Into the Purple Zone: Deconstructing the Politics/Administration Distinction

Abstract

We add new data to the long-standing debate about the interface between politics and administration, deploying theory and evidence indicating that it varies. It can be either a “purple zone” of interaction between the red of politics and the blue of administration, or a clear line. We use survey responses from 1,012 mostly senior public managers in the United Kingdom, Australia, and New Zealand, along with semi-structured interviews with 42 of them, to examine the extent to which public managers perceive that they “cross” the line or go into a zone, and the ways in which they do so. Our inclusion of a zone as well as a line recasts how roles and relationships between politicians and administrators can be conceived. Moreover, it raises questions about how particular contingencies affect whether public managers perceive and work with a line or a zone.

Keywords

politics/administration dichotomy political management political astuteness political skills

Introduction

The perennial debate about the validity and extent of the separation between politics and administration has given rise to numerous interpretations and labels since Woodrow Wilson’s (1887) classic formulation and Weber’s (1922/1994) theory of bureaucracy. Empirically, the central argument is about whether such a line between the domains of politics and administration exists (Campbell & Peters, 1988; Svara, 2006a), although the dispute takes many shapes. Contending interpretations of whether Wilson really meant a dichotomy, a separation, or a boundary jostle with alternative terms such as complementarity, separability, independence, or subordination, each with its own implications for the nature of the interface (Svara, 2006b). Probably the majority but certainly not exclusive view is that the dichotomy is theoretically and empirically questionable (Aberbach & Rockman, 1994; Behn, 2001; Demir & Nyhan, 2008; Overeem, 2005), and that both politicians and administrators often breach it in practice (Sayre, 1958; Svara, 1999). Instead of a clear delineation between the domain of administration and that of politics, public managers (and indeed politicians) may encounter a blurred area, in which there is a degree of indeterminacy about the roles and relationship between the two domains (see Figure 1a). Sometimes they may perceive a shifting line, where the boundary moves. Svara (2006b) proposes that a continuum may better describe the variety of relations between elected politicians and public servants.

Figure 1.

Images of the dichotomy.

Normatively, there is contention about how legitimate it is for public managers to cross that line or blurred area. Although at various times since Wilson, the traditional position has supported the dichotomy on democratic grounds, in recent decades, its legitimacy is under challenge. There is a widespread view that public managers as policy experts have a duty to be involved in policy deliberation, by placing their expertise in the service of the elected government—and by implication, the citizenry (Moore, 1995; J. Wilson, 1989). At the same time, other scholars see the separation as “a useful fiction” (Peters, 2001, p. 82), and suggest that it plays a legitimizing role.

In consequence, the literature is replete with contending positions. It lacks a broadly settled view about the actual or desirable relationship between the political and administrative realms (Demir, 2009). Much of the literature involves scholars arguing about binary choices or pursuing “one-size-fits-all” constructs: Is the interface between politics and administration typically a clear line or a blurred one? For the sake of exposition, we refer to the blurred or shifting line as a purple zone, representing where the “red” of political activity overlaps with the “blue” of administration (see Figure 1b).

Based on our interviews, we see the zone as an area of transition. It was explicitly articulated in New Zealand (Matheson, Scanlan, & Tanner, 1997) as “an amalgam of separation and integration . . . an arena of conversation between ministers and their senior officials” (Matheson et al., 1997, p. 5). However, some studies have identified not universally fixed but varying types of interface between politics and administration (e.g., Aberbach, Putnam, & Rockman, 1981; Svara, 2006b)—an approach likely to yield a more realistic understanding of the roles and relationships.

Also rarely considered in the literature is one of the key activities and functions of the relationship: the making of decisions (and of “non-decisions”; Bachrach & Baratz, 1962; Lukes, 1974). The relative influence of managers and politicians in political matters usually takes shape in the form of decisions or decision-shaping, but hardly any of the research looks at how decisions (or non-decisions) are made and in whose interests, and therefore how power is distributed between the two realms.

This article seeks to broaden our understanding and offer a new metaphor. First, with respect to the “line” or “zone,” we argue instead for a variable approach, in which neither is the dominant arrangement, but the nature of the interface varies according to circumstances. Our inclusion of a zone as well as a line broadens the conceptualization of roles and relationships between politicians and administrators.

Second, we illuminate decision making by deconstructing the steps through which a decision passes. We deploy a widely used framework for thinking about policy making, the policy cycle, to explore this. We also apply the distinction elaborated by Bachrach and Baratz (1962) and Lukes (1974) between a “decision” and a “non-decision.”

Drawing on 42 depth interviews with mostly senior public managers in the United Kingdom, Australia, and New Zealand, and data from an associated survey of 1,012 public managers, we explore the extent to which public managers perceive that they “cross” the line or zone, and the ways in which they do so, by addressing the three following questions:

What is the nature of the relationship or interface between politics and public administration: a clear line, a blurred zone, or varying between the two?

In what ways do public managers say they engage in politics? Why?

How legitimate do they report this engagement to be?

After defining key terms and outlining the contours of the debate, we explain our methodology. Then our main findings shed light on how public managers themselves perceive and react to the zone as well as the line. We explore their accounts of the contextual factors giving rise to those responses.

We find that the normal functioning of government requires public managers to engage with politics in various ways to create what they see as value for the public.¹ Managers’ forays into the political domain are influenced by their concern to implement tasks and deliver good quality services in the public interest. Many report working in a zone with some degree of ambiguity, but they see this as necessary given the complexity of tasks and roles. At the same time, they also need to be sensitive to the boundaries beyond which they should not push. How this plays out in practice varies according to the situation. Our notion of a zone conceptualizes complex and interconnected tasks where roles interact, contrasting with earlier ideas of governance based on legally defined roles where a “line” can be policed. We submit that this notion better encompasses these interactions.

