Gender mainstreaming (GM) refers to the “(re) organisation, improvement, development and evaluation of policy processes, so that a gender equality perspective is incorporated in all policies at all levels at all stages, by the actors normally involved in policy making” (Council of Europe 1998, 15). Emerging from the 1995 United Nations (UN) Beijing Platform for Action on women's rights, it has been advanced as an important tool for developing policies that advance gender equality and the empowerment of women. Today, more than 160 governments and international/regional institutions have made GM a key component of their policy-making processes. In Canada, for example, federal and provincial governments require that gender considerations be taken into account in all aspects of policy work, including policy formulation, implementation, and evaluation. Yet, despite initial optimism about GM's transformative potential, recent studies show that it has fallen short of its goals (see Caglar Reference Caglar2013 for a brief review).
Problems of policy implementation are often cited as the reasons for GM's disappointing results for gender equality. Most studies have used a macro-level approach to explain how GM is implemented in different countries and to what effect. Conceptualizing policy implementation as a “top-down” process, driven primarily by centrally located policy designers, these studies have tended to focus on the ambiguous wording of GM policy, “soft” or legally nonbinding policy instruments, poor compliance measures, as well as insufficient human, organizational, or financial resources. Some studies (Daly Reference Daly2005; Mazey Reference Mazey2002; Meier and Celis Reference Meier and Celis2011; Squires Reference Squires2005), for example, have highlighted the uncertainty surrounding the meaning of “gender” and “equality,” and the difficulties of translating these terms into action. Other scholars (see Subrahmanian Reference Subrahmanian2004) have also pointed to the institutional challenges of mainstreaming gender considerations into government practices that privilege neutrality, technocracy, and rule adherence as opposed to social justice goals. While these studies have advanced our understanding of the macro-level forces that shape the prospects and challenges of implementing GM, what remains understudied is how these forces are understood and expressed at the local level in the day-to-day work of actors tasked with enacting GM policy “on the ground.” Policy scholars remind us that policy meanings and outcomes are not a given; rather, they are constituted in the local contexts in which they are implemented (Yanow Reference Yanow1993). Understanding the success or failure of GM, therefore, requires that we direct our attention to the micro level and the stories articulated by the civil servants who translate GM into their everyday practice.
Drawing on implementation research and narrative analysis, this article explores the micro-level dynamics that significantly shape the character and outcome of GM implementation. Adopting an interpretivist lens (Stone Reference Stone2002; Yanow Reference Yanow1996), we seek to uncover how gender analysts in Canada's federal public service understand and conduct their work in implementing GM through storytelling. As noted by previous work, GM is a policy about policy (McNutt Reference McNutt2010; Paterson Reference Paterson2010). It provides a conceptual guide and practical toolkit for integrating gender considerations into the development of policies and programs. However, while GM, like other policy initiatives, is designed by senior officials, it is enacted by local bureaucratic actors through their interpretation and application (e.g., Ball Reference Ball1993). Moreover, the micro-level dynamics of GM implementation cannot be understood without taking into account broader political and managerial prerogatives that are changing the nature of policy work in general (Colebatch, Hoppe, and Noordegraaf Reference Colebatch, Hoppe, Noordegraaf, Colebatch, Hoppe and Noordegraaf2010). By bringing the experiences of gender analysts to the fore, we can gain a better understanding of how macro- and micro-level forces interact to influence the implementation and outcome of GM.
In this article, we offer a brief overview of GM in the Canadian federal public service. In the following section, we highlight the institutional context and central precepts of GM and the broader forces that are shaping the policy work of gender analysts in the Canadian bureaucracy. We then explain our choice of theoretical framework and methodology. Given the article's focus on the micro-level experiences of gender analysts, narrative analysis is used to understand their individual motives and actions. As Maynard-Moody and Musheno explain, narratives “illustrate the consequences of following, bending or ignoring rules and practices. They bring institutions to life; they provide a glimpse of what it is like to [work there]” (Reference Maynard-Moody and Musheno2003, 30). We then present an analysis of the narratives and highlight how macro-level forces are being negotiated, reproduced, or contested in the micro-level work of analysts. We conclude with a discussion of the findings and some reflections on future research endeavors on GM.
THE RESEARCH CONTEXT: GENDER MAINSTREAMING IN CANADA
Since the mid-1990s, gender mainstreaming has become a key pillar of gender policy architectures around the world. Emerging from a long history of feminist activism and advocacy in the international arena and taking form at the Fourth World Conference on Women in Beijing 1995, GM more generally is based on the (feminist) understanding that policies have the potential to impact men and women differently, thereby creating and sustaining unequal power relations (Bacchi and Eveline Reference Bacchi and Eveline2010; Rankin and Vickers Reference Rankin and Vickers2001). As practice, GM entails integrating a gender equality perspective in the formulation of legislation, policies, and programs to address and eliminate differential effects of government interventions on women and men.
In 1995, Canada signed on to the UN Beijing Platform for Action, which required all member states to “seek to ensure that before policy decisions are taken, an analysis of their impact on women and men, respectively, is carried out” (UN 1995, 86). Shortly thereafter, the Canadian federal government introduced a gender mainstreaming strategy, known as Gender-Based Analysis (GBA), which required gender considerations to be incorporated into all aspects of policy work, including design, analysis, and evaluation (Paterson Reference Paterson2010). The GM strategy would soon displace women's policy machinery and women's movement organizations, which were already in decline by the mid-1990s, as the key vehicles for advancing women's issues within the state (Haussman and Sauer Reference Haussman and Sauer2007). While Status of Women Canada (SWC), the federal-level women's policy agency, plays the lead role in GM development and training, the actual “doing” of GM is left to individual public servants. Moreover, departments are given broad scope to develop GM frameworks that suit organizational needs, while central agencies such as the Privy Council Office and the Treasury Board Secretariat perform a challenge function, informing public servants of their duty to take gender and other factors into consideration “where relevant” (SWC 2009).
