated previously, and following the Hierarchy of Influences Model (Shoemaker & Reese, 1996, we found that the internal pressures that publishers and editorsin-chief exert on the coverage of organized crime is different with respect to other topics. On the one hand, managers, in this sense, are aware that journalists may be subject to serious threats and violence (see also González & Rodelo, 2020). But, on the other, they also highlight the impact that these issues can have given the relationship between organized crime and other spheres of Mexican political and business life, which, as Márquez-Ramírez & Hughes (2016) points out, can certainly influence journalists.
Therefore, reporting on this beat demands high levels of accuracy, which both reinforces journalists' safety and prevents accusations of libel. Consequently, the main difference between covering organized crime and other beats is found in presenting a more precise and factual account of the events (see De León Vázquez et al., 2018;González, 2020;. Internal pressures, then, are mainly related to higher professional standards. Likewise, the evidence suggests other aspects that may represent a serious threat to Mexican journalism, such as the presence of infiltrators. In this way, our informants acknowledged that some reporters work for both a news outlet and a criminal group. As also discussed by previous studies (e.g. González, 2018;González & Cepeda, 2021;Merchant, 2018), this is particularly evident in the regions that are most affected by violence, as well as in media organizations that offer their staff poor working conditions.
Interestingly, the majority of the interviewees recognized that salaries do not differ according to the beats they cover. As a result, generalized low wages may lead to low quality content (see also Cepeda, 2018, Del Palacio, 2018González & Cepeda, 2021;Reyna, 2017). This is because many journalists have to work for several outlets, or even engage in other activities, including receipt of the extra income that cartels offer. Even though precarious labor conditions exist nationwide, freelancers and journalists who work for small local news organizations far from major metropolitan areas are the most vulnerable.
Regarding external pressures, cartels stand out as the main source because a compromising or poorly investigated story can be its author's death sentence. In order to confront this situation, participants in this study often adopt diverse protection strategies that aim to minimize risks, whether personal or organizational. The most common of these measures is the publication of content that has been painstakingly compared and contrasted with diverse and identified sources. In doing so, reporters cannot be blamed for the content because they just collected and included different sides of the story; which is also consistent with the literature revised before (for instance De León Vázquez et al., 2018;González, 2020;. Besides cartels, other actors can present threats, including advertisers, both commercial and governmental (see also De León Vázquez et al., 2018;Espino, 2016;González, 2018). The former exerts indirect pressures when they demand not to be associated with crime news, which includes making sure their advertisements do not appear alongside it. Some may even act as gatekeepers when they threaten to withdraw their ads if the news outlet publishes that kind of information.
Because government advertising is the largest source of income for the Mexican media, its advertising power represents the most significant external pressure, at least in the analyzed answers. Thus, governments (federal, state, and local) stand out in terms of direct and indirect pressure on journalists, given their role in advertising income (González, 2018;Salazar, 2018).
In short, the findings indicate that Mexican journalists who cover organized crime face pressures, both internal and external, that determine their job because the content of any news story related to this sensitive issue results from a constant struggle against diverse interests. The empirical evidence herein suggests that the most influential forces are associated with the organizational level (such as editorial line, institutional censorship, precarious labor conditions, and newsrooms infiltrated by cartel spokespersons) and the extramedia level (e.g., anti-press violence from cartels/ authorities, commercial advertising loss, and government advertising contracts).
Like any other academic endeavor, this inquiry has its limitations. First, this is a qualitative study with a reduced sample size and, although our findings are consistent with previous research on anti-press violence in Mexico, we cannot conclude that our findings represent Mexican journalism as a whole. Nonetheless, the experiences and insights that our interviewees shared provide a clear image of the conditions that their peers across the country are familiar with. Second, we analyzed the phenomenon from the news workers' perspective and, hence, we draw on the Hierarchy of Influences Model (Shoemaker & Reese, 1996. As previously noted, this approach does not focus on the content of news stories as such, but on the factors and actors that determine its production. That is, the responses collected herein make up an account of the internal and external conditions that the participants constantly face. Third, using in-depth interviews as a methodological tool involves a potential mismatch between what respondents say they do, and what they actually do. Despite these limitations, the findings have a twofold contribution to the growing literature on Mexican journalism. On the one hand, interviewees' answers provide empirical evidence that reinforces the idea of the uneven nature of the Mexican media system (González & Echeverría, 2018;Guerrero & Márquez, 2014;Salazar, 2018). This approach argues that there are significant differences between the media located in the country's capital-and other large urban centers such as Guadalajara, Tijuana, and Monterrey-and those that operate in less developed areas (González & Echeverría, 2018;Salazar, 2018). In other words, journalistic practice in Mexico is not homogeneous and is impacted by differences linked to each region's political, economic, social, and cultural context. Consequently, the national mainstream media, although not completely immune to the aforementioned pressures, do have more resources, both financial and journalistic, to deal with them (González & Echeverría, 2018;Guerrero & Márquez, 2014;Salazar, 2018). In this study, we have emphasized at least four factors that illustrate this point: Distinct professional performance, precarious labor conditions, government advertising contracts, and political alignment.
On the other hand, and closely connected with the previous point, our results also provide evidence to the discussion on subnational authoritarianisms and their impact on journalism (e.g. Salazar, 2019Salazar, , 2020Salazar, 2018). That is, as an outcome of the so-called political transition (when the hegemonic Institutional Revolutionary Party-PRIlost the presidency for the first time in 2000), a fragmentation of political power has occurred as well. However, since the head of the federal government could no longer exert control over the whole country, as their PRI predecessors did because of this party's historical centralization of power, local political strongholds gained salience. That is, whilst the presidential figure weakened, mayors and governors strengthened their positions and therefore, power was concentrated without a system of checks and balances (Del Palacio et al., 2020;González & Echeverría, 2018;Guerrero & Márquez, 2014;Salazar, 2019Salazar, , 2020Salazar, 2018). As we have discussed here, the discretional allocation of government advertising, the collusion between authorities and organized crime, and constant aggressions against the press by both armed forces and gangs exemplify this phenomenon.