Definitions and Constructs

Obviously, a central issue is what we mean by “politics". Drawing on our earlier work on political astuteness, politics is defined as

about mobilising support for, and consent to, action in the context of diverse, and sometimes competing interests and may involve either collaboration or competition depending on purpose; that politics can be legitimate as well as illegitimate; that it can be about pursuing either or both of self-interest and organisational interests; and that these activities can take place externally or internally to the organisation. (Hartley & Fletcher, 2008)

Two aspects of this general definition require elaboration. First, it covers politics both internal and external to the organization, because the authorizing environment confronting each manager is likely to contain other, sometimes more senior managers in the same department as well as actors outside the organization. (Consequently, it involves interacting with the public domain, and with the institutions, interests, and processes within that domain, such as the public interest, public trust, and fiduciary duty. However, these phenomena are not our central focus.) Second, it encompasses public managers’ relations with a diverse range of stakeholders, but concentrates particularly on their work with formal, elected politicians—as our focus is on the dichotomy—while taking the others into account where relevant.

Political behavior in public managers contrasts with what are typically seen as administrative behaviors and stances. These are neutral, non-partisan, scientific or relying on technical expertise, and relate to the gathering of information, the smooth running of government agencies, and the practical implementation of policies and decisions made by politicians (Svara, 1999). Politicians do of course engage in managerial or administrative behavior (i.e., entering the purple zone from the “red” side) but it is beyond the scope of this article to encompass this issue (see Di Francesco, 2012, for a discussion of the managerial role of ministers).

Against this backdrop, we focus on the roles of public managers in stages of the policy cycle and the pattern of decisions or non-decisions they make within each. We start from the simple premise that managers engage in politics to influence decision making, and that their perceptions of how much they do this, and in what manner, will provide insight into their political roles.

This raises the issue of what to call the public servants and their political masters. The literature offers many meanings, borne of complex debates. Rather than try to find the “correct” definition, for the sake of consistency, we adopt the following terminology. First, we use the term public managers—that is, officials who are appointed and who serve the government of the day (or in the case of some local/devolved governments, serve all elected politicians on the council or assembly), often serving several governments over the course of their career. Where others use the terms senior public servants or administrators, we will use them as substitutes, although our own preference is to recognize the complexity of managerial work by labeling them managers.² Second, we focus on people called politicians—that is, officials who are elected.³ As our sample mainly comprises senior managers working to portfolio ministers in state and national governments, the politicians we discuss here are largely the limited class who exercise executive authority (i.e., members of the political executive). To a lesser extent, we also discuss local government chief executives working to local councillors and leaders, and those managers dealing with politicians representing their constituencies (i.e., backbenchers).

Finally, we use the term non-decision. In a nutshell, if a “decision” is “a choice among alternative modes of action,” a non-decision is made when an actor affects the surrounding consciousness, structures, processes, or other factors that shape what is perceived as possible or desirable, so that those who might have expressed or exercised their preferences never even get to do so (Bachrach & Baratz, 1962).

The Politics/Administration Dichotomy

This article focuses particularly on Anglo-American (with data from Westminster) systems, and thus its point of departure is Woodrow Wilson’s “Study of Administration.” Contrary to common attribution, Wilson himself did not use the term dichotomy, and the idea was not fully crystallized until the 1940s (see Svara, 1999).⁴ Its primary focus was separation, positing a clearly defined line between elected politicians, whose role was to decide policy, and administrators, whose job was to carry it out—each remaining in his or her own territory (Goodnow, 1900).⁵ From the 1920s, the separation hardened: “Scholars . . . uncritically accepted the existence of a comprehensive dichotomy model” (Svara, 1999, p. 684).

However, from the 1950s, the dichotomy was the target of numerous critiques, on the grounds that it lacked empirical validity because it is often breached in practice—by political leaders as well as public managers (Sayre, 1958; Svara, 2001; Waldo, 1984). The dichotomy has been extensively debated since then, with the weight of the argument tending to view it with skepticism (Demir & Nyhan, 2008; Overeem, 2005; Peters, 1987; Svara, 1999, 2001, 2006a).

Another perspective is that there are necessary interconnections between the two realms. Svara’s (1999, p. 678; 2001) complementarity of politics and administration is “characterized by interdependency, extensive interaction, distinct but overlapping roles, and political supremacy and administrative subordination coexisting with reciprocity of influence.” Although the separation theme tended to overshadow the connection idea, Svara (2006a) argues that the latter informed the theories of many public administration scholars.

Also important, but peripheral to these conceptualizations, is the literature on “public service bargains,” which posits various models of interaction between public servants and politicians based on agreed exchanges between them of both tangible and intangible goods (see, for example, Hood & Lodge, 2006). It implicitly argues that the normative validity of public managers’ encroachment into politics varies from regime to regime. However, this school of thought, although promising, has attracted only modest empirical research.

This variability is also evident in a few cross-national studies. Aberbach et al. (1981) defined four “images,” ranging from the starkest dichotomy (Image I), to convergence between politics and the bureaucracy (Image IV). However, their classic study was published more than 35 years ago, and its data were collected about 45 years ago. Since then, major contextual changes have occurred. Indeed, Aberbach et al. (1981) themselves describe that period as “a time of some turbulence.” We have seen the emergence of “wicked problems” (Rittel & Webber, 1973), the reorientation of political forces, and most particularly the development of the New Public Management (NPM), which gave the dichotomy a new lease of life (MacDermott, 2008) with its separation of “steering from rowing” (Osborne & Gaebler, 1993). Then came new public governance, which has blurred it again (Torfing & Triantafillou, 2013). Other studies similarly suffer from being dated (Heclo & Wildavsky, 1974; Kaufman, 1969; Peters, 1987; Schaffer, 1973). More recently, both Hood (2002) and Svara (2006b) have constructed typologies of the relationships, with some types paralleling those in Aberbach et al. (1981). But these later studies are theoretical rather than data based.