Despite the Canadian government's formal commitment to GM, the strategy has produced limited results for women's empowerment and gender equality. Scholars have advanced a number of reasons for its limited impact, including a weak lead agency for GM (Grace Reference Grace1997); a gender-only perspective on diversity that excludes other markers of social and economic stratification such as race, age, sexuality, etc. (Hankivsky Reference Hankivsky2005); and the use of an expert-driven approach that favors technical knowledge and economic efficiency in policy deliberations over community input (Rankin and Vickers Reference Rankin and Vickers2001). In Canada, the focus on the means of gender mainstreaming rather than the ends has limited its potential to challenge bureaucratic and androcentric discourses pervading bureaucratic and social practice in Canada (Paterson Reference Paterson2010).
Issues related to implementation were also cited by the federal government's own assessment of GBA. In 2009, Canada's Office of the Auditor General (OAG) reported a number of problems plaguing GBA in the federal government. Based on its audit of GBA in nine federal departments, the OAG found that the extent to which GBA frameworks were implemented varied considerably across departments, as did the quality of the analysis. The OAG concluded that reporting of gender impacts could be improved with guidance and clearer communication of expectations from the Centre (OAG 2009). Following the report, an action plan was developed to facilitate the implementation of GBA, which included providing more support for SWC to assist in skills development, training, and assessments of departmental frameworks. Despite these initiatives, a recent report (OAG 2016) on GBA reveals an implementation gap continues to persist.Footnote 1
Beyond program specific reasons, broader, macro-level forces have also shaped the nature and outcome of GBA in Canada. Between 2006 and 2015, neoliberal ideology and the politics of austerity dampened political support for GBA. The Conservative government, as part of its austerity agenda, significantly cut funding to groups and programs aimed at improving the status of Canadian women. For example, in 2006, SWC's budget was reduced by 43%, resulting in the closure of 12 of its 16 regional offices (O'Manique Reference O'Manique2015). Moreover, within the neoliberal state, gendered politics and gender equality claims-making were marginalized and even displaced, as agencies were instructed to remove references to “gender equality” from their websites (O'Manique Reference O'Manique2015). In this context, the institutional capacity and political will to implement GBA, key elements of a successful strategy, were sorely lacking.
New public management ideas and culture and their effects on policy work also shape GM implementation. Policy work encompasses a broad range of activities, such as producing policy knowledge, providing policy advice to decision makers, and forging and maintaining relations among relevant stakeholders (Howlett, Wellstead, and Craft Reference Howlett, Wellstead and Craft2017). With the rise of the audit culture (Power Reference Power1994) and its emphasis on performance measurement and quality control, policy work is increasingly about “box ticking,” focusing “attention inwards towards systems, inputs and outputs rather than broader outcomes, turning means into ends” (Taylor Reference Taylor, McLaughlin, Osborne and Ferlie2002, 111). These developments are unfolding in the Canadian federal bureaucracy, as public servants find themselves spending more time meeting performance benchmarks and fulfilling accountability requirements than creating policy expertise.
Moreover, between 2006 and 2015, the policy-making role of public servants was further circumscribed because of strained relations between the civil service and Stephen Harper's Conservative government. In Canada, like other Westminster-style parliamentary systems, civil servants are tasked with providing objective and nonpartisan advice to decision makers based on evidence and expertise. This changed under the Conservative government, which, mistrustful of the bureaucracy, instituted a top-down, partisan approach to policy making that limited the influence of civil servants. It is within this context, shaped by uncertainty and ambiguity, that gender analysts are “making sense” of GM. This study reveals how these broader, macro-level forces in Canada are being played out at the micro level, in the everyday work and experiences of gender analysts in the federal bureaucracy. Our aim is to uncover how macro- and micro-level forces interact to shape the implementation and outcome of gender mainstreaming.
THEORY AND METHODOLOGY: IMPLEMENTATION RESEARCH AND NARRATIVE ANALYSIS
The scholarly literature on policy implementation identifies two main and often competing approaches to studying and explaining policy implementation: top-down versus bottom-up approaches. The top-down approach, which dates back 40 years to the germinal work of Pressman and Wildavsky (Reference Pressman and Wildavsky1973), regards implementation as the carrying out of formal policy decisions (Mazmanian and Sabatier Reference Mazmanian and Sabatier1983). Viewing implementation as a “governing elite phenomenon” (deLeon and deLeon Reference deLeon and deLeon2002, 468), proponents of this approach focus their analytical attention to policy makers' intentions, with lower-level administrators primarily viewed as dealing with “mundane decisions and interactions” (Van Meter and Van Horn Reference Van Meter and Van Horn1975, 445). When policies do not achieve their intended outcomes, “top-down” scholars typically turn to insufficient resources, poor policy design, and weak compliance measures as explanations.
By the 1980s, the top-down approach was criticized for ignoring variables embedded in the local context of policy implementation, such as working conditions and the attitudes and behavior of actors tasked with translating policies into action. Rejecting the conceptualization of policy implementation as a purely administrative process, the bottom-up approach views policy making as a more complex and integrated process. As explained by Lipsky, “policy making does not simply end once a policy is set out” (Reference Lipsky1980, x); rather, the content and outcomes of policy are continually shaped and reshaped by lower-level public servants responsible for implementing policy in local settings (Barrett and Fudge Reference Barrett, Fudge, Barrett and Fudge1981; Hudson Reference Hudson and Hill1993). Early scholarship on street-level bureaucracy revealed how frontline bureaucrats, driven by self-interest, interpret policies and allocate resources to protect their working environment (Lipsky Reference Lipsky1980). Building on this research, more recent work by interpretivist scholars (Durose Reference Durose2009; Maynard-Moody and Musheno Reference Maynard-Moody and Musheno2003; Yanow Reference Yanow1996) shows how civil servants are not neutral processors or passive recipients of policy decisions; rather, they are active agents who shape the meaning of policies in their day-to-day activities. By shifting the analytical focus to the micro-level, interpretivist scholars shed light on how local civil servants make sense of policies and their outcomes within the broader context of public sector reforms that are (re)shaping the work of bureaucracies (Ball et al. Reference Ball, Hoskins, Maguire and Braun2011; Bevir and Trentmann Reference Bevir and Trentmann2007).