Thus, although this discourse has generated a substantial literature, only a small proportion has proffered much empirical evidence (Aberbach et al., 1981; Carboni, 2010; Demir & Reddick, 2012; Miller & Wright, 2011; Mouritzen & Svara, 2002; Page, 2012; Stocker & Thompson-Fawcett, 2014; van der Wal, 2013; Zhang, 2014). Moreover, of those works offering such evidence, not many look in detail at public managers’ attitudes, beliefs and behaviors in respect of the dichotomy—that is, what they actually think and do when they are straying into politics.

The metaphor of “the line” has been prevalent in the literature, but our data indicate first that many public managers regard it more as a “zone”—a shared space between politicians and public managers, rather than one in which politicians inhabit one realm and public managers a separate one. At the same time, it is more realistic based on our data to say that relationships between politicians and public managers vary in type.

Method

This article uses mainly qualitative but also quantitative data from a large study examining the political astuteness of public managers. We surveyed 1,012 public managers in three Westminster countries: Australia, New Zealand, and the United Kingdom,⁶ and interviewed 42 public managers. The wider study looked at a variety of issues concerning public managers and politics (Hartley, Alford, Hughes, & Yates, 2013), but here, we focus specifically on the dichotomy. Our intention was to understand how managers see politics and what kinds of activities public managers engage in when they say they are being political. The interviews enabled us to obtain a nuanced view of participants’ self-reported perceptions and behaviors in their relationships with politicians.

The interviewees were a selection of volunteers from our survey sample, encompassing national, state, and local governments, and covering central and line agencies as well as a range of professional services (e.g., policing, environment, and human services) and all regions in each country. Additional U.K. interviews came from a related research project (Manzie & Hartley, 2013). Participants were assured that responses would be anonymized. The interviews were semi-structured and followed up some of the areas of the survey in more depth. All interviews except three were audio-recorded (detailed notes were taken for the rest). We used NVivo to undertake thematic coding, and also performed keyword searches on the transcripts. We did not use interviews to assess the frequency of particular views, but rather in an interpretive way to understand the perceptions of public managers. The survey question about in what situations managers found it valuable to use political skills was a 5-point Likert-type scale from no value to extremely valuable.

A challenge in analyzing a potentially controversial but also pervasive phenomenon is that respondents’ statements about it are necessarily summary accounts, which may be “stylized” or modified to fit more comfortably into their perceptions of what they are expected to say. We addressed this by first deconstructing public managers’ work into its constituent elements and asking about concrete behaviors and practices for them. We then asked directly about managers’ views of politics and the dichotomy in their jobs.

Public Managers’ Political Activities

To deconstruct the ways public managers “do” politics, we used the stages of the policy cycle. Since its original formulation by Lasswell (1958), it now has many versions (e.g., Althaus, Bridgman, & Davis, 2012; Bardach, 2008; Weimer & Vining, 1995), but fundamentally it sets out the steps in the life of a public policy from problem identification or issue framing through a series of processes to implementation, albeit in a stylized fashion (Table 1). The stages do not necessarily occur in lockstep order, and are typically more variable than in textbooks. For analytical purposes, this framework is more relevant than alternatives because it identifies domains that are typically political, administrative, or shared (Skok, 1995).

Table 1.

Elements and Aspects of Public Managers’ Decisions in Each Stage of the Policy Cycle.

Stage of policy cycle	Typical “political” behaviors of managers	Decisions, non-decisions or both?	Engagement supporting or opposing political master(s)?	Extent of public managers’ political engagement
Identifying problems/issue framing	Controlling information Defining terms Managing time Influencing third-party stakeholders	Mainly decisions, some non-decisions	Mainly supporting	Mixed
Generating options and criteria	Shaping options Controlling information Utilizing processes Adding or curbing alternative policy capabilities	Both	Mostly supporting	High
Deciding policies	Utilizing processes Adding or curbing alternative policy capabilities Influencing other stakeholders	Mainly decisions, some non-decisions	Mainly supporting, sometimes disputing	Substantial
Implementation	Utilizing processes Managing time Influencing other stakeholders	Mainly decisions, some non-decisions	Mainly supporting	Mixed

Within each stage of the cycle lie actions that public managers take that can either advance or retard a given purpose—as managerial politics might be directed toward either progressing or stifling a policy development. Analysis of the interview data shows several types of managerial political behavior (see Table 1):

Controlling information, for example, selecting or burying relevant facts for consideration by politicians or other stakeholders;

Supplementing or reducing alternative policy capabilities for the politician;

Using processes, for instance, persuading parties to submit the issue to a cabinet committee, a consultative forum, or a judicial proceeding likely to delay or distort the deliberations;

Shaping options in a way that highlights or favors some and downplays others;

Managing time to either delay an urgent decision or speed up something that requires more thorough and lengthy consideration;

Defining terms in ways that clarify, obfuscate, or distort key issues;

Persuading or influencing third-party stakeholders—beyond the politician’s office—in ways that shape their authorizing environments. This limits, enhances, or distorts politicians’ room to maneuver and constitutes an intervention in what is usually the domain of the politician.

Dealing With Political Actors

We first draw from findings in each of the four stages of the policy cycle. Then, we broaden the lens, to ask more generally about perceptions of the dichotomy. But before we go into detail for each stage of the policy cycle, we should stress that managers’ interactions with third-party stakeholders cut across all of these stages. Third-party stakeholders include other politicians or public sector organizations; other governments at national, sub-national, or local levels; lobby groups; suppliers and contractors to the organization; relevant regulatory bodies; the media; and significant bodies of public opinion—what Moore (1995) has termed the authorizing environment.

Dealing with these other actors is generally thought to be a primary function of elected politicians, and more than a few of the interview respondents made it clear that this was a useful role politicians played—insulating public managers to a degree, persuading stakeholders of positions, playing a part in negotiating outcomes, and bestowing ministerial authority on decisions.

P04: . . . there’s no way that I’m in the political domain in that sense. The actor on that stage is very much the minister and the minister’s supporters.

But it was equally necessary, and from our interviews, it seemed common, for public managers also to deal with key stakeholders in various ways.

P06: So much of our work involves the broader community, different layers of government, a whole range of different external stakeholders, and again it’s about how you engage with those stakeholders in order to achieve what you’re trying to achieve for the organization as a whole in support of what community objectives are.

A quite common phenomenon is for ministers to conduct the main dealings, but for public managers to provide them with background information about the substance of the issue and also about the players, their interests and influences (P01: “I need to be able to serve them up the information that they use to then deal with those people”).

In summary, although officially the job of dealing with external policy actors is the preserve and duty of elected politicians, in practice, they share this role with senior public servants to varying degrees. We turn now to the four stages.

Problem Identification/Issue Framing

Kingdon (2011) found in his research three decades ago that the identification or framing of an issue—that is, its emergence (or non-emergence) into the consciousness of policy actors—was largely the preserve of (“visible”) politicians rather than (“hidden”) civil servants. However, our interviews revealed a more complex picture: Although managers felt that primary responsibility for raising issues rested with politicians, sometimes managers played a role in either identifying or framing an issue. Contrary to what might be expected, they reported only a modest level of interest-based conflict with politicians. Rather, dealings were often about assisting them when they lacked capabilities.

To the extent that they did seek to push ministers in a particular direction, they tended to use non-decision-making processes such as providing or controlling information, defining terms, clarifying agendas, managing time, and influencing other stakeholders. One variant occurs when ministers don’t know quite what they want (see Di Francesco, 2000), and need the help or encouragement of expert public service advisers to crystallize or clarify their position:

P09: . . . if you have a minister who really doesn’t know what they want, the real issue is to be able to try to craft what some of the policy propositions are and you go through different sorts of processes, walk them through . . . why you think this should be a priority, and would you like to take it in this order, well we’ll do this first, so effectively you try and create the agenda for them as well.

P30: I think you need a skill where you are genuinely enabling politicians to reach a view on what they want to do, and [the skill also] allows you to help them find how they want to do it.

A version of “not knowing” was where the policy was internally contradictory, perhaps because it reflected the party platform or had not been thought through.

P01: Uh, no they’re not clear about what they want and they often say contradictory things because they’re trying to appease the whole . . . group of people.

A further gap in elected politicians’ awareness concerned issues for the long term, beyond the current political horizon—for instance, whether and how to fund the burgeoning requirement for baby boomers’ retirement incomes. This requires considering the longer-term value of policies and actions initiated now, but which may have consequences beyond the life spans of most voters.

P03: . . . in some ways it’s a classic short-term long-term thing, like they’re trying to prosecute their stuff almost on a daily basis, whether it’s surviving Question Time or the next whatever that they’re worried about, and we’re trying to look at well you’ve said that your aspiration is a long term reform of this sort, so our advice . . . is set in a longer term context.

In short, although ministers’ authority to set the agenda is broadly accepted, and they do so to a considerable degree, public managers find there are situations where they intervene in political issue framing when ministers are unwilling or unable to perform this role properly due to lack of knowledge or a short-term viewpoint. In these situations, public managers typically seek to help politicians rather than bypass or even challenge them, but in doing so they can sometimes stray well into the purple zone, effectively substituting their own policy preferences for their political masters. Only some of these situations entailed “non-decisions.”

Generating Options and Criteria

The next policy stage involves identifying options and assessing their relative value against criteria (Sidney, 2006). This is traditionally the preserve of public service policy specialists, and has the potential to influence actors in the political arena. The generation or framing of policy options can improve, limit, or distort the choices available to politicians. The extent to which this occurred was partly related to the outlook of the managers themselves and partly related to how much they had to compete with alternative sources of advice sought by the minister.

On the question of public managers’ outlooks, the short-term/long-term theme offers a clear case of non-decisions, with several interviewees seeing their role in formulating options as one of marrying political reality with longer-term objectives:

P36: I always say that my role, when I am at my best, is to be able, within the political parameters of the party politics, to lay out a series of options that will allow them to get to the outcome they want. It’s for them then to choose the outcome . . . to choose the option. . . . The more “nous,” the clearer I am about the political constraints, the clearer I can be about the options and the advice I can give in terms of how to move forward.

This effectively elevated some positions and side-lined others. Non-decision making was also prevalent where some of the otherwise viable options might be selectively left out by advisers.

P20: . . . depending on the information that you give people and how you present it, you sometimes realize that that might influence an outcome, but you still need to be careful. You certainly wouldn’t lie or anything, but you might just put a particular emphasis on something if you’re trying to get a better outcome for a client.

As with “problem identification,” the encroachment of public administrators into the zone was mainly based on their concern to help the minister be clear about what they want—but at the same time, politicians could be vulnerable to even subconscious differences between their mind-sets and those of public servants.