Within the gender mainstreaming literature, implementation has been identified as a crucial factor in its success (e.g., Meier and Celis Reference Meier and Celis2011). To date, however, much of the scholarship on gender mainstreaming has focused on program design issues (i.e., goals, rules, resources, tools, etc.) and their consequences for gender equality and transformative politics (e.g., Benschop and Verloo Reference Benschop and Verloo2006; Verloo Reference Verloo2001; Meier and Celis Reference Meier and Celis2011). Less is known about the human dimension of GM implementation, namely, the activities and experiences of gender analysts who must translate their understanding of GM into their day-to-day work, a process that is shaped by broader managerial and political forces. This article addresses this gap by exploring the micro-level dynamics of GM implementation in Canada. Using narrative analysis (Ginger Reference Ginger, Yanow and Schwartz-Shea2006; Hajer Reference Hajer, Fischer and Forester1993; Roe Reference Roe1994), we illuminate how gender analysts operate and “make sense” of their work within the context of broader political and administrative forces that often are not kind to feminist interventions (Ferguson Reference Ferguson1984; Findlay Reference Findlay2015; Stivers Reference Stivers2002).
As a mode of policy analysis, narrative analysis has been used in a variety of ways, revealing the ways in which policies are understood and become “actionable” (e.g., Hajer Reference Hajer, Fischer and Forester1993; Ingram, Ingram, and Lejano Reference Ingram, Ingram and Lejano2015; Stone Reference Stone1988), how policy controversies are potentially “solved” (see Yanow Reference Yanow2000 for a discussion; see also Roe Reference Roe1994; Schön and Rein Reference Schön and Rein1994), and how policies are lived and experienced in the everyday (e.g., Neysmith, Bezanson, and O'Connell Reference Neysmith, Bezanson and O'Connell2005; Orsini and Scala Reference Orsini and Scala2006). In all of this work, narratives are conceived as ways of knowing (Clandinin and Connelly Reference Clandinin and Connelly2000, 45). Moreover, narratives enable action and assign roles and responsibilities to various actors (Bevir and Rhodes 2006; Dinham and Lowndes Reference Dinham and Lowndes2008; Ingram, Ingram, and Lejano Reference Ingram, Ingram and Lejano2015; Stone Reference Stone1988). Therefore, they link agent to structure, illuminating the points at which actors adhere to or challenge institutional discourses and practice. Yanow explains, “In their telling … they also become, themselves, sources of meaning, even when their storied nature is neither explicit nor, at times, recognized” (Reference Yanow2000, 58). Therefore, power works through narratives as they serve to represent individuals and issues in ways that reproduce or, alternatively, transform social relations.
Narratives can take several forms, including stories, nonstories, counterstories and metanarratives (Roe Reference Roe1994, 3–4). In this study, we are most concerned with metanarratives, narratives that transverse individual accounts. Our use of metanarratives resembles Hajer's (Reference Hajer, Fischer and Forester1993) concept of “shared storylines,” which provide discursive “short hand” for groups discussing complex and/or ambiguous issues. From this perspective, various actors might put forth seemingly different stories that, upon closer scrutiny, share a discursive affinity with one another. Institutions and organizations shape and are shaped by broad narratives that inform shared meanings and identities therein (e.g., Lowndes and Roberts Reference Lowndes and Roberts2013; Stone Reference Stone1988). They also offer discursive foundations for understanding new and extant practices. Consider, for example, that gender mainstreaming is a contested concept, subject to both feminist and rationalist-instrumentalist interpretations (e.g., Daly Reference Daly2005; Meier and Celis Reference Meier and Celis2011; Verloo Reference Verloo2005; Walby Reference Walby2005; Woodward Reference Woodward, Christiansen and Piattoni2004). As noted by Caglar, GM is a concept that “can mean all things to all people” (Reference Caglar2013, 338). Despite its origins in feminist theory and activism, GM suffers from a lack of conceptual clarity, enabling its coherence with bureaucratic discourse. Whereas feminist interpretations emphasize the substantive dimensions of GM, effectively linking it to social justice, rationalist-instrumentalist interpretations tend to depoliticize gender as a category and lived experience, emphasizing instead the “rules and tools” of policy analysis (Meier and Celis Reference Meier and Celis2011). Thus, the ways in which new and contested initiatives such as GM adhere to or resist these organizational narratives offer insights into not only how GM moves from theory to practice but also the transformative potential of GM. These shared storylines, or metanarratives, are potential mechanisms through which analysts “make sense” of their work, reconciling conflicting interpretations, identifying perceived problems with GM, and illuminating potential solutions to those problems (e.g., Wagenaar Reference Wagenaar1995; see also Bevir Reference Bevir2004; Ginger Reference Ginger, Yanow and Schwartz-Shea2006; Maynard Moody-Musheno Reference Maynard-Moody and Musheno2003). They render GM actionable.
Our narrative analysis focuses on interviews with public servants conducted during 2013 and early 2014. It is important to note that the interviews took place at the height of potentially drastic cuts to the public service, which is likely to have impacted the responses in important ways, as we discuss later. We used a number of strategies to recruit participants for our study. Initial participants were identified and recruited through our personal contacts and by consulting the Government of Canada Employment Directory. Using the snowballing technique, we then recruited additional participants by asking initial participants to provide the names of individuals who might be interested in participating in the study. Finally, at our request, the director of the gender-based analysis unit at SWC sent out a letter of invitation to participate in the study across its intergovernmental network.
Our recruitment efforts resulted in a total of 13 participants. To minimize selection bias, participants were drawn from different line departments and central agencies and worked on different aspects of the GBA process, including development, training, and application. They were also at different stages of their professional careers, ranging from 4 to 40 years of experience in the public service. Each participant was the primary GBA person in their organization, and with few exceptions, GBA made up a large part of their duties. All of the participants had postgraduate degrees, and about one-third had doctoral or law degrees. Most of our respondents (11) self-identified as feminists. Names were changed and identifying references were removed to protect the anonymity and confidentiality of participants.
The semistructured interviews, which lasted between 60 to 85 minutes, were conversational in nature and conducted using an interview guide consisting of questions related to work history and experience as gender analysts. One person completed the interview questionnaire in print because of scheduling issues. We conducted six individual interviews and two small-group interviews that included three participants each. Each interview was audio recorded, transcribed verbatim, and returned to the participant for member checking. In our analysis, we drew on interview transcripts and recordings to piece together both individual and shared narratives among the participants. To reveal shared narratives, we looked for common words, phrases, and/or narrative themes that characterized the participants' experiences. The stories presented in the following section reveal narratives of isolation, disempowerment, and resistance that cut across individual accounts. These metanarratives provide insight into the forces shaping and being shaped by the local context of GM implementation in Canada.
STORIES OF ISOLATION: GM AS EVERYWHERE/NOWHERE
It's like sometimes people didn't know: “Who is the GBA Unit … where are they?” Because ‘they’ were hidden in the weeds. (Alexis)
In setting the scene for GM, the interviews painted a fairly bleak picture of its status in the Canadian public service. Several former and current GM analysts discussed how they had witnessed the federal government's commitment to gender issues and gender-based analysis wane in recent years, as budgetary cuts and departmental reorganizations led to the disappearance of GM units and gender focal points in many departments. The individual narratives revealed a shared storyline of isolation addressing both analysts and the institutional configuration of GBA. Underlying the story of isolation is a story of decline. As noted by Stone (Reference Stone2002, 138), the general arc of stories of decline is that things were once good but have since worsened. Indeed, the diminished status and presence of GM was especially noticeable in large departments that were once at the vanguard of gender analysis. After government cuts and layoffs, analysts who once worked as a member of a GM team found themselves as the sole GM focal point for an entire department.
The gender focal points that remained were relocated away from points of influence within their organization. No longer in policy development units, the remaining gender focal points expressed frustration at their limited access to resources and key policy actors within their departments. Jamie, for example, spoke of not being able to engage in meaningful gender-related work or to influence policy making once relocated to another division after the layoffs.
[A]fter the cuts, for instance, my role as the gender person, I would move from a shop where I could have access to all this data to a branch quite removed from the policy process, not engaged in policy, and it really limited my ability to be able to influence or engage with people, to be able to influence the work that they were doing. (Jamie)
The above quote illustrates that for Jamie, doing GM meant being able to influence others to incorporate gender considerations in their policy work in a meaningful way. However, after the layoffs and relocation were implemented, Jamie found the gender analyst's sphere of influence to be diminished.
Once the gender units were dismantled, “doing GM” would be left to lone individuals within departments. For many of the respondents we spoke with, GM was not part of their official job descriptions. While some of the individuals we interviewed were appointed as gender focal points by their superiors, others volunteered to take on the mantle for GM in their units because of their previous experience and/or their belief in the value of the work for their units and the population they served. Morgan, for example, recounted how she became the de facto gender focal point in her unit after she kept raising questions about the gendered impact of policies and programs in her unit. “I guess I just kept asking those questions, and then my director general said: ‘Look, this is going to be in your file. If we need to have any GBA done, then that's your role,’ basically. So, it wasn't written in my job description … it was just me kind of putting my hand up.” Other gender analysts told more action-oriented stories about taking on gender-based analysis as one of their official duties. Cameron, who positioned herself as a “maverick” within her unit, described her inability to let go of GM in the following way:
[T]o me, it is absolutely essential, especially because of the department that we are. I'm like … if anyone is going to be doing GBA, it's going to be us, right? So for me that's why I'm having a hard time letting go of that piece of meat, even though I was told that it's no longer there. I'm still doing it. I'm still cooking and I'm still hunting.
In this text, Cameron depicts gender-based analysis as an organizational imperative; to abandon it would mean abandoning what she regards as an integral component of the department's values and mission. She uses the hunting metaphor to describe her lone efforts to identify opportunities to build support for GM within her organization.
Many of the civil servants we spoke with attributed the invisibility of GM to the perception that it has been successfully mainstreamed within the federal civil service. As discussed earlier, gender mainstreaming requires all staff members to integrate gender considerations in all aspects of their work. However, one of the challenges of assigning responsibility to all employees for gender mainstreaming is that no one is ultimately held accountable for its success or failure. Moreover, an expectation to incorporate gender considerations does not necessarily translate into action. As Tiessen explains, “The ultimate challenge is to make sure that gender mainstreaming does not become everyone's but no one's responsibility or that gender mainstreaming gives the appearance that it is happening everywhere when it's really happening nowhere” (Reference Tiessen2007, 77).
The “everywhere/nowhere” phenomenon identified by Tiessen was reflected in how the interviewees explained the decline of GM in the federal public service. Many of the respondents contended that the prevailing view among senior management and staff members was that with regard to GM, “the battle has been won” (Jamie). The availability of gender analysis training as well as the existence of accountability measures for GM belied the need for dedicated gender units or analysts. “They say it's been mainstreamed … that there are enough people in the department who have been trained and know how to do it that there's no need to have people actively teaching or guiding anyone” (Jamie).
The disappearance of GM champions was also noted by many of the interviewees. The literature on gender mainstreaming underscores the central role that senior management plays in building capacity for gender analysis. Located in key leadership positions in the public service, GM “champions” are senior managers who actively use their positional authority and political capital to rally resources and support for gender work in their organizations. Among most of the interviewees, GM awareness and support among senior managers was regarded as the critical factor in maintaining the visibility of gender considerations and analysis in their organizations. When asked about the conditions that facilitate or hinder their work as gender analysts, Cameron replied, “You know the expression, poop trickles down, not up? Well, it's kind of the same. Like if your manager says [it] is important and needs to be done, you're gonna care. If your DG [director general] either never talks about, doesn't know what it is or says: ‘eh whatever, isn't that just for women anyways? We don't do any of that.’ Then you're not going to care.”