Deciding on a Policy Approach

This is typically a multi-faceted part of the policy cycle, a key point at which public managers might encroach on politicians’ roles. Most obviously, it entails recommending a choice based on the options provided—or sometimes, opposing a particular course of action. This is mainly about active decisions, but some participants related their experiences with non-decisions too.

Our participants reported using most of the methods listed in Table 1 for influencing ministers to adopt particular policies, or modifying ones already under discussion. These methods were not just about tendering good advice based on sound evidence to make a case for a particular position. One manager (P24) found that the minister was showing decision papers to third-party stakeholders, who gave well thought-out responses. She reflected, “we got quite upset about this until we learned how to game with it,” which meant taking “wider soundings” of the policy environment, and determining how to influence (or at least take account) of those stakeholders who in turn had influence over the minister.

On occasion, a combination of influence techniques—such as controlling information, asking questions, and managing time—can make it difficult for a particular view to gain traction—sometimes even eliminating it from consideration. The following is a clear example of a non-decision, where a manager thinks a particular request is undesirable and uses a technique of thoughtful questioning to make his interlocutors reconsider what they wanted in the first place:

P21: I’ll . . . just keep asking questions, and when you think you’ve got the answer, think of some more questions to ask. And often then the person you’re asking the questions of, they come in to get an answer, you ask them a lot of questions, and if they go away and say “oh listen I’ll need to think about this a bit more, I’ll come back to you again when I work out what question I want to do,” you’ve done your job.

This manager often found that in this way he could reveal the real problem underlying their request, and recalibrate the issue accordingly.

Not only do public managers help shape content categories but they also influence the decision-making process or the time frame for the decision, for example, in scheduling meetings or deadlines:

P03: What we tend to do with the current minister is to have what are called small group discussions, so we’ll identify something that’s not quite cooked or that needs sorting in terms of where are we going to drive the truck, both us and him, and we need to get back in—we need to get in tune, we need to sort of know that the way we’re going to do it is going to work for him and vice versa, so that we’ll organize a small group meeting . . . and talk it through. And that works quite well.

This is the typical pattern for some politicians, who may as a result be swayed in a particular direction. But more experienced ministers are usually less susceptible to these devices, and are more able to put their stamp on departments’ policy direction.

Implementation

Implementing policies is the raison d’etre of public managers, and therefore well into the blue of administration. Deploying and directing staff, facilities, and funding are clearly administrative tasks. But both the literature and our data make it clear that it is difficult to perform that function without calling on authorization or capabilities from actors in the political environment (Pressman & Wildavsky, 1973). Among the aspects of implementation that called for political engagement are securing financial resources, securing and operationalizing legislation (“legal resources”), and mobilizing staff and other stakeholders.

Many interviewees indicated that they found it hard to do their jobs of putting policy into action without the support, cooperation, effort or resources of various stakeholders—not only those with formal authority such as ministers or finance departments:

P08: What wasn’t sought in the project before I started, and probably what I didn’t do as good a job as I would like in retrospect, was building bureaucratic political support for the project. [A major department said to me] “Oh dear, [name], we’re the Department of [X]. There are over 100,000 people working for us. We’ve ignored Cabinet decisions in the past, we will ignore them into the future, so you coming in here with your little ‘Cabinet told me that we have to do it’ thing is just not going to cut it, it’s not good enough.” And at that point I realized I’d significantly politically misplayed the situation.

P19: I had to go in and actually find out why these [stakeholders] were concerned. Why was this policy that seemed on the face of it pretty vanilla, very non-controversial, creating a lot of angst and concern? . . . and eventually I just rang up people I knew in the sector and said look tell me what the hell the story is here, what’s driving it? It was something else [worrying them] . . . I mean you actually had to go behind and find out what was driving it, and then tell the manager involved to change it, even though he saw no particular reason to do so. Because it was obvious that this was going to hurt the minister big-time.

Thus, although implementation is commonly seen as separate from politics and policy, our participants found that often it was hard to draw a line between them. Apparently, administrative matters tended to burgeon into political ones.

A variant on this was where public managers calculatedly wielded an externally endowed source of influence, such as the politician’s fear of adverse publicity. One local government manager (P38) warned his political master that if he continued to insist on a policy action that would substantially deplete council funds, he would put on the record that the policy was financially imprudent and against his advice.

A Line or a Zone?

The central issue in this article is whether public managers themselves see their interface with elected officials as a line or a zone. The strong implication from analyzing the stages of the policy cycle is that this varies between managers rather than being common across all of them. This is borne out by our survey and interview data about public managers’ views of the interface, which vary considerably.

First, to understand the context of the public managers’ work, the survey asked respondents about when they found it valuable to use political skills. Respondents gave most prominence to situations where they were dealing with elected politicians: in the Australia/New Zealand case, ministers (3.44 on a scale of 0-4) or central agencies (3.14), and in the U.K. case, working with central government (3.12) or regional/local government (3.2). In all three countries, they gave almost as much value (3.1) to public opinion.

This indicates that the interface between public managers and politicians is very salient to their work. The next question, therefore, is whether they see that as crossing a line (an implicitly less legitimate behavior) or entering a zone, which they understand as being shared with politicians. Interviewees had various responses on this issue. The clear message from most was that they saw no sharp line. A few talked explicitly of a zone rather than a line, also used the terms no man’s land (P11), a gray line (P10), and blurry (P07). Some saw it as a shifting line, incorporating either more politics or more administration at different times (depending on what kind of minister they had, for example):

P06: I don’t think it is particularly clear. It never is, when people talk about us they talk about the separation of powers, but . . . to the credit of this government and our minister, he’s basically said that “I’ll leave you to run the organization, but when it comes around to the policy direction, that’s what government- and that’s what I’ll be responsible for,” and I don’t think we really had that as clearly articulated with our previous ministers.

Others saw a gap between rhetoric and reality:

P11: I think there’s a clear line formally, I think informally there is a little bit of overlap. There has to be really when you’re expected to provide advice, you do fudge it a bit sometimes . . . you know I could be very strict and say I’m giving you this advice and that’s it, I have no opinion . . . and these are our recommendations in ranked order. I could do that. But I don’t believe that happens . . . I think we do express viewpoints. I wouldn’t say they were strongly party-political, but we probably express a viewpoint according to our particular beliefs I suppose.

P07: Look I think over the years it’s become more and more unclear, you know the whole Westminster principles and the Westminster system and the stuff that you sort of learn in government 101 when you first start, and has got very blurry.

However some participants, when asked “is there a clear line between the two or is it an overlapping zone,” affirmed the dichotomy in principle and often in practice as well:

P04: I’m not entering their domain at all . . . I suppose [I’m] the person behind the scenes who does everything that can be done to make implementation of what was thought of possible.

P05: I like to keep a clear line between the two . . . How do I explain what that means for me as an executive? [pause] I think for me it’s about articulating the policy and the direction of our department and being able to relate that back to the direction being set by government rather than being out there promoting or advancing the agenda of the government directly if that makes sense.

Even when managers knew that there was substantial ambiguity in the dichotomy, some pointed to the problems involved in going too far into the zone:

P07: I think some of that is very blurry . . . in terms of looking at the boundaries I get very nervous about people trying to cleverly second-guess what government wants . . . maybe I’m being naive but I think it’s really really dangerous, I think there is still a role for providing clear advice . . . I think if you start guessing what the minister might want, you’re getting into very dangerous territory.

Thus, in some cases, managers who conceded that the dichotomy did not really hold in practice found that to be a cause for anxiety or regret. Overall, most interviewees did not see a clear line between policy and administration or a clear line between what they did and what their political leaders did. Rather, they tended to see a zone as a more accurate representation of their reality. Those who mentioned difficulties were worried about what they saw as “politicization” of the public service, but there was only minority agreement with the idea of a simple, clear line between what they did and what political leaders did or should do. Instead, many public managers found themselves affected by or engaging in politics in various ways.

How Far Should Managers Be Political?

All of this raises the question of how there can be confidence that public managers will not take advantage of their access to the purple zone to peddle their own priorities and usurp the decision-making prerogatives of politicians. Our analysis suggests some safeguards, none of them foolproof, but together amounting to a credible set of constraints on managers.

First is the simple fact that ministers have more influence over managers’ employment tenure than vice versa. Indeed, the weight of this power is increasingly on the side of the politicians under NPM-style reforms such as term contracts and performance incentives (Hood & Lodge, 2006). Second, the purple zone is a shared zone, in which both politicians and managers operate, with the likely consequence that each is subject to pressures from the other. The influence is a two-way street, not just one in which the public servant has sway over the politician. Third, our interviewees understood that even the zone has limits: Public managers can only press so far through the purple and into the red zone.

Finally, we argue that the extent to which public managers encroach into territory beyond the purple zone is influenced by values and skills of those managers. Findings from our interview data lend weight to this argument. Interviewees reported that they (and their colleagues) were more aligned to achieving public purposes than to pursuing their own agendas at the expense of the public or politicians. Of course, there could be an element of self-serving bias in their survey or interview responses, but the former was anonymous, and in the latter, they also spoke of dilemmas, mistakes they had made, and concerns that preoccupied them at times. This seemed to indicate that self-serving behaviors and discourses, if present, were attenuated. We know from some of our other work (Hartley et al., 2013) that the more senior managers report higher levels of political skills, suggesting that they become more sensitive to political dilemmas with experience.

Discussion and Conclusion

This article began by reviewing and questioning the “politics/administration dichotomy” and arguing that although it might function as a useful myth, the empirical reality is more complex. A number of writers have proposed various models to account for the limits of the dichotomy and to express different perspectives on the boundaries between politicians and administrators (e.g., Aberbach & Rockman, 1994; Svara, 2001, 2006a; Zalmanovitch, 2014). However, the issues still tend to be discussed using words such as “line” or “boundary.” We have argued for adding to (not replacing) the line with the concept of the zone, which differs metaphorically and conceptually because it connotes an intermediate space, with a gradual transition rather than a sharp boundary. Svara (2006b) comes close when he discusses overlapping roles, but his work still refers to lines and boundaries.

The zone concept adds a degree of fluidity to public managers’ movements in and out of potentially political spaces and recognizes that politics can be informal as well as formal. In the absence of clear markers, there is a degree of ambiguity in the intermediate space, so that managers might find they have gone too far across the zone. In addition, the transitional zone is itself dynamic, so that sometimes managers will find themselves in a different position than they envisaged whereas at others they will have failed to move out of their traditional role sufficiently to advance valuable purposes.

We have also argued that a proper picture of what is happening in each stage is more likely if non-decisions as well as decisions are taken into account. This is especially apposite in this field with its institutional intricacies and the political astuteness of its players.

Our empirical evidence shows that many public managers—even in Westminster systems where they are particularly expected to be impartial—find it necessary to push into the shared area of the purple zone to do their jobs better. They say they use political skills at all stages in the policy cycle, depending on a variety of factors.

Furthermore, these forays into politics have a degree of legitimacy in the eyes of public managers: Their purpose, as they report it, is not to usurp the role of politicians, but rather to assist them in achieving public purposes, particularly where issues are complex (wicked) and pressured. They endorse prosocial views and uses of politics rather than self-interested ones, and use political skills in a variety of contexts, with a range of stakeholders. This is not to say that “the line” is outdated or irrelevant. A minority of managers said they were uncomfortable with the idea of a zone, and preferred to stay firmly within the technical aspect of their role that has always been a key part of any public manager’s job. Others said they respected it at some times, but that work or external circumstances often called for positioning themselves in the purple zone.