For several respondents, their ability to engage in gender work within their units was facilitated by the presence of senior managers knowledgeable about GM and committed to this type of analysis. In describing a former champion who was particularly committed to gender and diversity work, Cameron nostalgically recalled, “he was amazing. He totally got it and was able to speak to it really well.” Many of the stories that emerged from the interviews, however, portrayed senior mangers as indifferent, hostile or resistant to GM. Another respondent, Tristan, described numerous efforts to get senior managers on board with GM and how they ultimately failed. “[W]e had flagged very heavily the OAG report and the government-wide action plan to strengthen the use of GBA. And frankly, senior management couldn't have given a flying fuck.” Another respondent, Pat, who in the past found women in senior positions to be more receptive to GM now received an “eye-roll” reaction from both male and female senior managers. “Definitely get the eye-roll from both [male and female]. We've had meetings with senior management in departments where senior management was all women, and the room was enthusiastic, and we were like ‘sweet!’ But now it's just total skepticism.”
The reasons given for the disappearance of GM champions in recent years varied. Some respondents stated that there was a general perception that there was sufficient GM capacity in their organizations and therefore GBA champions were no longer needed. “I don't think the department's interested in capacity building right now … the department's stance is that GBA is mainstreamed [and] all employees should be mainstreaming GBA” (Alexis). Other respondents, however, attributed the lack of support for gender analysis among senior managers to a general but subtle skepticism toward gender specialists, diversity issues, and feminist activism in the Canadian civil service.
You know, we're part of our own society. I think there is an anti-feminist backlash: “All those gender specialists, they have so much power … I think we're not going to challenge them directly on this because that would be politically incorrect because we're progressive.” But I do feel that there is an underlying sentiment around this that people don't voice but is nonetheless there. (Tristan)
The foregoing passage echoes previous accounts (see Chappell Reference Chappell2003) of the prejudice and resistance experienced by Canadian femocrats in a public service that privileges anonymity and nonpartisan neutrality and has “little tolerance of advocates in general and feminists in particular” (Chappell Reference Chappell2003, 106). The stories of gender analysts demonstrate that while in the past, resistance to gender-related issues may have been overt, in recent years, it has taken on a subtler form.
As the presence of GM champions, units, and specialists receded, the visibility of GM work diminished in the public sector. Respondents relayed how, in recent years, their work as gender analysts had moved away from analysis and policy development and toward a greater preoccupation with performance measurement, compliance imperatives and training and education. For most of the respondents we interviewed, “doing GM” in the federal public service meant complying with accountability frameworks. Describing the in-house training in one department, Alexis explained how most of the analysts enrolled in the session were primarily interested in learning how to meet administrative imperatives rather than conduct robust analysis: “It's for treasury board submission—very specific. That's about it. Because I think that's what most analysts were very interested in. It was not necessarily the history and background on GBA, nor the definitions between gender-integrated, gender-neutral, gender-specific.” Cameron described a situation in which she openly expressed her annoyance with staff members who, rushed to complete their Memoranda to Cabinet (MCs), would ask her for the gender analysis “template language”: “Oh, we have the committee meeting in a week and we gotta get those signed off so what's the template language that you guys have for the MCs, which to me, sends me in a rage. I'm like: ‘We don't do template language! I don't know where you heard it from but that is not what we say.’”
In these texts, we see how gender and diversity concerns, reduced to accountability measures, were often perceived as added layers of red tape within organizations. They also revealed a sense of disempowerment especially among participants, who, nostalgic about the past, were now disengaged from “real” policy analysis. “It's like we could create a drop down menu for gender analysis … before we used to help write those sections … But then what we were seeing at one point, years ago, was sort of stock paragraphs that would just be listed—so there was no real analysis being done” (Sidney; emphasis added).
These accounts of GM reveal a shared story of isolation that speaks to the decline of support for equality work within the public service. Once a part of GM teams, gender analysts now found themselves working alone, located away from where policy decisions are made in their organizations. This organizational isolation made it more difficult for analysts to engage in the relational aspects of gender work, such as capacity building and networking within and beyond their departments. Analysts also conveyed a sense of abandonment when discussing the current lack of commitment for GM among senior officials. In their stories of isolation, gender analysts positioned senior managers who championed GM as the protagonists and those who did not as the antagonists. Many of the analysts saw senior managers as playing a crucial role in enhancing the use and legitimacy of gender work in their organizations. Lacking support from senior management, gender analysts would be left to their own devices to find opportunities to make gender considerations matter in their departments.
STORIES OF DISEMPOWERMENT: THE DISAPPEARING SPACE FOR POLICY ANALYSIS
It's the thinking, the substance, the evidence … I always felt that we could carve out a space [for that]. I've found in the last couple of years that space increasingly disappeared. And that's one of the reasons that I disappeared … finally. (Sam)
In recent years, many policy scholars and practitioners have witnessed the decline of evidence-based policy advice within government, despite official rhetoric to the contrary. The bureaucratic monopoly of policy expertise of the past has given way to a burgeoning marketplace of policy advice emanating from a variety of different sources, including think tanks, epistemic communities, universities, and ordinary citizens. This new marketplace of policy ideas, along with greater adherence to the principles of efficiency and effectiveness within the public service, has significantly impacted the nature of policy work in the bureaucracy (Colebatch Reference Colebatch2007). The emphasis on managerial accountability and on maintaining a clearer distinction between politics and administration has diminished the presence and influence of policy analysts and researchers in the public service (Rhodes Reference Rhodes1994).