This all relates to another potentially very useful implication of our study: It offers a platform for further research on when and why public managers work in the purple zone. From all appearances, the extent to which public managers encroach into the purple zone varies according to the policy issue and the context. This is consistent with the literature and our interview evidence. For any given issue, sometimes the politician and the public servant both play a role in the same activity, and sometimes one does so whereas the other is more quiescent. Given the variability of the phenomena under consideration, it is clear that a contingent approach would add important insights.

Such research would usefully probe the factors influencing public managers’ forays into the zone, which appear to be inherent in the contexts, structures, or processes of the public sector. First, the degree of controversy over a policy issue is a contextual factor that public managers said made them more likely to use their political skills, partly to understand and engage a variety of stakeholders on a topic, and partly to help the politician navigate choppy waters. Second, managers sometimes compensate for politicians’ inability or unwillingness to crystallize their own position on an issue. In this sense, the duo is a dancing partnership where the public manager may guide, for a short while, the steps of the politician (Manzie & Hartley, 2013). However, this is not the puppetry sometimes alluded to in the literature (or on “Yes Minister!”) of public managers usurping the role of politicians. Both their own values and their political sensitivity to context suggest that they give a helping hand but no more. The third contingency to investigate is the relative power of the two parties, which is a function both of the public manager’s and politician’s capacities and of the extent to which politicians have alternative sources of advice available to them. Differences across varied countries and therefore governance regimes would also be valuable.

Where situations are routine, stable, and uncontested, the orderly separation of roles inherent in the dichotomy may apply. However, Rittel and Webber (1973) and others note the increasing prominence of “wicked” problems. This makes the work of both politician and public manager more complex than in the past, requiring engagement with a wider variety of stakeholders. Kettl (2009) also notes the different approaches to management in government according to whether problems are routine or non-routine. Behn (2001) argues against the acceptance of “the fallacy of efficient, non-political administration” (p. 49). Finally Moynihan and Ingraham (2010) argue in the U.S. context that “the nature of governance has changed in important ways, incorporating a greater range of tasks, more complex tasks, and more complex structural characteristics” (p. 229). We can see this in non-U.S. governments as well.

The “line” between politicians and administrators may have been devised at a time when problems were largely conceptualized as “tame” (Rittel & Webber, 1973), that is, solvable with technical solutions and by “administrators.” The concept of a zone, which is a transition space, arguably better fits governmental roles and relationships where issues are wicked and where management skills are called for. There are clearly going to be limits, for democratic accountability reasons, but the debate about limits may be better served by the metaphor of a zone than a line.

Footnotes

Acknowledgements

The authors would like to thank Michael Di Francesco for his suggestions for this article.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was made possible with the assistance of a grant from the Australia and New Zealand School of Government.

Notes

Author Biographies

John Alford, PhD, is a professor of public sector management at the University of Melbourne and at the Australia and New Zealand School of Government.

Jean Hartley, PhD, is a professor in public leadership at the Open University Business School, UK.

Sophie Yates, MAppLing, is a research fellow at the Australia and New Zealand School of Government.

Owen Hughes, PhD, is a professor and dean of students at RMIT University, Australia.

References

Aberbach

Putnam

Rockman

(1981). Bureacrats and politicians in Western democracies. New York, NY: Columbia University Press.

Aberbach

Rockman

(1994). Civil servants and policymakers: Neutral or responsive competence? Governance, 7, 461-469.

Althaus

Bridgman

Davis

(2012). The Australian policy handbook (5th ed.). Crows Nest, Australia: Allen & Unwin.

Bachrach

Baratz

(1962). The two faces of power. American Political Science Review, 56, 641-651.

Bardach

(2008). A practical guide for policy analysis (3rd ed.). New York, NY: Chatham House.

Behn

(2001). Rethinking democratic accountability. Washington, DC: Brookings Institution Press.

Campbell

Peters

B. G.

(1988). The politics/administration dichotomy: Death or merely change? Governance, 1, 79-99.

Carboni

(2010). Changing relationships between politicians and bureaucrats in contemporary democracies: An empirical analysis of the Italian experience. International Public Management Review, 11, 90-109.

Demir

(2009). The complementarity view: Exploring a continuum in political-administrative relations. Public Administration Review, 69, 876-888.

10.

Demir

Nyhan

(2008). The politics-administration dichotomy: An empirical search for correspondence between theory and practice. Public Administration Review, 68, 81-96.

11.

Demir

Reddick

(2012). Understanding shared roles in policy and administration: An empirical study of council-manager relations. Public Administration Review, 72, 526-536.

12.

Di Francesco

. (2000). An evaluation crucible: Evaluating policy advice in Australian central agencies. Australian Journal of Public Administration, 59, 36-48.

13.

Di Francesco

. (2012). Grand designs? The “managerial” role of ministers within Westminster-based public management policy. Australian Journal of Public Administration, 71, 257-268.

14.

Goodnow

(1900). Politics and administration. New York, NY: Russell and Russell.

15.

Hartley

Alford

Hughes

Yates

(2013). Leading with political astuteness: A study of public managers in Australia, New Zealand and the United Kingdom. Melbourne: ANZSOG and CMI.

16.

Hartley

Fletcher

(2008). Leading with political awareness: Leadership across diverse interests inside and outside the organization. In James

Collins

(Eds.), Leadership perspectives: Knowledge into action (pp. 163-176). London, England: Palgrave.