These broader trends in the public service figured prominently in the individual stories of gender analysts. Indeed, from these stories emerged a shared narrative of disempowerment in which participants frequently spoke of having limited “space” to conduct gender analysis in the federal bureaucracy. Similar to the story of isolation, discussed earlier, underlying the story of disempowerment is a story of decline. Within this context, all of the respondents spoke of the challenges posed by administrative and political imperatives in the Canadian civil service for conducting gender analysis in their respective units. For many, the lack of commitment to GM was not only a result of general misgivings about “feminist” or diversity issues or initiatives among public officials but also a product of the diminished status of policy advice and expertise within the civil service in general. The outsourcing of policy advice to think tanks and consultants, greater public scrutiny of bureaucratic operations, and a growing “trust gap” between public servants and political officials (Côté, Baird, and Green Reference Côté, Baird and Green2007) are among the trends identified as (re)shaping and circumscribing the role of public servants in policy development. The experiences of many of the interviewees are reflected in the following statement by Morgan, who described how her analytical work was being shaped by the priorities and preferences of political officials. “In my experience in developing policy … and providing evidence for supporting policy—it really felt as though it was more top down. PMO [Prime Minister's Office] decides what they want the policy to say. Then it's the analyst's job to find the evidence to support that. It's not the other way around.” The interviews substantiated Wellstead and Stedman's observation that in this climate, “public officials are now inclined both to promote easy policy options that are certain to be preferred by politicians, and to engage in “fire fighting” by focusing on immediate political issues. Relatively neglected is long-term policy planning” (Reference Wellstead and Stedman2010, 896). Jessie, who had worked on GM for many years, described this situation as senior managers abandoning their responsibility in a democracy to “speak truth to power”:
[T]hey've abandoned their responsibility to provide fearless advice. And instead [they] are just tailoring their advice to what the minister/the minister's office is going to like. And that's across the board, that's not just gender, that's everything, but gender falls victim to that. The public service is not fulfilling its obligation to Canadian citizens because that's our role in the democracy. (emphasis added)
At the more junior levels, interviewees described how these broader political trends were shaping the organizational context and daily work life of gender analysts. Many of the stories, including Jessie's, spoke of “shrinking space” or “gutted policy capacity” in describing the challenges they faced in engaging in gender-based work in the federal bureaucracy.
I would say analysis at all levels is—I mean the space to do analytical work, at all, within our department, was shrinking. So it was increasingly challenging just to provide the level of detail that we would assume you would want, if you wanted to know about anything. … . . it was just getting increasingly hard to drill down to the level of detail we would be collecting if you were doing good GBA. (Jessie)
Along with lack of senior management support, time constraints figured prominently in the stories of disempowerment. For some gender analysts, time constraints were seen as part and parcel of the policy development process. Like all forms of analytical work in government, gender analysis is performed in a bureaucratic context characterized by administrative rules, routines, and operating procedures that often limit time for analytical work (Meltsner Reference Meltsner1976). Rather than providing input into policy or program development, gender analysts found themselves present only in the final stages of the process to “check off” gender and diversity “boxes.” As Cameron explained, “If I'm looking at overall trends in terms of what I'm asked, for help or anything, it's usually in the context of an MC. That's the first you hear of it. Even though we push that GBAs should be thought of way, way before that, that's usually when it comes into play because now they have a form they have to fill out.” When asked whether her involvement in policy development was reactive, Cameron responded, “Absolutely. And it's mostly with not that much time because we have to go to [a] committee meeting in three weeks and we need the form done. So, what do we do? So, it's usually a quick and dirty preliminary analysis.” This response, which captures the experiences of many of the interviewees, substantiates other research that has found public servants in the Canadian bureaucracy to be dedicating more of their time on internal performance reporting and managing political sensitivities rather than developing in-house policy expertise (Côté, Baird, and Green Reference Côté, Baird and Green2007).
The disappearing space for consultations with communities also emerged from the interviews. Consultations figure prominently in gender analysis as a means of bringing the voices of citizens and stakeholders into the research and policy process (Rankin and Wilcox 2004). In practice, however, the mechanisms and resources to engage societal groups in analytical work and program development have disappeared over the past years. “So there were these mechanisms in place, whether formally or informally, that would help to get those voices in play, whether through direct consultation or through some of these other means, right? Not there anymore” (Sam). Another analyst, Cameron, explained that while consultations with communities are regarded as an integral component of gender analysis in her organization, they are often not feasible given time constraints and lack of human and financial resources.
So we advocate for that [consultations], we always say it's one of the most important parts of GBA. However, the way the government works right now, we never have money, there's never time—time is also very funny. It takes time to consult and money and we have none of those apparently. Well, for years, actually, but it's been dwindling.
In general, the portrait of GM that emerged from the interviews was one that did not include the input of citizens and civil society groups, which limits its potential to expose and remedy intersectional inequalities, as observed by the literature on intersectionality and gender mainstreaming (e.g., Hankivsky Reference Hankivsky2005; Hankivsky and Cormier Reference Hankivsky and Cormier2011; Squires Reference Squires2005, Reference Squires2009; Verloo and Walby Reference Verloo and Walby2012).
Across these individual accounts, a shared storyline of decline and disempowerment emerges. In the wake of significant changes to the public sector, gender analysts found it increasingly difficult to engage in the type of analytical work they feel they should be doing. When describing their day-to-day work, gender analysts revealed a sense of alienation from the programs developed and administered by their departments. They were no longer actively engaged in providing strategic input into policy development; rather, they spent the majority of their time meeting accountability requirements. However, as the next section reveals, analysts would use these challenges as opportunities for advancing GM in their department.
STORIES OF RESISTANCE: IN SEARCH OF “HOOKS,” “LEVERS,” AND “ENTRY POINTS” FOR GM
Many of the respondents spoke about how their ability to conduct gender analysis was hampered by organizational and political constraints. They bemoaned the fact that GM had been reduced to an accountability requirement—often competing with demands for greater economic efficiency in the public sector. Despite these impediments, their stories also revealed a shared narrative of resistance as they took advantage of organizational opportunities and dominant bureaucratic discourses to advance gender and diversity work in their organizations. Such a narrative follows a similar story arc to the story of control, as defined by Stone (Reference Stone2002, 143), in which narrators reveal ways of controlling situations that were previously thought uncontrollable. In this context, gender analysts found ways to resist bureaucratic constraints and create space, albeit limited, to advance GM within the public service.
For many of the gender analysts we interviewed, the 2009 OAG report became an entry point for persuading senior managers and coworkers to build GM capacity in their organization. As mentioned earlier, the report found considerable variation in the development and implementation of GM frameworks across the federal public service. Following the report, central agencies and line departments were asked to establish a plan to fulfill their commitments to gender-based analysis. For many respondents, the Auditor General report and the action plan that followed it represented a window of opportunity to increase the visibility of gender analysis within their agencies. Alexis, whose department was among the hardest hit in the report, referred to the report as “one of the levers” used to rationalize the need for more training in gender analysis. The OAG report was an especially effective “hook” for middle managers to advocate for gender analysis to their superiors. As Sam recalled,
What's interesting, middle-management, though, was willing to push gender analysis requirements in the comments they made because … we had this greater hook from the government of Canada accountability requirements … So we leveraged it up … so our middle management was willing to carry it up. And I think if we didn't have this government of Canada requirements and it was just talking our [own] gender policy, forget it. But because we had this other thing, and trust me we used it, it was phenomenal. (emphasis added)
Over time, however, this external push was met with countervailing forces that limited the “space” in which gender analysis and any other forms of analytical work could be performed, as noted earlier. Faced with the “squeezing” out of analysis from their work, most of the respondents engaged in strategic behavior to build support and capacity for GM in their agencies. For example, many interviewees described how they strategically framed the issue of GM within the dominant discourses of efficiency and “good policy analysis” that prevail within the Canadian public service. The following statement from Pat captures how gender analysts discursively positioned themselves as “good” policy analysts: “If the best advice is that we have to look at all of these different communities, or all these different groups, and the intersecting factors of diversity, then you're doing your job. I would argue that if you're not doing what we call intersectionality then you're probably not doing your job very well.” This passage substantiates Rankin and Wilcox's observation of femocrats, who in the current political and bureaucratic climate have to “adopt an economic-rationalist rhetoric and to justify their actions to advance equality within a market-defined discourse emphasizing the principles of efficiency, accountability and affordability” (Reference Rankin and Wilcox2004, 54).
Linking GM to traditional rational policy analysis was a common discursive strategy used by analysts to garner support for their work. Another related strategy centered on delinking gender analysis from its feminist roots to get buy-in from their co-workers and senior managers. As Alexis explained, “You want to sell it, and you want to sell it to people who don't know how to do GBA, particularly men. I will say—I have no problem saying that—and old school. So you have to kind of have to separate that notion of feminist from GBA.” Pat, who was involved in training, recognized the link between feminist theory and GM but also thought it prudent to avoid using the feminist label during training sessions.
Sometimes it's better to avoid those labels because then people—with all due respect—when you say the word “feminist” people get a little confused about its meaning. In the training and when we do awareness sessions and that kind of thing, we often just make the point that “you're doing this already. We just want to give you the tools to make sure that you are doing it consistently and make sure that you are aware of your obligations.”
While some of the respondents regarded middle or senior managers as entry points for advancing GM, others focused their strategic efforts on “embedding” gender work at the programming level. As discussed in the previous section, several interviewees complained about what they regarded as the template or “drop-down menu” to gender analysis in their organizations. For Rory, however, integrating gender and diversity considerations in programming templates constituted a subtle yet effective strategy to advance gender work in her organization. “[W]e were femocrats. And what I learned is, don't go to [senior management] and say: ‘We need to have mandatory reporting in the system.’ We take the computer guys [out] for coffee … When are you doing the next update? Do you think we could get three little boxes? And they call you and they'd say: ‘Yeah yeah, it's coming.’ What were your boxes going to look like? Those are the people you talk to.”
Within the texts presented here, we see how individual analysts created a space for gender-based analysis by adapting, disrupting and re-inscribing dominant bureaucratic and managerial discourses. In their stories, analysts described how they used the accountability requirements and the OAG report as discursive and institutional resources to garner support for GM. This strategy mainly targeted senior managers who, in analysts' stories, were portrayed as either the bottlenecks or facilitators of gender analysis in their organizations. To enlist organizational support for GM, the majority of analysts also downplayed its feminist origins, focusing instead on framing GM as “good policy analysis.” Among these stories, Rory's story of resistance stood out in that it focused primarily on working horizontally, that is, across functional units, rather than vertically, with senior managers, to integrate gender in the work of her organization. Moreover, in describing herself as a “femocrat,” Rory, unlike the others, openly positioned herself as both a feminist and bureaucrat when advocating for GM. Collectively, these stories echo Eyben's (Reference Eyben, Sandler and Rao2012) description of gender specialists as “conservative instrumentalists” who, not able to strive for organizational or social change, work within the rules of the game to bring gender and diversity concerns into policy deliberations.
DISCUSSION AND CONCLUSION
Implementation research tells us that there is often a gap between what a policy says and how it is enacted in the day-to-day practices of local bureaucratic actors. This gap, however, continues to be a theoretical and empirical oversight in the gender mainstreaming scholarship (however, see Lombardo and Mergaert Reference Lombardo and Mergaert2013, Reference Lombardo, Mergaert, Bustelo, Ferguson and Forest2016; Meier and Celis Reference Meier and Celis2011). Conceptualizing implementation as a top-down process, GM scholars have, for the most part, cited insufficient resources, poor design, and ambiguous policy language as the main culprits for GM's limited gains. The micro-level dynamics of implementation, including the local actors that enact GM in their everyday work, have largely been ignored (for exceptions, see Lombardo and Mergaert Reference Lombardo and Mergaert2013, Reference Lombardo, Mergaert, Bustelo, Ferguson and Forest2016; Meier and Celis Reference Meier and Celis2011). Emphasizing metanarratives, this study addresses this gap by focusing on the local dynamics of policy implementation, showing how they both reflect and constitute GM more broadly.