17.

Heclo

Wildavsky

(1974). The private government of public money: Community and policy inside British politics. London, England: Macmillan.

18.

Hood

(2002). Control, bargains, and cheating: The politics of public-service reform. Journal of Public Administration Research and Theory, 12, 309-332.

19.

Hood

Lodge

(2006). The politics of public service bargain: Reward, competency, loyalty and blame. Oxford, UK: Oxford University Press.

20.

Kaufman

(1969). Administrative decentralization and political power. Public Administration Review, 29, 3-15.

21.

Kettl

(2009). The next government of the United States. New York, NY: W.W. Norton.

22.

Kingdon

(2011). Agendas, alternatives, and public policies (2nd ed.). London, England: Longman.

23.

Lasswell

(1958). Politics: Who gets what, when, how. Cleveland, OH: Meridian Books.

24.

Lukes

(1974). Power: A radical view. London, England: Macmillan.

25.

MacDermott

(2008). Whatever happened to frank and fearless? The impact of new public management on the Australian Public Service. Canberra, Australia: ANU Press.

26.

Manzie

Hartley

(2013). Dancing on ice: Leadership with political astuteness by senior public servants in the UK. Milton Keynes: The Open University Business School.

27.

Matheson

Scanlan

Tanner

(1997). Strategic management in government: Extending the reform model in New Zealand. Paris, France: OECD. Retrieved from http://www.oecd.org/newzealand/1902913.pdf

28.

Miller

Wright

(2011). The politics-administration continuum in American State Governments: Administrative accountability as a positive-sum pattern. Current Topics in Management, 15, 225-240.

29.

Moore

(1995). Creating public value: Strategic management in government. Cambridge, MA: Harvard University Press.

30.

Mouritzen

Svara

(2002). Leadership at the apex: Politicians and administrators in Western local governments. Pittsburgh, PA: University of Pittsburgh Press.

31.

Moynihan

Ingraham

(2010). The suspect handmaiden: The evolution of politics and administration in the American state. Public Administration Review, 70(Suppl. 1), 229-237.

32.

Osborne

Gaebler

(1993). Reinventing government. New York, NY: Plume.

33.

Overeem

(2005). The value of the dichotomy: Politics, administration and the political neutrality of administrators. Administrative Theory & Praxis, 27, 311-329.

34.

Page

(2012). Policy without politicians: Bureaucratic influence in comparative perspective. Oxford, UK: Oxford University Press.

35.

Peters

B. G.

(1987). Politicians and bureaucrats in the politics of policy making. In Lane

(Ed.), Bureaucracy and public choice (pp. 256-282). London, England: Sage.

36.

Peters

B. G.

(2001). The politics of bureaucracy. London, England: Routledge.

37.

Pressman

Wildavsky

(1973). Implementation: How great expectations in Washington are dashed in Oakland. Berkeley: University of California Press.

38.

Rittel

Webber

(1973). Dilemmas in a general theory of planning. Policy Sciences, 4, 155-169.

39.

Sayre

W. S.

(1958). Premises of public administration: Past and emerging. Public Administration Review, 18, 102-105.

40.

Schaffer

(1973). The administrative factor. London, England: Frank Cass.

41.

Sidney

(2006). Policy formulation: Design and tools. In Fischer

Miller

Sidney

(Eds.), Handbook of public policy analysis: Theory, politics and methods (pp. 79-87). Boca Raton, FL: CRC Press.

42.

Skok

(1995). Policy issue networks and the public policy cycle: A structural-functional framework for public administration. Public Administration Review, 55, 325-332.

43.

Stocker

Thompson-Fawcett

(2014). “It’s not like never the twain shall meet”: Politician-staff relationship structures in local government. Local Government Studies, 40, 791-808. doi:10.1080/03003930.2014.887563

44.

Svara

(1999). The politics-administration dichotomy model as aberration. Public Administration Review, 58, 51-58.

45.

Svara

(2001). The myth of the dichotomy: Complementarity of politics and administration in the past and future of public administration. Public Administration Review, 61, 176-183.

46.

Svara

(2006a). Complexity in political-administrative relations and the limits of the dichotomy concept. Administrative Theory & Praxis, 28, 121-139.

47.

Svara

(2006b). Introduction: Politicians and administrators in the political process: A review of themes and issues in the literature. International Journal of Public Administration, 29, 953-976.

48.

Torfing

Triantafillou

(2013). What’s in a name? Grasping new public governance as a political-administrative system. International Review of Public Administration, 18(2), 9-25.

49.

van der Wal

. (2013). Mandarins versus Machiavellians? On differences between work motivations of administrative and political elites. Public Administration Review, 73, 749-759.

50.

Waldo

(1984). The administrative state: A study of the political theory of American public administration (2nd ed.). New York, NY: Holmes & Meier.

51.

Weber

(1994). The profession and vocation of politics. In Lassman

Speirs

(Eds.), Weber: Political writings (pp. 309-369). Cambridge, UK: Cambridge University Press. (Original work published 1922).

52.

Weimer

Vining

. (1995). Policy analysis: Concepts and practice (3rd ed.). Englewood Cliffs, NJ: Prentice-Hall.

53.

Wilson

(1989). Bureaucracy: What government agencies do and why they do it. New York, NY: Basic Books.

54.

Wilson

(1887). The study of administration. Political Science Quarterly, 2, 197-222.

55.

Zalmanovitch

(2014). Don’t reinvent the wheel: The search for an identity for public administration. International Review of Administrative Sciences, 80, 808-826.

56.

Zhang

(2014). The city manager’s role in policy-making: A perspective beyond substitution collaboration models. The American Review of Public Administration, 44, 358-372.