This study illustrates that the local context matters to GM implementation, for it is here that macro- and micro-level forces interact to influence how local actors individually and collectively “make sense” of GM. Through narrative analysis, the article identifies a number of shared storylines among gender analysts in Canada that reveal how neoliberalism and the politics of austerity (re)constituted the space in which they are enacting GM. The stories of isolation, for example, show how following program cuts and the internal reorganization of government agencies, gender analysts found themselves physically and intellectually distanced from the locus of policy making. This, in turn, would also keep gender considerations at the margins of policy making. The sense of isolation was further compounded after GM units were dismantled and analysts who were once part of a larger team were dispersed across organizations. Moreover, funding cuts to women's advocacy groups and the broader diminishment of feminist engagement with the state left gender analysts without outside allies or partners in their GM work. Once a collaborative exercise, “doing” gender analysis now meant going it alone, with little support from senior management and little input from outside experts and women's groups. While GM scholars have stressed the importance of feminist research and activism for a transformative GM strategy (Hankivsky Reference Hankivsky2008; Squires Reference Squires2005; Verloo Reference Verloo2005; Woodward Reference Woodward, Christiansen and Piattoni2004), the stories of isolation reveal a continuing disconnect between the theory of GM and its practice in the Canadian bureaucracy.
The stories of disempowerment also highlight the broader administrative and political forces that are shaping GM “on the ground.” As a group, respondents shared their frustrations about the diminished capacity and space for engaging in “real” policy analysis and their persistent attempts to reaffirm their legitimacy as participants in policy development. From these stories, a narrative of decline emerged, with respondents attributing GM's limited gains to the changing nature of policy work more broadly. In telling their stories, gender analysts would compare their current “box-ticking” tasks to the “substantive” work they were able to perform in the past, under previous and “friendlier” governments. While not explicitly stated, the stories represented the election of the Conservative Party as a significant “point in time” for GM, bringing about changes to gender politics and the political-bureaucratic dynamic within the Canadian federal government.
While gender analysts lamented the technical-rationalist orientation of GM, at the same time, they used it to make gender matter in their agencies. The stories of resistance uncovered in this study show how, through challenge and reinscription, gender analysts created opportunities to “trouble,” if not discredit, dominant managerial discourses that depict gender equality as a top-down, bureaucratic activity. These strategies, however, often appealed to notions of accountability and evidence-based policy making, thereby leaving the dominant managerial discourses intact. Moreover, by delinking feminism from gender analysis and appealing to notions of “good” policy analysis, the majority of analysts drew upon a rational understanding of policy work and of policy development in general, thus muting the gender politics from which GM emerged (Prügl Reference Prügl2011; Teghtsoonian Reference Teghtsoonian2003). The transformative or emancipatory potential of GM did not come up during the interviews; rather, GM was represented as an effective technocratic tool for achieving program outcomes.
Our interpretative study, while historically and geographically situated in Canada, offers broader insights for GM researchers and practitioners. The study highlights the importance of local context and local actors in shaping the content and outcome of gender mainstreaming strategies within the public sector. Our findings suggest that by shifting the analytical focus away from centralized policy makers to local implementation actors, researchers can develop a better understanding of how macro- and micro-level forces interact to hinder or enhance GM's transformative potential. In our study, for example, a micro-level lens allowed us to uncover how neoliberal principles and austerity politics (e.g., economic efficiency; deficit reduction; reduced program spending, etc.) and changing political-bureaucratic relations helped shape how gender analysts made sense of GM. Faced with a political climate hostile to gender politics and to expert advice, gender analysts acted as “conservative instrumentalists” by using audit language and performance measurements to make gender considerations matter in their organizations.
Our findings also represent a cautionary tale for a gender mainstreaming strategy that is not backed by a robust women's policy machinery. GM scholars (McBride and Mazur Reference McBride and Mazur2010) have long argued that gender experts and women's agencies within the state play a central role in advancing a transformative GM strategy. However, in Canada, similar to other countries, women's policy agencies and women's movement groups have receded as gender equality becomes increasingly viewed as “everyone's” responsibility. Challenging the view that gender considerations have been integrated into every aspect of policy work, the stories presented in our study show an alternative reality marked by disappearing GM units, focal points, and champions and ongoing struggles to make gender a priority in government agencies. Without institutional backing, gender analysts found themselves “going it alone” and struggling to make themselves and their work visible in their agencies.
These findings offer both hope and concern for GM advocates and practitioners. They offer hope in that the analysts in our study were keenly aware of the challenges facing GM and acted strategically to incorporate gender in everyday policy work and bureaucratic practice. However, the findings raise concerns about the degree to which GM is, or can be, transformative. As much of the GM scholarship has observed, the potential for technocratic bias in shaping GM initiatives often binds its potential for socially just outcomes (e.g., Haffner-Burton and Pollack Reference Hafner-Burton and Pollack2002; Meier and Celis Reference Meier and Celis2011). Moreover, the role of non–gender analysts in GM processes stymies progress (Benschop and Verloo Reference Benschop and Verloo2006; Lombardo and Mergaert Reference Lombardo and Mergaert2013, Reference Lombardo, Mergaert, Bustelo, Ferguson and Forest2016), as gender analysts not only compromise in their attempts to get things done but also rely on senior managers to support their work and signal the importance of GM. In sum, our findings support an observation made by Lombardo and Mergaert (Reference Lombardo and Mergaert2013, 309) that perhaps the key to unlocking the transformative potential of GM is to anticipate these concerns when designing and reforming GM strategies. As we have shown here, understanding how analysts “make sense” of GM is a necessary first step.
